| [stillwater-sc.com](http://www.stillwater-sc.com/) | Document comprehension and association with word2vec |
| Channel 4 | 
| [channel4.com](http://www.channel4.com/) | Recommendation engine |
| Amazon | 
| [amazon.com](http://www.amazon.com/) | Document similarity|
+| SiteGround Hosting | 
| [siteground.com](https://www.siteground.com/) | An ensemble search engine which uses different embeddings models and similarities, including word2vec, WMD, and LDA. |
-------
diff --git a/continuous_integration/travis/flake8_diff.sh b/continuous_integration/travis/flake8_diff.sh
index 119e383fa6..50b2103e58 100755
--- a/continuous_integration/travis/flake8_diff.sh
+++ b/continuous_integration/travis/flake8_diff.sh
@@ -133,6 +133,6 @@ check_files() {
if [[ "$MODIFIED_FILES" == "no_match" ]]; then
echo "No file has been modified"
else
- check_files "$(echo "$MODIFIED_FILES" )" "--ignore=E501,E731,E12,W503 --exclude=*.sh,*.md,*.yml"
+ check_files "$(echo "$MODIFIED_FILES" )" "--ignore=E501,E731,E12,W503 --exclude=*.sh,*.md,*.yml,*.rst,*.ipynb,*.vec,Dockerfile*"
fi
echo -e "No problem detected by flake8\n"
diff --git a/docs/notebooks/Corpora_and_Vector_Spaces.ipynb b/docs/notebooks/Corpora_and_Vector_Spaces.ipynb
index 44560f952d..bccfa9cc91 100644
--- a/docs/notebooks/Corpora_and_Vector_Spaces.ipynb
+++ b/docs/notebooks/Corpora_and_Vector_Spaces.ipynb
@@ -17,7 +17,7 @@
},
{
"cell_type": "code",
- "execution_count": 1,
+ "execution_count": null,
"metadata": {
"collapsed": true
},
@@ -27,6 +27,18 @@
"logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s', level=logging.INFO)"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import os\n",
+ "import tempfile\n",
+ "TEMP_FOLDER = tempfile.gettempdir()\n",
+ "print('Folder \"{}\" will be used to save temporary dictionary and corpus.'.format(TEMP_FOLDER))"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -149,7 +161,7 @@
],
"source": [
"dictionary = corpora.Dictionary(texts)\n",
- "dictionary.save('/tmp/deerwester.dict') # store the dictionary, for future reference\n",
+ "dictionary.save(os.path.join(TEMP_FOLDER, 'deerwester.dict')) # store the dictionary, for future reference\n",
"print(dictionary)"
]
},
@@ -211,7 +223,11 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "The function `doc2bow()` simply counts the number of occurrences of each distinct word, converts the word to its integer word id and returns the result as a sparse vector. The sparse vector `[(word_id, 1), (word_id, 1)]` therefore reads: in the document *“Human computer interaction”*, the words *\"computer\"* and *\"human\"*, identified by an integer id given by the built dictionary, appear once; the other ten dictionary words appear (implicitly) zero times. Check their id at the dictionary displayed in the previous cell and see that they match."
+ "The function `doc2bow()` simply counts the number of occurrences of each distinct word, converts the word to its integer word id and returns the result as a bag-of-words--a sparse vector, in the form of `[(word_id, word_count), ...]`. \n",
+ "\n",
+ "As the token_id is 0 for *\"human\"* and 2 for *\"computer\"*, the new document *“Human computer interaction”* will be transformed to [(0, 1), (2, 1)]. The words *\"computer\"* and *\"human\"* exist in the dictionary and appear once. Thus, they become (0, 1), (2, 1) respectively in the sparse vector. The word *\"interaction\"* doesn't exist in the dictionary and, thus, will not show up in the sparse vector. The other ten dictionary words, that appear (implicitly) zero times, will not show up in the sparse vector and , ,there will never be a element in the sparse vector like (3, 0).\n",
+ "\n",
+ "For people familiar with scikit learn, `doc2bow()` has similar behaviors as calling `transform()` on [`CountVectorizer`](http://scikit-learn.org/stable/modules/generated/sklearn.feature_extraction.text.CountVectorizer.html). `doc2bow()` can behave like `fit_transform()` as well. For more details, please look at [gensim API Doc](https://radimrehurek.com/gensim/corpora/dictionary.html#gensim.corpora.dictionary.Dictionary.doc2bow)."
]
},
{
@@ -239,7 +255,7 @@
],
"source": [
"corpus = [dictionary.doc2bow(text) for text in texts]\n",
- "corpora.MmCorpus.serialize('/tmp/deerwester.mm', corpus) # store to disk, for later use\n",
+ "corpora.MmCorpus.serialize(os.path.join(TEMP_FOLDER, 'deerwester.mm'), corpus) # store to disk, for later use\n",
"for c in corpus:\n",
" print(c)"
]
@@ -338,7 +354,7 @@
"source": [
"Although the output is the same as for the plain Python list, the corpus is now much more memory friendly, because at most one vector resides in RAM at a time. Your corpus can now be as large as you want.\n",
"\n",
- "Similarly, to construct the dictionary without loading all texts into memory:"
+ "We are going to create the dictionary from the mycorpus.txt file without loading the entire file into memory. Then, we will generate the list of token ids to remove from this dictionary by querying the dictionary for the token ids of the stop words, and by querying the document frequencies dictionary (dictionary.dfs) for token ids that only appear once. Finally, we will filter these token ids out of our dictionary and call dictionary.compactify() to remove the gaps in the token id series."
]
},
{
@@ -399,14 +415,14 @@
"# create a toy corpus of 2 documents, as a plain Python list\n",
"corpus = [[(1, 0.5)], []] # make one document empty, for the heck of it\n",
"\n",
- "corpora.MmCorpus.serialize('/tmp/corpus.mm', corpus)"
+ "corpora.MmCorpus.serialize(os.path.join(TEMP_FOLDER, 'corpus.mm'), corpus)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "Other formats include [Joachim’s SVMlight format](http://svmlight.joachims.org/), [Blei’s LDA-C format](http://www.cs.princeton.edu/~blei/lda-c/) and [GibbsLDA++ format](http://gibbslda.sourceforge.net/)."
+ "Other formats include [Joachim’s SVMlight format](http://svmlight.joachims.org/), [Blei’s LDA-C format](http://www.cs.columbia.edu/~blei/lda-c/) and [GibbsLDA++ format](http://gibbslda.sourceforge.net/)."
]
},
{
@@ -417,9 +433,9 @@
},
"outputs": [],
"source": [
- "corpora.SvmLightCorpus.serialize('/tmp/corpus.svmlight', corpus)\n",
- "corpora.BleiCorpus.serialize('/tmp/corpus.lda-c', corpus)\n",
- "corpora.LowCorpus.serialize('/tmp/corpus.low', corpus)"
+ "corpora.SvmLightCorpus.serialize(os.path.join(TEMP_FOLDER, 'corpus.svmlight'), corpus)\n",
+ "corpora.BleiCorpus.serialize(os.path.join(TEMP_FOLDER, 'corpus.lda-c'), corpus)\n",
+ "corpora.LowCorpus.serialize(os.path.join(TEMP_FOLDER, 'corpus.low'), corpus)"
]
},
{
@@ -437,7 +453,7 @@
},
"outputs": [],
"source": [
- "corpus = corpora.MmCorpus('/tmp/corpus.mm')"
+ "corpus = corpora.MmCorpus(os.path.join(TEMP_FOLDER, 'corpus.mm'))"
]
},
{
@@ -539,7 +555,7 @@
},
"outputs": [],
"source": [
- "corpora.BleiCorpus.serialize('/tmp/corpus.lda-c', corpus)"
+ "corpora.BleiCorpus.serialize(os.path.join(TEMP_FOLDER, 'corpus.lda-c'), corpus)"
]
},
{
@@ -600,23 +616,23 @@
],
"metadata": {
"kernelspec": {
- "display_name": "Python 3",
+ "display_name": "Python 2",
"language": "python",
- "name": "python3"
+ "name": "python2"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
- "version": 3
+ "version": 2
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
- "pygments_lexer": "ipython3",
- "version": "3.6.0"
+ "pygments_lexer": "ipython2",
+ "version": "2.7.6"
}
},
"nbformat": 4,
- "nbformat_minor": 0
+ "nbformat_minor": 1
}
diff --git a/docs/notebooks/Similarity_Queries.ipynb b/docs/notebooks/Similarity_Queries.ipynb
index 88f69da85f..34a367e8ee 100644
--- a/docs/notebooks/Similarity_Queries.ipynb
+++ b/docs/notebooks/Similarity_Queries.ipynb
@@ -16,7 +16,7 @@
},
{
"cell_type": "code",
- "execution_count": 2,
+ "execution_count": 5,
"metadata": {
"collapsed": true
},
@@ -26,6 +26,28 @@
"logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s', level=logging.INFO)"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Folder \"C:\\Users\\chaor\\AppData\\Local\\Temp\" will be used to save temporary dictionary and corpus.\n"
+ ]
+ }
+ ],
+ "source": [
+ "import os\n",
+ "import tempfile\n",
+ "TEMP_FOLDER = tempfile.gettempdir()\n",
+ "print('Folder \"{}\" will be used to save temporary dictionary and corpus.'.format(TEMP_FOLDER))"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -44,11 +66,22 @@
},
{
"cell_type": "code",
- "execution_count": 12,
+ "execution_count": 7,
"metadata": {
"collapsed": false
},
"outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "2017-05-22 14:27:18,911 : INFO : loading Dictionary object from C:\\Users\\chaor\\AppData\\Local\\Temp\\deerwester.dict\n",
+ "2017-05-22 14:27:18,911 : INFO : loaded C:\\Users\\chaor\\AppData\\Local\\Temp\\deerwester.dict\n",
+ "2017-05-22 14:27:18,921 : INFO : loaded corpus index from C:\\Users\\chaor\\AppData\\Local\\Temp\\deerwester.mm.index\n",
+ "2017-05-22 14:27:18,924 : INFO : initializing corpus reader from C:\\Users\\chaor\\AppData\\Local\\Temp\\deerwester.mm\n",
+ "2017-05-22 14:27:18,929 : INFO : accepted corpus with 9 documents, 12 features, 28 non-zero entries\n"
+ ]
+ },
{
"name": "stdout",
"output_type": "stream",
@@ -60,8 +93,8 @@
"source": [
"from gensim import corpora, models, similarities\n",
"\n",
- "dictionary = corpora.Dictionary.load('/tmp/deerwester.dict')\n",
- "corpus = corpora.MmCorpus('/tmp/deerwester.mm') # comes from the first tutorial, \"From strings to vectors\"\n",
+ "dictionary = corpora.Dictionary.load(os.path.join(TEMP_FOLDER, 'deerwester.dict'))\n",
+ "corpus = corpora.MmCorpus(os.path.join(TEMP_FOLDER, 'deerwester.mm')) # comes from the first tutorial, \"From strings to vectors\"\n",
"print(corpus)"
]
},
@@ -74,11 +107,30 @@
},
{
"cell_type": "code",
- "execution_count": 13,
+ "execution_count": 8,
"metadata": {
"collapsed": false
},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "2017-05-22 14:27:19,048 : INFO : using serial LSI version on this node\n",
+ "2017-05-22 14:27:19,050 : INFO : updating model with new documents\n",
+ "2017-05-22 14:27:19,054 : INFO : preparing a new chunk of documents\n",
+ "2017-05-22 14:27:19,057 : INFO : using 100 extra samples and 2 power iterations\n",
+ "2017-05-22 14:27:19,060 : INFO : 1st phase: constructing (12, 102) action matrix\n",
+ "2017-05-22 14:27:19,064 : INFO : orthonormalizing (12, 102) action matrix\n",
+ "2017-05-22 14:27:19,068 : INFO : 2nd phase: running dense svd on (12, 9) matrix\n",
+ "2017-05-22 14:27:19,070 : INFO : computing the final decomposition\n",
+ "2017-05-22 14:27:19,073 : INFO : keeping 2 factors (discarding 43.156% of energy spectrum)\n",
+ "2017-05-22 14:27:19,076 : INFO : processed documents up to #9\n",
+ "2017-05-22 14:27:19,078 : INFO : topic #0(3.341): 0.644*\"system\" + 0.404*\"user\" + 0.301*\"eps\" + 0.265*\"response\" + 0.265*\"time\" + 0.240*\"computer\" + 0.221*\"human\" + 0.206*\"survey\" + 0.198*\"interface\" + 0.036*\"graph\"\n",
+ "2017-05-22 14:27:19,081 : INFO : topic #1(2.542): 0.623*\"graph\" + 0.490*\"trees\" + 0.451*\"minors\" + 0.274*\"survey\" + -0.167*\"system\" + -0.141*\"eps\" + -0.113*\"human\" + 0.107*\"response\" + 0.107*\"time\" + -0.072*\"interface\"\n"
+ ]
+ }
+ ],
"source": [
"lsi = models.LsiModel(corpus, id2word=dictionary, num_topics=2)"
]
@@ -92,7 +144,7 @@
},
{
"cell_type": "code",
- "execution_count": 14,
+ "execution_count": 9,
"metadata": {
"collapsed": false
},
@@ -101,7 +153,7 @@
"name": "stdout",
"output_type": "stream",
"text": [
- "[(0, 0.46182100453271591), (1, 0.070027665279000534)]\n"
+ "[(0, 0.46182100453271535), (1, -0.070027665279000437)]\n"
]
}
],
@@ -135,11 +187,20 @@
},
{
"cell_type": "code",
- "execution_count": 15,
+ "execution_count": 10,
"metadata": {
"collapsed": false
},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "2017-05-22 14:27:19,299 : WARNING : scanning corpus to determine the number of features (consider setting `num_features` explicitly)\n",
+ "2017-05-22 14:27:19,358 : INFO : creating matrix with 9 documents and 2 features\n"
+ ]
+ }
+ ],
"source": [
"index = similarities.MatrixSimilarity(lsi[corpus]) # transform corpus to LSI space and index it"
]
@@ -157,14 +218,23 @@
},
{
"cell_type": "code",
- "execution_count": 18,
+ "execution_count": 12,
"metadata": {
- "collapsed": true
+ "collapsed": false
},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "2017-05-22 14:27:52,760 : INFO : saving MatrixSimilarity object under C:\\Users\\chaor\\AppData\\Local\\Temp\\deerwester.index, separately None\n",
+ "2017-05-22 14:27:52,772 : INFO : saved C:\\Users\\chaor\\AppData\\Local\\Temp\\deerwester.index\n"
+ ]
+ }
+ ],
"source": [
- "index.save('/tmp/deerwester.index')\n",
- "index = similarities.MatrixSimilarity.load('/tmp/deerwester.index')"
+ "index.save(os.path.join(TEMP_FOLDER, 'deerwester.index'))\n",
+ "#index = similarities.MatrixSimilarity.load(os.path.join(TEMP_FOLDER, 'index'))"
]
},
{
@@ -190,7 +260,7 @@
},
{
"cell_type": "code",
- "execution_count": 21,
+ "execution_count": 13,
"metadata": {
"collapsed": false
},
@@ -199,7 +269,7 @@
"name": "stdout",
"output_type": "stream",
"text": [
- "[(0, 0.99809301), (1, 0.93748635), (2, 0.99844527), (3, 0.9865886), (4, 0.90755945), (5, -0.12416792), (6, -0.10639259), (7, -0.098794632), (8, 0.050041769)]\n"
+ "[(0, 0.99809301), (1, 0.93748635), (2, 0.99844527), (3, 0.9865886), (4, 0.90755945), (5, -0.12416792), (6, -0.10639259), (7, -0.098794639), (8, 0.050041765)]\n"
]
}
],
@@ -246,25 +316,34 @@
"* your **feedback is most welcome** and appreciated (and it’s not just the code!): [idea contributions](https://github.com/piskvorky/gensim/wiki/Ideas-&-Features-proposals), [bug reports](https://github.com/piskvorky/gensim/issues) or just consider contributing [user stories and general questions](http://groups.google.com/group/gensim/topics).\n",
"Gensim has no ambition to become an all-encompassing framework, across all NLP (or even Machine Learning) subfields. Its mission is to help NLP practicioners try out popular topic modelling algorithms on large datasets easily, and to facilitate prototyping of new algorithms for researchers."
]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": []
}
],
"metadata": {
"kernelspec": {
- "display_name": "Python 2",
+ "display_name": "Python 3",
"language": "python",
- "name": "python2"
+ "name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
- "version": 2
+ "version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
- "pygments_lexer": "ipython2",
- "version": "2.7.10"
+ "pygments_lexer": "ipython3",
+ "version": "3.6.0"
}
},
"nbformat": 4,
diff --git a/docs/notebooks/Topics_and_Transformations.ipynb b/docs/notebooks/Topics_and_Transformations.ipynb
index 4c78cd6037..5a8ec7f985 100644
--- a/docs/notebooks/Topics_and_Transformations.ipynb
+++ b/docs/notebooks/Topics_and_Transformations.ipynb
@@ -16,18 +16,40 @@
},
{
"cell_type": "code",
- "execution_count": 1,
+ "execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"import logging\n",
- "import os.path\n",
"\n",
"logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s', level=logging.INFO)"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Folder \"C:\\Users\\chaor\\AppData\\Local\\Temp\" will be used to save temporary dictionary and corpus.\n"
+ ]
+ }
+ ],
+ "source": [
+ "import tempfile\n",
+ "import os.path\n",
+ "\n",
+ "TEMP_FOLDER = tempfile.gettempdir()\n",
+ "print('Folder \"{}\" will be used to save temporary dictionary and corpus.'.format(TEMP_FOLDER))"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -41,24 +63,16 @@
},
{
"cell_type": "code",
- "execution_count": 2,
+ "execution_count": null,
"metadata": {
"collapsed": false
},
- "outputs": [
- {
- "name": "stdout",
- "output_type": "stream",
- "text": [
- "Used files generated from first tutorial\n"
- ]
- }
- ],
+ "outputs": [],
"source": [
"from gensim import corpora, models, similarities\n",
- "if (os.path.exists(\"/tmp/deerwester.dict\")):\n",
- " dictionary = corpora.Dictionary.load('/tmp/deerwester.dict')\n",
- " corpus = corpora.MmCorpus('/tmp/deerwester.mm')\n",
+ "if os.path.isfile(os.path.join(TEMP_FOLDER, 'deerwester.dict')):\n",
+ " dictionary = corpora.Dictionary.load(os.path.join(TEMP_FOLDER, 'deerwester.dict'))\n",
+ " corpus = corpora.MmCorpus(os.path.join(TEMP_FOLDER, 'deerwester.mm'))\n",
" print(\"Used files generated from first tutorial\")\n",
"else:\n",
" print(\"Please run first tutorial to generate data set\")"
@@ -66,7 +80,7 @@
},
{
"cell_type": "code",
- "execution_count": 3,
+ "execution_count": 4,
"metadata": {
"collapsed": false
},
@@ -75,16 +89,16 @@
"name": "stdout",
"output_type": "stream",
"text": [
+ "human\n",
"interface\n",
- "computer\n",
- "human\n"
+ "computer\n"
]
}
],
"source": [
- "print (dictionary[0])\n",
- "print (dictionary[1])\n",
- "print (dictionary[2])"
+ "print(dictionary[0])\n",
+ "print(dictionary[1])\n",
+ "print(dictionary[2])"
]
},
{
@@ -103,9 +117,9 @@
},
{
"cell_type": "code",
- "execution_count": 4,
+ "execution_count": 5,
"metadata": {
- "collapsed": true
+ "collapsed": false
},
"outputs": [],
"source": [
@@ -124,7 +138,7 @@
},
{
"cell_type": "code",
- "execution_count": 5,
+ "execution_count": 6,
"metadata": {
"collapsed": false
},
@@ -151,7 +165,7 @@
},
{
"cell_type": "code",
- "execution_count": 6,
+ "execution_count": 7,
"metadata": {
"collapsed": false
},
@@ -161,14 +175,14 @@
"output_type": "stream",
"text": [
"[(0, 0.5773502691896257), (1, 0.5773502691896257), (2, 0.5773502691896257)]\n",
- "[(1, 0.44424552527467476), (3, 0.44424552527467476), (4, 0.44424552527467476), (5, 0.44424552527467476), (6, 0.3244870206138555), (7, 0.3244870206138555)]\n",
- "[(0, 0.5710059809418182), (6, 0.4170757362022777), (7, 0.4170757362022777), (8, 0.5710059809418182)]\n",
- "[(2, 0.49182558987264147), (6, 0.7184811607083769), (8, 0.49182558987264147)]\n",
- "[(3, 0.6282580468670046), (4, 0.6282580468670046), (7, 0.45889394536615247)]\n",
+ "[(2, 0.44424552527467476), (3, 0.44424552527467476), (4, 0.3244870206138555), (5, 0.3244870206138555), (6, 0.44424552527467476), (7, 0.44424552527467476)]\n",
+ "[(1, 0.5710059809418182), (4, 0.4170757362022777), (5, 0.4170757362022777), (8, 0.5710059809418182)]\n",
+ "[(0, 0.49182558987264147), (5, 0.7184811607083769), (8, 0.49182558987264147)]\n",
+ "[(4, 0.45889394536615247), (6, 0.6282580468670046), (7, 0.6282580468670046)]\n",
"[(9, 1.0)]\n",
"[(9, 0.7071067811865475), (10, 0.7071067811865475)]\n",
"[(9, 0.5080429008916749), (10, 0.5080429008916749), (11, 0.695546419520037)]\n",
- "[(5, 0.6282580468670046), (10, 0.45889394536615247), (11, 0.6282580468670046)]\n"
+ "[(3, 0.6282580468670046), (10, 0.45889394536615247), (11, 0.6282580468670046)]\n"
]
}
],
@@ -192,9 +206,9 @@
},
{
"cell_type": "code",
- "execution_count": 7,
+ "execution_count": 8,
"metadata": {
- "collapsed": true
+ "collapsed": false
},
"outputs": [],
"source": [
@@ -211,7 +225,7 @@
},
{
"cell_type": "code",
- "execution_count": 8,
+ "execution_count": 9,
"metadata": {
"collapsed": false
},
@@ -220,12 +234,12 @@
"data": {
"text/plain": [
"[(0,\n",
- " u'0.703*\"trees\" + 0.538*\"graph\" + 0.402*\"minors\" + 0.187*\"survey\" + 0.061*\"system\" + 0.060*\"time\" + 0.060*\"response\" + 0.058*\"user\" + 0.049*\"computer\" + 0.035*\"interface\"'),\n",
+ " '0.703*\"trees\" + 0.538*\"graph\" + 0.402*\"minors\" + 0.187*\"survey\" + 0.061*\"system\" + 0.060*\"response\" + 0.060*\"time\" + 0.058*\"user\" + 0.049*\"computer\" + 0.035*\"interface\"'),\n",
" (1,\n",
- " u'-0.460*\"system\" + -0.373*\"user\" + -0.332*\"eps\" + -0.328*\"interface\" + -0.320*\"response\" + -0.320*\"time\" + -0.293*\"computer\" + -0.280*\"human\" + -0.171*\"survey\" + 0.161*\"trees\"')]"
+ " '-0.460*\"system\" + -0.373*\"user\" + -0.332*\"eps\" + -0.328*\"interface\" + -0.320*\"time\" + -0.320*\"response\" + -0.293*\"computer\" + -0.280*\"human\" + -0.171*\"survey\" + 0.161*\"trees\"')]"
]
},
- "execution_count": 8,
+ "execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
@@ -245,7 +259,7 @@
},
{
"cell_type": "code",
- "execution_count": 9,
+ "execution_count": 10,
"metadata": {
"collapsed": false
},
@@ -254,15 +268,15 @@
"name": "stdout",
"output_type": "stream",
"text": [
- "[(0, 0.066007833960903733), (1, -0.52007033063618524)]\n",
- "[(0, 0.19667592859142552), (1, -0.76095631677000408)]\n",
- "[(0, 0.089926399724463882), (1, -0.72418606267525099)]\n",
- "[(0, 0.075858476521781265), (1, -0.63205515860034311)]\n",
- "[(0, 0.10150299184980148), (1, -0.57373084830029475)]\n",
- "[(0, 0.70321089393783065), (1, 0.16115180214025804)]\n",
- "[(0, 0.87747876731198282), (1, 0.16758906864659434)]\n",
- "[(0, 0.90986246868185772), (1, 0.14086553628719053)]\n",
- "[(0, 0.6165825350569285), (1, -0.053929075663893253)]\n"
+ "[(0, 0.066007833960904788), (1, -0.52007033063618524)]\n",
+ "[(0, 0.19667592859142516), (1, -0.76095631677000508)]\n",
+ "[(0, 0.089926399724465478), (1, -0.72418606267525021)]\n",
+ "[(0, 0.075858476521782792), (1, -0.63205515860034234)]\n",
+ "[(0, 0.10150299184980102), (1, -0.57373084830029564)]\n",
+ "[(0, 0.70321089393783121), (1, 0.16115180214025845)]\n",
+ "[(0, 0.87747876731198327), (1, 0.16758906864659459)]\n",
+ "[(0, 0.90986246868185772), (1, 0.14086553628719056)]\n",
+ "[(0, 0.61658253505692806), (1, -0.053929075663893752)]\n"
]
}
],
@@ -273,14 +287,14 @@
},
{
"cell_type": "code",
- "execution_count": 10,
+ "execution_count": 14,
"metadata": {
- "collapsed": true
+ "collapsed": false
},
"outputs": [],
"source": [
- "lsi.save('/tmp/model.lsi') # same for tfidf, lda, ...\n",
- "lsi = models.LsiModel.load('/tmp/model.lsi')"
+ "lsi.save(os.path.join(TEMP_FOLDER, 'model.lsi')) # same for tfidf, lda, ...\n",
+ "#lsi = models.LsiModel.load(os.path.join(TEMP_FOLDER, 'model.lsi'))"
]
},
{
@@ -304,7 +318,7 @@
},
{
"cell_type": "code",
- "execution_count": 11,
+ "execution_count": 15,
"metadata": {
"collapsed": false
},
@@ -323,7 +337,7 @@
},
{
"cell_type": "code",
- "execution_count": 12,
+ "execution_count": 16,
"metadata": {
"collapsed": false
},
@@ -372,7 +386,7 @@
},
{
"cell_type": "code",
- "execution_count": 13,
+ "execution_count": 17,
"metadata": {
"collapsed": false
},
@@ -391,7 +405,7 @@
},
{
"cell_type": "code",
- "execution_count": 14,
+ "execution_count": 18,
"metadata": {
"collapsed": false
},
@@ -417,7 +431,7 @@
},
{
"cell_type": "code",
- "execution_count": 15,
+ "execution_count": 19,
"metadata": {
"collapsed": false
},
@@ -455,6 +469,15 @@
"[4]\tHalko, Martinsson, Tropp. 2009. Finding structure with randomness. \n",
"[5]\tŘehůřek. 2011. Subspace tracking for Latent Semantic Analysis. "
]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": []
}
],
"metadata": {
diff --git a/docs/notebooks/Wordrank_comparisons.ipynb b/docs/notebooks/Wordrank_comparisons.ipynb
index e035e4f2b5..2aecbd3911 100644
--- a/docs/notebooks/Wordrank_comparisons.ipynb
+++ b/docs/notebooks/Wordrank_comparisons.ipynb
@@ -23,7 +23,9 @@
{
"cell_type": "code",
"execution_count": 1,
- "metadata": {},
+ "metadata": {
+ "collapsed": false
+ },
"outputs": [
{
"name": "stdout",
@@ -76,7 +78,9 @@
{
"cell_type": "code",
"execution_count": 2,
- "metadata": {},
+ "metadata": {
+ "collapsed": false
+ },
"outputs": [
{
"name": "stdout",
@@ -169,9 +173,10 @@
" \n",
" # Train using wordrank\n",
" output_file = '{:s}_wr'.format(output_name)\n",
+ " output_dir = 'wordrank_model' # directory to save embeddings and metadata to\n",
" if not os.path.isfile(os.path.join(MODELS_DIR, '{:s}.vec'.format(output_file))):\n",
" print('\\nTraining wordrank on {:s} corpus..'.format(corpus_file))\n",
- " %time wr_model = Wordrank.train(WR_HOME, corpus_file, **wr_params); wr_model\n",
+ " %time wr_model = Wordrank.train(WR_HOME, corpus_file, output_dir, **wr_params); wr_model\n",
" locals()['wr_model'].save_word2vec_format(os.path.join(MODELS_DIR, '{:s}.vec'.format(output_file)))\n",
" print('\\nSaved wordrank model as {:s}.vec'.format(output_file))\n",
" else:\n",
@@ -193,7 +198,9 @@
{
"cell_type": "code",
"execution_count": 3,
- "metadata": {},
+ "metadata": {
+ "collapsed": false
+ },
"outputs": [
{
"name": "stdout",
@@ -235,7 +242,9 @@
{
"cell_type": "code",
"execution_count": 1,
- "metadata": {},
+ "metadata": {
+ "collapsed": true
+ },
"outputs": [],
"source": [
"import logging\n",
@@ -263,7 +272,9 @@
{
"cell_type": "code",
"execution_count": 2,
- "metadata": {},
+ "metadata": {
+ "collapsed": false
+ },
"outputs": [
{
"name": "stderr",
@@ -587,7 +598,9 @@
{
"cell_type": "code",
"execution_count": 3,
- "metadata": {},
+ "metadata": {
+ "collapsed": false
+ },
"outputs": [
{
"name": "stderr",
@@ -973,14 +986,16 @@
{
"cell_type": "code",
"execution_count": 9,
- "metadata": {},
+ "metadata": {
+ "collapsed": false
+ },
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAnYAAATbCAYAAAAUBvjzAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nKAAAAABJRU5ErkJggg==\n"
},
- "output_type": "display_data",
- "metadata": {}
+ "metadata": {},
+ "output_type": "display_data"
}
],
"source": [
@@ -1111,14 +1126,16 @@
{
"cell_type": "code",
"execution_count": 10,
- "metadata": {},
+ "metadata": {
+ "collapsed": false
+ },
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAm4AAATbCAYAAAAgfznvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n"
},
- "output_type": "display_data",
- "metadata": {}
+ "metadata": {},
+ "output_type": "display_data"
}
],
"source": [
@@ -1197,16 +1214,16 @@
"language_info": {
"codemirror_mode": {
"name": "ipython",
- "version": 2.0
+ "version": 2
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython2",
- "version": "2.7.12"
+ "version": "2.7.13"
}
},
"nbformat": 4,
"nbformat_minor": 2
-}
\ No newline at end of file
+}
diff --git a/docs/notebooks/annoytutorial-text8.ipynb b/docs/notebooks/annoytutorial-text8.ipynb
new file mode 100644
index 0000000000..61fa6a8508
--- /dev/null
+++ b/docs/notebooks/annoytutorial-text8.ipynb
@@ -0,0 +1,743 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Similarity Queries using Annoy Tutorial"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "This tutorial is about using the ([Annoy Approximate Nearest Neighbors Oh Yeah](https://github.com/spotify/annoy \"Link to annoy repo\")) library for similarity queries with a Word2Vec model built with gensim.\n",
+ "\n",
+ "## Why use Annoy?\n",
+ "The current implementation for finding k nearest neighbors in a vector space in gensim has linear complexity via brute force in the number of indexed documents, although with extremely low constant factors. The retrieved results are exact, which is an overkill in many applications: approximate results retrieved in sub-linear time may be enough. Annoy can find approximate nearest neighbors much faster.\n",
+ "\n",
+ "\n",
+ "## Prerequisites\n",
+ "Additional libraries needed for this tutorial:\n",
+ "- annoy\n",
+ "- psutil\n",
+ "- matplotlib\n",
+ "\n",
+ "## Outline\n",
+ "1. Download Text8 Corpus\n",
+ "2. Build Word2Vec Model\n",
+ "3. Construct AnnoyIndex with model & make a similarity query\n",
+ "4. Verify & Evaluate performance\n",
+ "5. Evaluate relationship of `num_trees` to initialization time and accuracy"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "CPython 3.6.1\n",
+ "IPython 6.0.0\n",
+ "\n",
+ "gensim 2.1.0\n",
+ "numpy 1.12.1\n",
+ "scipy 0.19.0\n",
+ "psutil 5.2.2\n",
+ "matplotlib 2.0.2\n",
+ "\n",
+ "compiler : GCC 4.2.1 Compatible Apple LLVM 6.0 (clang-600.0.57)\n",
+ "system : Darwin\n",
+ "release : 14.5.0\n",
+ "machine : x86_64\n",
+ "processor : i386\n",
+ "CPU cores : 8\n",
+ "interpreter: 64bit\n"
+ ]
+ }
+ ],
+ "source": [
+ "# pip install watermark\n",
+ "%reload_ext watermark\n",
+ "%watermark -v -m -p gensim,numpy,scipy,psutil,matplotlib"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### 1. Download Text8 Corpus"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import os.path\n",
+ "if not os.path.isfile('text8'):\n",
+ " !wget -c http://mattmahoney.net/dc/text8.zip\n",
+ " !unzip text8.zip"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Import & Set up Logging\n",
+ "I'm not going to set up logging due to the verbose input displaying in notebooks, but if you want that, uncomment the lines in the cell below."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "LOGS = False\n",
+ "\n",
+ "if LOGS:\n",
+ " import logging\n",
+ " logging.basicConfig(format='%(asctime)s : %(levelname)s : %(message)s', level=logging.INFO)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### 2. Build Word2Vec Model"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {
+ "scrolled": true
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Word2Vec(vocab=71290, size=100, alpha=0.05)\n"
+ ]
+ }
+ ],
+ "source": [
+ "from gensim.models import Word2Vec, KeyedVectors\n",
+ "from gensim.models.word2vec import Text8Corpus\n",
+ "\n",
+ "# using params from Word2Vec_FastText_Comparison\n",
+ "\n",
+ "lr = 0.05\n",
+ "dim = 100\n",
+ "ws = 5\n",
+ "epoch = 5\n",
+ "minCount = 5\n",
+ "neg = 5\n",
+ "loss = 'ns'\n",
+ "t = 1e-4\n",
+ "\n",
+ "# Same values as used for fastText training above\n",
+ "params = {\n",
+ " 'alpha': lr,\n",
+ " 'size': dim,\n",
+ " 'window': ws,\n",
+ " 'iter': epoch,\n",
+ " 'min_count': minCount,\n",
+ " 'sample': t,\n",
+ " 'sg': 1,\n",
+ " 'hs': 0,\n",
+ " 'negative': neg\n",
+ "}\n",
+ "\n",
+ "model = Word2Vec(Text8Corpus('text8'), **params)\n",
+ "print(model)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "See the [Word2Vec tutorial](https://github.com/RaRe-Technologies/gensim/blob/develop/docs/notebooks/word2vec.ipynb) for how to initialize and save this model."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Comparing the traditional implementation and the Annoy approximation"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "#Set up the model and vector that we are using in the comparison\n",
+ "try:\n",
+ " from gensim.similarities.index import AnnoyIndexer\n",
+ "except ImportError:\n",
+ " raise ValueError(\"SKIP: Please install the annoy indexer\")\n",
+ "\n",
+ "model.init_sims()\n",
+ "annoy_index = AnnoyIndexer(model, 100)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[('the', 0.9999998807907104),\n",
+ " ('of', 0.8208043575286865),\n",
+ " ('in', 0.8024208545684814),\n",
+ " ('a', 0.7661813497543335),\n",
+ " ('and', 0.7392199039459229)]"
+ ]
+ },
+ "execution_count": 6,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# Dry run to make sure both indices are fully in RAM\n",
+ "vector = model.wv.syn0norm[0]\n",
+ "model.most_similar([vector], topn=5, indexer=annoy_index)\n",
+ "model.most_similar([vector], topn=5)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "import time\n",
+ "import numpy as np"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 8,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "def avg_query_time(annoy_index=None, queries=1000):\n",
+ " \"\"\"\n",
+ " Average query time of a most_similar method over 1000 random queries,\n",
+ " uses annoy if given an indexer\n",
+ " \"\"\"\n",
+ " total_time = 0\n",
+ " for _ in range(queries):\n",
+ " rand_vec = model.wv.syn0norm[np.random.randint(0, len(model.wv.vocab))]\n",
+ " start_time = time.clock()\n",
+ " model.most_similar([rand_vec], topn=5, indexer=annoy_index)\n",
+ " total_time += time.clock() - start_time\n",
+ " return total_time / queries"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 9,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Gensim (s/query):\t0.00571\n",
+ "Annoy (s/query):\t0.00028\n",
+ "\n",
+ "Annoy is 20.67 times faster on average on this particular run\n"
+ ]
+ }
+ ],
+ "source": [
+ "queries = 10000\n",
+ "\n",
+ "gensim_time = avg_query_time(queries=queries)\n",
+ "annoy_time = avg_query_time(annoy_index, queries=queries)\n",
+ "print(\"Gensim (s/query):\\t{0:.5f}\".format(gensim_time))\n",
+ "print(\"Annoy (s/query):\\t{0:.5f}\".format(annoy_time))\n",
+ "speed_improvement = gensim_time / annoy_time\n",
+ "print (\"\\nAnnoy is {0:.2f} times faster on average on this particular run\".format(speed_improvement))"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "\n",
+ "**This speedup factor is by no means constant** and will vary greatly from run to run and is particular to this data set, BLAS setup, Annoy parameters(as tree size increases speedup factor decreases), machine specifications, among other factors.\n",
+ "\n",
+ ">**Note**: Initialization time for the annoy indexer was not included in the times. The optimal knn algorithm for you to use will depend on how many queries you need to make and the size of the corpus. If you are making very few similarity queries, the time taken to initialize the annoy indexer will be longer than the time it would take the brute force method to retrieve results. If you are making many queries however, the time it takes to initialize the annoy indexer will be made up for by the incredibly fast retrieval times for queries once the indexer has been initialized\n",
+ "\n",
+ ">**Note** : Gensim's 'most_similar' method is using numpy operations in the form of dot product whereas Annoy's method isnt. If 'numpy' on your machine is using one of the BLAS libraries like ATLAS or LAPACK, it'll run on multiple cores(only if your machine has multicore support ). Check [SciPy Cookbook](http://scipy-cookbook.readthedocs.io/items/ParallelProgramming.html) for more details."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## 3. Construct AnnoyIndex with model & make a similarity query"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Creating an indexer\n",
+ "An instance of `AnnoyIndexer` needs to be created in order to use Annoy in gensim. The `AnnoyIndexer` class is located in `gensim.similarities.index`\n",
+ "\n",
+ "`AnnoyIndexer()` takes two parameters:\n",
+ "\n",
+ "**`model`**: A `Word2Vec` or `Doc2Vec` model\n",
+ "\n",
+ "**`num_trees`**: A positive integer. `num_trees` effects the build time and the index size. **A larger value will give more accurate results, but larger indexes**. More information on what trees in Annoy do can be found [here](https://github.com/spotify/annoy#how-does-it-work). The relationship between `num_trees`, build time, and accuracy will be investigated later in the tutorial. \n",
+ "\n",
+ "Now that we are ready to make a query, lets find the top 5 most similar words to \"science\" in the Text8 corpus. To make a similarity query we call `Word2Vec.most_similar` like we would traditionally, but with an added parameter, `indexer`. The only supported indexer in gensim as of now is Annoy. "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Approximate Neighbors\n",
+ "('science', 1.0)\n",
+ "('interdisciplinary', 0.6099119782447815)\n",
+ "('astrobiology', 0.5975957810878754)\n",
+ "('actuarial', 0.596003383398056)\n",
+ "('robotics', 0.5942946970462799)\n",
+ "('sciences', 0.59312504529953)\n",
+ "('scientific', 0.5900688469409943)\n",
+ "('psychohistory', 0.5890524089336395)\n",
+ "('bioethics', 0.5867903232574463)\n",
+ "('cryobiology', 0.5854728817939758)\n",
+ "('xenobiology', 0.5836742520332336)\n",
+ "\n",
+ "Normal (not Annoy-indexed) Neighbors\n",
+ "('science', 1.0)\n",
+ "('fiction', 0.7495021224021912)\n",
+ "('interdisciplinary', 0.6956626772880554)\n",
+ "('astrobiology', 0.6761417388916016)\n",
+ "('actuarial', 0.6735734343528748)\n",
+ "('robotics', 0.6708062887191772)\n",
+ "('sciences', 0.6689055562019348)\n",
+ "('scientific', 0.6639128923416138)\n",
+ "('psychohistory', 0.6622439622879028)\n",
+ "('bioethics', 0.6585155129432678)\n",
+ "('vernor', 0.6571990251541138)\n"
+ ]
+ }
+ ],
+ "source": [
+ "# 100 trees are being used in this example\n",
+ "annoy_index = AnnoyIndexer(model, 100)\n",
+ "# Derive the vector for the word \"science\" in our model\n",
+ "vector = model[\"science\"]\n",
+ "# The instance of AnnoyIndexer we just created is passed \n",
+ "approximate_neighbors = model.most_similar([vector], topn=11, indexer=annoy_index)\n",
+ "# Neatly print the approximate_neighbors and their corresponding cosine similarity values\n",
+ "print(\"Approximate Neighbors\")\n",
+ "for neighbor in approximate_neighbors:\n",
+ " print(neighbor)\n",
+ "\n",
+ "normal_neighbors = model.most_similar([vector], topn=11)\n",
+ "print(\"\\nNormal (not Annoy-indexed) Neighbors\")\n",
+ "for neighbor in normal_neighbors:\n",
+ " print(neighbor)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Analyzing the results"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "The closer the cosine similarity of a vector is to 1, the more similar that word is to our query, which was the vector for \"science\". There are some differences in the ranking of similar words and the set of words included within the 10 most similar words."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### 4. Verify & Evaluate performance"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Persisting Indexes\n",
+ "You can save and load your indexes from/to disk to prevent having to construct them each time. This will create two files on disk, _fname_ and _fname.d_. Both files are needed to correctly restore all attributes. Before loading an index, you will have to create an empty AnnoyIndexer object."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 11,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "fname = 'index'\n",
+ "\n",
+ "# Persist index to disk\n",
+ "annoy_index.save(fname)\n",
+ "\n",
+ "# Load index back\n",
+ "if os.path.exists(fname):\n",
+ " annoy_index2 = AnnoyIndexer()\n",
+ " annoy_index2.load(fname)\n",
+ " annoy_index2.model = model"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 12,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "('science', 1.0)\n",
+ "('interdisciplinary', 0.6099119782447815)\n",
+ "('astrobiology', 0.5975957810878754)\n",
+ "('actuarial', 0.596003383398056)\n",
+ "('robotics', 0.5942946970462799)\n",
+ "('sciences', 0.59312504529953)\n",
+ "('scientific', 0.5900688469409943)\n",
+ "('psychohistory', 0.5890524089336395)\n",
+ "('bioethics', 0.5867903232574463)\n",
+ "('cryobiology', 0.5854728817939758)\n",
+ "('xenobiology', 0.5836742520332336)\n"
+ ]
+ }
+ ],
+ "source": [
+ "# Results should be identical to above\n",
+ "vector = model[\"science\"]\n",
+ "approximate_neighbors2 = model.most_similar([vector], topn=11, indexer=annoy_index2)\n",
+ "for neighbor in approximate_neighbors2:\n",
+ " print(neighbor)\n",
+ " \n",
+ "assert approximate_neighbors == approximate_neighbors2"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Be sure to use the same model at load that was used originally, otherwise you will get unexpected behaviors."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Save memory by memory-mapping indices saved to disk\n",
+ "\n",
+ "Annoy library has a useful feature that indices can be memory-mapped from disk. It saves memory when the same index is used by several processes.\n",
+ "\n",
+ "Below are two snippets of code. First one has a separate index for each process. The second snipped shares the index between two processes via memory-mapping. The second example uses less total RAM as it is shared."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 13,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "# Remove verbosity from code below (if logging active)\n",
+ "\n",
+ "if LOGS:\n",
+ " logging.disable(logging.CRITICAL)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 14,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "from multiprocessing import Process\n",
+ "import os\n",
+ "import psutil"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Bad Example: Two processes load the Word2vec model from disk and create there own Annoy indices from that model. "
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 15,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Process Id: 6452\n",
+ "\n",
+ "Memory used by process 6452: pmem(rss=425226240, vms=3491692544, pfaults=149035, pageins=0) \n",
+ "---\n",
+ "Process Id: 6460\n",
+ "\n",
+ "Memory used by process 6460: pmem(rss=425136128, vms=3491692544, pfaults=149020, pageins=0) \n",
+ "---\n",
+ "CPU times: user 489 ms, sys: 204 ms, total: 693 ms\n",
+ "Wall time: 29.3 s\n"
+ ]
+ }
+ ],
+ "source": [
+ "%%time\n",
+ "\n",
+ "model.save('/tmp/mymodel')\n",
+ "\n",
+ "def f(process_id):\n",
+ " print ('Process Id: ', os.getpid())\n",
+ " process = psutil.Process(os.getpid())\n",
+ " new_model = Word2Vec.load('/tmp/mymodel')\n",
+ " vector = new_model[\"science\"]\n",
+ " annoy_index = AnnoyIndexer(new_model,100)\n",
+ " approximate_neighbors = new_model.most_similar([vector], topn=5, indexer=annoy_index)\n",
+ " print('\\nMemory used by process {}: '.format(os.getpid()), process.memory_info(), \"\\n---\")\n",
+ "\n",
+ "# Creating and running two parallel process to share the same index file.\n",
+ "p1 = Process(target=f, args=('1',))\n",
+ "p1.start()\n",
+ "p1.join()\n",
+ "p2 = Process(target=f, args=('2',))\n",
+ "p2.start()\n",
+ "p2.join()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Good example. Two processes load both the Word2vec model and index from disk and memory-map the index\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 16,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Process Id: 6461\n",
+ "\n",
+ "Memory used by process 6461: pmem(rss=357363712, vms=3576012800, pfaults=105041, pageins=0) \n",
+ "---\n",
+ "Process Id: 6462\n",
+ "\n",
+ "Memory used by process 6462: pmem(rss=357097472, vms=3576012800, pfaults=104995, pageins=0) \n",
+ "---\n",
+ "CPU times: user 509 ms, sys: 181 ms, total: 690 ms\n",
+ "Wall time: 2.61 s\n"
+ ]
+ }
+ ],
+ "source": [
+ "%%time\n",
+ "\n",
+ "model.save('/tmp/mymodel')\n",
+ "\n",
+ "def f(process_id):\n",
+ " print('Process Id: ', os.getpid())\n",
+ " process = psutil.Process(os.getpid())\n",
+ " new_model = Word2Vec.load('/tmp/mymodel')\n",
+ " vector = new_model[\"science\"]\n",
+ " annoy_index = AnnoyIndexer()\n",
+ " annoy_index.load('index')\n",
+ " annoy_index.model = new_model\n",
+ " approximate_neighbors = new_model.most_similar([vector], topn=5, indexer=annoy_index)\n",
+ " print('\\nMemory used by process {}: '.format(os.getpid()), process.memory_info(), \"\\n---\")\n",
+ "\n",
+ "# Creating and running two parallel process to share the same index file.\n",
+ "p1 = Process(target=f, args=('1',))\n",
+ "p1.start()\n",
+ "p1.join()\n",
+ "p2 = Process(target=f, args=('2',))\n",
+ "p2.start()\n",
+ "p2.join()"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### 5. Evaluate relationship of `num_trees` to initialization time and accuracy"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 17,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "import matplotlib.pyplot as plt\n",
+ "%matplotlib inline"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Build dataset of Initialization times and accuracy measures"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 18,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "exact_results = [element[0] for element in model.most_similar([model.wv.syn0norm[0]], topn=100)]\n",
+ "\n",
+ "x_values = []\n",
+ "y_values_init = []\n",
+ "y_values_accuracy = []\n",
+ "\n",
+ "for x in range(1, 300, 10):\n",
+ " x_values.append(x)\n",
+ " start_time = time.time()\n",
+ " annoy_index = AnnoyIndexer(model, x)\n",
+ " y_values_init.append(time.time() - start_time)\n",
+ " approximate_results = model.most_similar([model.wv.syn0norm[0]], topn=100, indexer=annoy_index)\n",
+ " top_words = [result[0] for result in approximate_results]\n",
+ " y_values_accuracy.append(len(set(top_words).intersection(exact_results)))"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Plot results"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 19,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1gAAAGoCAYAAABbkkSYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
+ "text/plain": [
+ "--- {}\".format(\", \".join(int_tokens), \n", + " \", \".join(diff_tokens)) \n", + " for (int_tokens, diff_tokens) in row] \n", + " for row in annotation]\n", + " \n", + " data = go.Heatmap(z=mdiff, colorscale='RdBu', text=annotation_html)\n", + " layout = go.Layout(width=950, height=950, title=title,\n", + " xaxis=dict(title=\"topic\"), yaxis=dict(title=\"topic\"))\n", + " py.iplot(dict(data=[data], layout=layout))" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "In gensim, you can visualise topic different with matrix and annotation. For this purposes, you can use method `diff` from LdaModel.\n", + "\n", + "This function return matrix with distances mdiff and matrix with annotations annotation. Read the docstring for more detailed info.\n", + "\n", + "In cells mdiff[i][j] we can see a distance between topic_i from the first model and topic_j from the second model.\n", + "\n", + "In cells annotation[i][j] we can see [tokens from intersection, tokens from difference] between topic_i from first model and topic_j from the second model." + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": { + "collapsed": true + }, + "outputs": [], + "source": [ + "LdaMulticore.diff?" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Case 1: How topics in ONE model correlate with each other" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Short description:\n", + "- x-axis - topic\n", + "- y-axis - topic\n", + "- almost red cell - strongly decorrelated topics\n", + "- almost blue cell - strongly correlated topics\n", + "\n", + "In an ideal world, we would like to see different topics decorrelated between themselves. In this case, our matrix would look like this:\n" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "" + ], + "text/plain": [ + "
![](\"jupyter_home.png\")
\n",
+ "\n",
+ "This will open the corresponding notebook in a separate tab. The Python code in the notebook can be executed by selecting/clicking on a cell and pressing SHIFT + ENTER.\n",
+ "\n",
+ "![](\"jupyter_execute_cell.png\")
\n",
+ "\n",
+ "**Note:** The order of cell execution matters. Be sure to run all of the code cells in order from top to bottom, you you might encounter errors."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Core Concepts and Simple Example\n",
+ "\n",
+ "This section introduces the basic concepts and terms needed to understand and use `gensim` and provides a simple usage example. In particular, we will build a model that measures the importance of a particular word.\n",
+ "\n",
+ "At a very high-level, `gensim` is a tool for discovering the semantic structure of documents by examining the patterns of words (or higher-level structures such as entire sentences or documents). `gensim` accomplishes this by taking a *corpus*, a collection of text documents, and producing a *vector* representation of the text in the corpus. The vector representation can then be used to train a *model*, which is an algorithms to create different representations of the data, which are usually more semantic. These three concepts are key to understanding how `gensim` works so let's take a moment to explain what each of them means. At the same time, we'll work through a simple example that illustrates each of them.\n",
+ "\n",
+ "### Corpus\n",
+ "\n",
+ "A *corpus* is a collection of digital documents. This collection is the input to `gensim` from which it will infer the structure of the documents, their topics, etc. The latent structure inferred from the corpus can later be used to assign topics to new documents which were not present in the training corpus. For this reason, we also refer to this collection as the *training corpus*. No human intervention (such as tagging the documents by hand) is required - the topic classification is [unsupervised](https://en.wikipedia.org/wiki/Unsupervised_learning.html).\n",
+ "\n",
+ "For our corpus, we'll use a list of 9 strings, each consisting of only a single sentence."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {
+ "collapsed": true
+ },
+ "outputs": [],
+ "source": [
+ "raw_corpus = [\"Human machine interface for lab abc computer applications\",\n",
+ " \"A survey of user opinion of computer system response time\",\n",
+ " \"The EPS user interface management system\",\n",
+ " \"System and human system engineering testing of EPS\", \n",
+ " \"Relation of user perceived response time to error measurement\",\n",
+ " \"The generation of random binary unordered trees\",\n",
+ " \"The intersection graph of paths in trees\",\n",
+ " \"Graph minors IV Widths of trees and well quasi ordering\",\n",
+ " \"Graph minors A survey\"]"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "This is a particularly small example of a corpus for illustration purposes. Another example could be a list of all the plays written by Shakespeare, list of all wikipedia articles, or all tweets by a particular person of interest.\n",
+ "\n",
+ "After collecting our corpus, there are typically a number of preprocessing steps we want to undertake. We'll keep it simple and just remove some commonly used English words (such as 'the') and words that occur only once in the corpus. In the process of doing so, we'll [tokenize](https://en.wikipedia.org/wiki/Lexical_analysis#Tokenization)) our data. Tokenization breaks up the documents into words (in this case using space as a delimiter)."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 2,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[['human', 'interface', 'computer'],\n",
+ " ['survey', 'user', 'computer', 'system', 'response', 'time'],\n",
+ " ['eps', 'user', 'interface', 'system'],\n",
+ " ['system', 'human', 'system', 'eps'],\n",
+ " ['user', 'response', 'time'],\n",
+ " ['trees'],\n",
+ " ['graph', 'trees'],\n",
+ " ['graph', 'minors', 'trees'],\n",
+ " ['graph', 'minors', 'survey']]"
+ ]
+ },
+ "execution_count": 2,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "# Create a set of frequent words\n",
+ "stoplist = set('for a of the and to in'.split(' '))\n",
+ "# Lowercase each document, split it by white space and filter out stopwords\n",
+ "texts = [[word for word in document.lower().split() if word not in stoplist]\n",
+ " for document in raw_corpus]\n",
+ "\n",
+ "# Count word frequencies\n",
+ "from collections import defaultdict\n",
+ "frequency = defaultdict(int)\n",
+ "for text in texts:\n",
+ " for token in text:\n",
+ " frequency[token] += 1\n",
+ "\n",
+ "# Only keep words that appear more than once\n",
+ "processed_corpus = [[token for token in text if frequency[token] > 1] for text in texts]\n",
+ "processed_corpus"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Before proceeding, we want to associate each word in the corpus with a unique integer ID. We can do this using the `gensim.corpora.Dictionary` class. This dictionary defines the vocabulary of all words that our processing knows about."
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 3,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Dictionary(12 unique tokens: [u'minors', u'graph', u'system', u'trees', u'eps']...)\n"
+ ]
+ }
+ ],
+ "source": [
+ "from gensim import corpora\n",
+ "\n",
+ "dictionary = corpora.Dictionary(processed_corpus)\n",
+ "print(dictionary)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Because our corpus is small, there is only 12 different tokens in this `Dictionary`. For larger corpuses, dictionaries that contains hundreds of thousands of tokens are quite common."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Vector\n",
+ "\n",
+ "To infer the latent structure in our corpus we need a way to represent documents that we can manipulate mathematically. One approach is to represent each document as a vector. There are various approaches for creating a vector representation of a document but a simple example is the *bag-of-words model*. Under the bag-of-words model each document is represented by a vector containing the frequency counts of each word in the dictionary. For example, given a dictionary containing the words `['coffee', 'milk', 'sugar', 'spoon']` a document consisting of the string `\"coffee milk coffee\"` could be represented by the vector `[2, 1, 0, 0]` where the entries of the vector are (in order) the occurrences of \"coffee\", \"milk\", \"sugar\" and \"spoon\" in the document. The length of the vector is the number of entries in the dictionary. One of the main properties of the bag-of-words model is that it completely ignores the order of the tokens in the document that is encoded, which is where the name bag-of-words comes from.\n",
+ "\n",
+ "Our processed corpus has 12 unique words in it, which means that each document will be represented by a 12-dimensional vector under the bag-of-words model. We can use the dictionary to turn tokenized documents into these 12-dimensional vectors. We can see what these IDs correspond to:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 4,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "{u'minors': 11, u'graph': 10, u'system': 6, u'trees': 9, u'eps': 8, u'computer': 1, u'survey': 5, u'user': 7, u'human': 2, u'time': 4, u'interface': 0, u'response': 3}\n"
+ ]
+ }
+ ],
+ "source": [
+ "print(dictionary.token2id)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "For example, suppose we wanted to vectorize the phrase \"Human computer interaction\" (note that this phrase was not in our original corpus). We can create the bag-of-word representation for a document using the `doc2bow` method of the dictionary, which returns a sparse representation of the word counts:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 5,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[(1, 1), (2, 1)]"
+ ]
+ },
+ "execution_count": 5,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "new_doc = \"Human computer interaction\"\n",
+ "new_vec = dictionary.doc2bow(new_doc.lower().split())\n",
+ "new_vec"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "The first entry in each tuple corresponds to the ID of the token in the dictionary, the second corresponds to the count of this token."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Note that \"interaction\" did not occur in the original corpus and so it was not included in the vectorization. Also note that this vector only contains entries for words that actually appeared in the document. Because any given document will only contain a few words out of the many words in the dictionary, words that do not appear in the vectorization are represented as implicitly zero as a space saving measure.\n",
+ "\n",
+ "We can convert our entire original corpus to a list of vectors:"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 6,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[[(0, 1), (1, 1), (2, 1)],\n",
+ " [(1, 1), (3, 1), (4, 1), (5, 1), (6, 1), (7, 1)],\n",
+ " [(0, 1), (6, 1), (7, 1), (8, 1)],\n",
+ " [(2, 1), (6, 2), (8, 1)],\n",
+ " [(3, 1), (4, 1), (7, 1)],\n",
+ " [(9, 1)],\n",
+ " [(9, 1), (10, 1)],\n",
+ " [(9, 1), (10, 1), (11, 1)],\n",
+ " [(5, 1), (10, 1), (11, 1)]]"
+ ]
+ },
+ "execution_count": 6,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "bow_corpus = [dictionary.doc2bow(text) for text in processed_corpus]\n",
+ "bow_corpus"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Note that while this list lives entirely in memory, in most applications you will want a more scalable solution. Luckily, `gensim` allows you to use any iterator that returns a single document vector at a time. See the documentation for more details.\n",
+ "\n",
+ "### Model\n",
+ "\n",
+ "Now that we have vectorized our corpus we can begin to transform it using *models*. We use model as an abstract term referring to a transformation from one document representation to another. In `gensim`, documents are represented as vectors so a model can be thought of as a transformation between two [vector spaces](https://en.wikipedia.org/wiki/Vector_space). The details of this transformation are learned from the training corpus.\n",
+ "\n",
+ "One simple example of a model is [tf-idf](https://en.wikipedia.org/wiki/Tf%E2%80%93idf). The tf-idf model transforms vectors from the bag-of-words representation to a vector space, where the frequency counts are weighted according to the relative rarity of each word in the corpus.\n",
+ "\n",
+ "Here's a simple example. Let's initialize the tf-idf model, training it on our corpus and transforming the string \"system minors\":"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 7,
+ "metadata": {
+ "collapsed": false
+ },
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "[(6, 0.5898341626740045), (11, 0.8075244024440723)]"
+ ]
+ },
+ "execution_count": 7,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
+ "source": [
+ "from gensim import models\n",
+ "# train the model\n",
+ "tfidf = models.TfidfModel(bow_corpus)\n",
+ "# transform the \"system minors\" string\n",
+ "tfidf[dictionary.doc2bow(\"system minors\".lower().split())]"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "The `tfidf` model again returns a list of tuples, where the first entry is the token ID and the second entry is the tf-idf weighting. Note that the ID corresponding to \"system\" (which occurred 4 times in the original corpus) has been weighted lower than the ID corresponding to \"minors\" (which only occurred twice).\n",
+ "\n",
+ "`gensim` offers a number of different models/transformations. See [Transformations and Topics](https://github.com/RaRe-Technologies/gensim/blob/develop/docs/notebooks/Topics_and_Transformations.ipynb) for details."
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Next Steps\n",
+ "\n",
+ "Interested in learning more about `gensim`? Please read through the following notebooks.\n",
+ "\n",
+ "1. [Corpora_and_Vector_Spaces.ipynb](https://github.com/RaRe-Technologies/gensim/blob/develop/docs/notebooks/Corpora_and_Vector_Spaces.ipynb)\n",
+ "2. [word2vec.ipynb](https://github.com/RaRe-Technologies/gensim/blob/develop/docs/notebooks/word2vec.ipynb)"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python [Root]",
+ "language": "python",
+ "name": "Python [Root]"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.5.1"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 0
+}
diff --git a/gensim/__init__.py b/gensim/__init__.py
index 88e2e391f7..1999fc72ce 100644
--- a/gensim/__init__.py
+++ b/gensim/__init__.py
@@ -6,7 +6,7 @@
from gensim import parsing, matutils, interfaces, corpora, models, similarities, summarization
import logging
-__version__ = '2.0.0'
+__version__ = '2.1.0'
class NullHandler(logging.Handler):
"""For python versions <= 2.6; same as `logging.NullHandler` in 2.7."""
diff --git a/gensim/corpora/dictionary.py b/gensim/corpora/dictionary.py
index 484684c26d..1ff89a5b31 100644
--- a/gensim/corpora/dictionary.py
+++ b/gensim/corpora/dictionary.py
@@ -194,9 +194,11 @@ def filter_extremes(self, no_below=5, no_above=0.5, keep_n=100000, keep_tokens=N
# determine which tokens to keep
if keep_tokens:
keep_ids = [self.token2id[v] for v in keep_tokens if v in self.token2id]
- good_ids = (v for v in itervalues(self.token2id)
- if no_below <= self.dfs.get(v, 0) <= no_above_abs
- or v in keep_ids)
+ good_ids = (
+ v for v in itervalues(self.token2id)
+ if no_below <= self.dfs.get(v, 0) <= no_above_abs
+ or v in keep_ids
+ )
else:
good_ids = (
v for v in itervalues(self.token2id)
diff --git a/gensim/corpora/wikicorpus.py b/gensim/corpora/wikicorpus.py
index fb402da517..f5f30281eb 100755
--- a/gensim/corpora/wikicorpus.py
+++ b/gensim/corpora/wikicorpus.py
@@ -173,7 +173,7 @@ def tokenize(content):
"""
# TODO maybe ignore tokens with non-latin characters? (no chinese, arabic, russian etc.)
return [
- token.encode('utf8') for token in utils.tokenize(content, lower=True, errors='ignore')
+ utils.to_unicode(token) for token in utils.tokenize(content, lower=True, errors='ignore')
if 2 <= len(token) <= 15 and not token.startswith('_')
]
diff --git a/gensim/matutils.py b/gensim/matutils.py
index 93c750efd8..057e65a52f 100644
--- a/gensim/matutils.py
+++ b/gensim/matutils.py
@@ -532,6 +532,19 @@ def jaccard(vec1, vec2):
return 1 - float(len(intersection)) / float(len(union))
+def jaccard_distance(set1, set2):
+ """
+ Calculate a distance between set representation (1 minus the intersection divided by union).
+ Return a value in range <0, 1> where values closer to 0 mean smaller distance and thus higher similarity.
+ """
+
+ union_cardinality = len(set1 | set2)
+ if union_cardinality == 0: # Both sets are empty
+ return 1.
+
+ return 1. - float(len(set1 & set2)) / float(union_cardinality)
+
+
def dirichlet_expectation(alpha):
"""
For a vector `theta~Dir(alpha)`, compute `E[log(theta)]`.
diff --git a/gensim/models/doc2vec.py b/gensim/models/doc2vec.py
index e57945ce67..da1f5d01a8 100644
--- a/gensim/models/doc2vec.py
+++ b/gensim/models/doc2vec.py
@@ -580,7 +580,7 @@ def __init__(self, documents=None, dm_mean=None,
need about 1GB of RAM. Set to `None` for no limit (default).
`sample` = threshold for configuring which higher-frequency words are randomly downsampled;
- default is 0 (off), useful value is 1e-5.
+ default is 1e-3, useful value is 1e-5.
`workers` = use this many worker threads to train the model (=faster training with multicore machines).
@@ -597,7 +597,7 @@ def __init__(self, documents=None, dm_mean=None,
`dm_concat` = if 1, use concatenation of context vectors rather than sum/average;
default is 0 (off). Note concatenation results in a much-larger model, as the input
- is no longer the size of one (sampled or arithmatically combined) word vector, but the
+ is no longer the size of one (sampled or arithmetically combined) word vector, but the
size of the tag(s) and all words in the context strung together.
`dm_tag_count` = expected constant number of document tags per document, when using
@@ -614,9 +614,13 @@ def __init__(self, documents=None, dm_mean=None,
of the model.
"""
+ if 'sentences' in kwargs:
+ raise DeprecationWarning("'sentences' in doc2vec was renamed to 'documents'. Please use documents parameter.")
+
super(Doc2Vec, self).__init__(
sg=(1 + dm) % 2,
- null_word=dm_concat, **kwargs)
+ null_word=dm_concat,
+ **kwargs)
self.load = call_on_class_only
diff --git a/gensim/models/hdpmodel.py b/gensim/models/hdpmodel.py
index 2c74d15a15..6937d928d4 100755
--- a/gensim/models/hdpmodel.py
+++ b/gensim/models/hdpmodel.py
@@ -33,7 +33,9 @@
from __future__ import with_statement
-import logging, time
+import logging
+import time
+import warnings
import numpy as np
from scipy.special import gammaln, psi # gamma function utils
@@ -47,7 +49,6 @@
meanchangethresh = 0.00001
rhot_bound = 0.0
-
def expect_log_sticks(sticks):
"""
For stick-breaking hdp, return the E[log(sticks)]
@@ -436,7 +437,7 @@ def update_expectations(self):
self.m_timestamp[:] = self.m_updatect
self.m_status_up_to_date = True
- def show_topic(self, topic_id, num_words=20, log=False, formatted=False):
+ def show_topic(self, topic_id, topn=20, log=False, formatted=False, num_words=None):
"""
Print the `num_words` most probable words for topic `topic_id`.
@@ -444,12 +445,17 @@ def show_topic(self, topic_id, num_words=20, log=False, formatted=False):
`False` as lists of (weight, word) pairs.
"""
+ if num_words is not None: # deprecated num_words is used
+ logger.warning("The parameter num_words for show_topic() would be deprecated in the updated version.")
+ logger.warning("Please use topn instead.")
+ topn = num_words
+
if not self.m_status_up_to_date:
self.update_expectations()
betas = self.m_lambda + self.m_eta
hdp_formatter = HdpTopicFormatter(self.id2word, betas)
- return hdp_formatter.show_topic(topic_id, num_words, log, formatted)
-
+ return hdp_formatter.show_topic(topic_id, topn, log, formatted)
+
def show_topics(self, num_topics=20, num_words=20, log=False, formatted=True):
"""
Print the `num_words` most probable words for `num_topics` number of topics.
@@ -608,10 +614,17 @@ def show_topics(self, num_topics=10, num_words=10, log=False, formatted=True):
return shown
- def print_topic(self, topic_id, num_words):
- return self.show_topic(topic_id, num_words, formatted=True)
+ def print_topic(self, topic_id, topn= None, num_words=None):
+ if num_words is not None: # deprecated num_words is used
+ warnings.warn("The parameter num_words for print_topic() would be deprecated in the updated version. Please use topn instead.")
+ topn = num_words
+
+ return self.show_topic(topic_id, topn, formatted=True)
- def show_topic(self, topic_id, num_words, log=False, formatted=False):
+ def show_topic(self, topic_id, topn=20, log=False, formatted=False, num_words= None,):
+ if num_words is not None: # deprecated num_words is used
+ warnings.warn("The parameter num_words for show_topic() would be deprecated in the updated version. Please use topn instead.")
+ topn = num_words
lambdak = list(self.data[topic_id, :])
lambdak = lambdak / sum(lambdak)
@@ -619,7 +632,7 @@ def show_topic(self, topic_id, num_words, log=False, formatted=False):
temp = zip(lambdak, xrange(len(lambdak)))
temp = sorted(temp, key=lambda x: x[0], reverse=True)
- topic_terms = self.show_topic_terms(temp, num_words)
+ topic_terms = self.show_topic_terms(temp, topn)
if formatted:
topic = self.format_topic(topic_id, topic_terms)
diff --git a/gensim/models/keyedvectors.py b/gensim/models/keyedvectors.py
index 4d187bd7dd..e55e553d43 100644
--- a/gensim/models/keyedvectors.py
+++ b/gensim/models/keyedvectors.py
@@ -21,7 +21,7 @@
The vectors can also be instantiated from an existing file on disk in the original Google's word2vec C format as a KeyedVectors instance::
- >>> from gensim.keyedvectors import KeyedVectors
+ >>> from gensim.models.keyedvectors import KeyedVectors
>>> word_vectors = KeyedVectors.load_word2vec_format('/tmp/vectors.txt', binary=False) # C text format
>>> word_vectors = KeyedVectors.load_word2vec_format('/tmp/vectors.bin', binary=True) # C binary format
diff --git a/gensim/models/ldamodel.py b/gensim/models/ldamodel.py
index 67398ab099..77f7e68be9 100755
--- a/gensim/models/ldamodel.py
+++ b/gensim/models/ldamodel.py
@@ -33,11 +33,13 @@
import logging
import numpy as np
import numbers
+from random import sample
import os
from gensim import interfaces, utils, matutils
from gensim.matutils import dirichlet_expectation
from gensim.models import basemodel
+from gensim.matutils import kullback_leibler, hellinger, jaccard_distance
from itertools import chain
from scipy.special import gammaln, psi # gamma function utils
@@ -965,6 +967,74 @@ def get_term_topics(self, word_id, minimum_probability=None):
return values
+ def diff(self, other, distance="kulback_leibler", num_words=100, n_ann_terms=10, normed=True):
+ """
+ Calculate difference topic2topic between two Lda models
+ `other` instances of `LdaMulticore` or `LdaModel`
+ `distance` is function that will be applied to calculate difference between any topic pair.
+ Available values: `kulback_leibler`, `hellinger` and `jaccard`
+ `num_words` is quantity of most relevant words that used if distance == `jaccard` (also used for annotation)
+ `n_ann_terms` is max quantity of words in intersection/symmetric difference between topics (used for annotation)
+ Returns a matrix Z with shape (m1.num_topics, m2.num_topics), where Z[i][j] - difference between topic_i and topic_j
+ and matrix annotation with shape (m1.num_topics, m2.num_topics, 2, None),
+ where
+ annotation[i][j] = [[`int_1`, `int_2`, ...], [`diff_1`, `diff_2`, ...]] and
+ `int_k` is word from intersection of `topic_i` and `topic_j` and
+ `diff_l` is word from symmetric difference of `topic_i` and `topic_j`
+ `normed` is a flag. If `true`, matrix Z will be normalized
+ Example:
+ >>> m1, m2 = LdaMulticore.load(path_1), LdaMulticore.load(path_2)
+ >>> mdiff, annotation = m1.diff(m2)
+ >>> print(mdiff) # get matrix with difference for each topic pair from `m1` and `m2`
+ >>> print(annotation) # get array with positive/negative words for each topic pair from `m1` and `m2`
+ """
+
+ distances = {
+ "kulback_leibler": kullback_leibler,
+ "hellinger": hellinger,
+ "jaccard": jaccard_distance,
+ }
+
+ if distance not in distances:
+ valid_keys = ", ".join("`{}`".format(x) for x in distances.keys())
+ raise ValueError("Incorrect distance, valid only {}".format(valid_keys))
+
+ if not isinstance(other, self.__class__):
+ raise ValueError("The parameter `other` must be of type `{}`".format(self.__name__))
+
+ distance_func = distances[distance]
+ d1, d2 = self.state.get_lambda(), other.state.get_lambda()
+ t1_size, t2_size = d1.shape[0], d2.shape[0]
+
+ fst_topics = [{w for (w, _) in self.show_topic(topic, topn=num_words)} for topic in xrange(t1_size)]
+ snd_topics = [{w for (w, _) in other.show_topic(topic, topn=num_words)} for topic in xrange(t2_size)]
+
+ if distance == "jaccard":
+ d1, d2 = fst_topics, snd_topics
+
+ z = np.zeros((t1_size, t2_size))
+ for topic1 in range(t1_size):
+ for topic2 in range(t2_size):
+ z[topic1][topic2] = distance_func(d1[topic1], d2[topic2])
+
+ if normed:
+ if np.abs(np.max(z)) > 1e-8:
+ z /= np.max(z)
+
+ annotation = [[None] * t1_size for _ in range(t2_size)]
+
+ for topic1 in range(t1_size):
+ for topic2 in range(t2_size):
+ pos_tokens = fst_topics[topic1] & snd_topics[topic2]
+ neg_tokens = fst_topics[topic1].symmetric_difference(snd_topics[topic2])
+
+ pos_tokens = sample(pos_tokens, min(len(pos_tokens), n_ann_terms))
+ neg_tokens = sample(neg_tokens, min(len(neg_tokens), n_ann_terms))
+
+ annotation[topic1][topic2] = [pos_tokens, neg_tokens]
+
+ return z, annotation
+
def __getitem__(self, bow, eps=None):
"""
Return topic distribution for the given document `bow`, as a list of
@@ -1023,9 +1093,9 @@ def save(self, fname, ignore=['state', 'dispatcher'], separately=None, *args, **
separately_explicit = ['expElogbeta', 'sstats']
# Also add 'alpha' and 'eta' to separately list if they are set 'auto' or some
# array manually.
- if (isinstance(self.alpha, six.string_types) and self.alpha == 'auto') or len(self.alpha.shape) != 1:
+ if (isinstance(self.alpha, six.string_types) and self.alpha == 'auto') or (isinstance(self.alpha, np.ndarray) and len(self.alpha.shape) != 1):
separately_explicit.append('alpha')
- if (isinstance(self.eta, six.string_types) and self.eta == 'auto') or len(self.eta.shape) != 1:
+ if (isinstance(self.eta, six.string_types) and self.eta == 'auto') or (isinstance(self.eta, np.ndarray) and len(self.eta.shape) != 1):
separately_explicit.append('eta')
# Merge separately_explicit with separately.
if separately:
@@ -1049,18 +1119,28 @@ def load(cls, fname, *args, **kwargs):
"""
kwargs['mmap'] = kwargs.get('mmap', None)
result = super(LdaModel, cls).load(fname, *args, **kwargs)
+
+ # check if `random_state` attribute has been set after main pickle load
+ # if set -> the model to be loaded was saved using a >= 0.13.2 version of Gensim
+ # if not set -> the model to be loaded was saved using a < 0.13.2 version of Gensim, so set `random_state` as the default value
+ if not hasattr(result, 'random_state'):
+ result.random_state = utils.get_random_state(None) # using default value `get_random_state(None)`
+ logging.warning("random_state not set so using default value")
+
state_fname = utils.smart_extension(fname, '.state')
try:
result.state = super(LdaModel, cls).load(state_fname, *args, **kwargs)
except Exception as e:
logging.warning("failed to load state from %s: %s", state_fname, e)
+
id2word_fname = utils.smart_extension(fname, '.id2word')
+ # check if `id2word_fname` file is present on disk
+ # if present -> the model to be loaded was saved using a >= 0.13.2 version of Gensim, so set `result.id2word` using the `id2word_fname` file
+ # if not present -> the model to be loaded was saved using a < 0.13.2 version of Gensim, so `result.id2word` already set after the main pickle load
if (os.path.isfile(id2word_fname)):
try:
result.id2word = utils.unpickle(id2word_fname)
except Exception as e:
logging.warning("failed to load id2word dictionary from %s: %s", id2word_fname, e)
- else:
- result.id2word = None
return result
# endclass LdaModel
diff --git a/gensim/models/phrases.py b/gensim/models/phrases.py
index 46e915af17..be735b865a 100644
--- a/gensim/models/phrases.py
+++ b/gensim/models/phrases.py
@@ -250,7 +250,7 @@ def export_phrases(self, sentences, out_delimiter=b' ', as_tuples=False):
yield (out_delimiter.join((word_a, word_b)), score)
last_bigram = True
continue
- last_bigram = False
+ last_bigram = False
def __getitem__(self, sentence):
"""
diff --git a/gensim/models/tfidfmodel.py b/gensim/models/tfidfmodel.py
index 592cc9537c..1b4aea863b 100644
--- a/gensim/models/tfidfmodel.py
+++ b/gensim/models/tfidfmodel.py
@@ -28,8 +28,7 @@ def precompute_idfs(wglobal, dfs, total_docs):
"""Precompute the inverse document frequency mapping for all terms."""
# not strictly necessary and could be computed on the fly in TfidfModel__getitem__.
# this method is here just to speed things up a little.
- return dict((termid, wglobal(df, total_docs))
- for termid, df in iteritems(dfs))
+ return dict((termid, wglobal(df, total_docs)) for termid, df in iteritems(dfs))
class TfidfModel(interfaces.TransformationABC):
@@ -49,8 +48,9 @@ class TfidfModel(interfaces.TransformationABC):
Model persistency is achieved via its load/save methods.
"""
- def __init__(self, corpus=None, id2word=None, dictionary=None,
- wlocal=utils.identity, wglobal=df2idf, normalize=True):
+ def __init__(
+ self, corpus=None, id2word=None, dictionary=None,
+ wlocal=utils.identity, wglobal=df2idf, normalize=True):
"""
Compute tf-idf by multiplying a local component (term frequency) with a
global component (inverse document frequency), and normalizing
@@ -87,11 +87,13 @@ def __init__(self, corpus=None, id2word=None, dictionary=None,
# statistics we need to construct the IDF mapping. we can skip the
# step that goes through the corpus (= an optimization).
if corpus is not None:
- logger.warning("constructor received both corpus and explicit "
- "inverse document frequencies; ignoring the corpus")
+ logger.warning(
+ "constructor received both corpus and explicit inverse document frequencies; ignoring the corpus")
self.num_docs, self.num_nnz = dictionary.num_docs, dictionary.num_nnz
self.dfs = dictionary.dfs.copy()
self.idfs = precompute_idfs(self.wglobal, self.dfs, self.num_docs)
+ if id2word is None:
+ self.id2word = dictionary
elif corpus is not None:
self.initialize(corpus)
else:
@@ -114,7 +116,7 @@ def initialize(self, corpus):
numnnz, docno = 0, -1
for docno, bow in enumerate(corpus):
if docno % 10000 == 0:
- logger.info("PROGRESS: processing document #%i" % docno)
+ logger.info("PROGRESS: processing document #%i", docno)
numnnz += len(bow)
for termid, _ in bow:
dfs[termid] = dfs.get(termid, 0) + 1
@@ -126,8 +128,9 @@ def initialize(self, corpus):
# and finally compute the idf weights
n_features = max(dfs) if dfs else 0
- logger.info("calculating IDF weights for %i documents and %i features (%i matrix non-zeros)" %
- (self.num_docs, n_features, self.num_nnz))
+ logger.info(
+ "calculating IDF weights for %i documents and %i features (%i matrix non-zeros)",
+ self.num_docs, n_features, self.num_nnz)
self.idfs = precompute_idfs(self.wglobal, self.dfs, self.num_docs)
@@ -142,8 +145,10 @@ def __getitem__(self, bow, eps=1e-12):
# unknown (new) terms will be given zero weight (NOT infinity/huge weight,
# as strict application of the IDF formula would dictate)
- vector = [(termid, self.wlocal(tf) * self.idfs.get(termid))
- for termid, tf in bow if self.idfs.get(termid, 0.0) != 0.0]
+ vector = [
+ (termid, self.wlocal(tf) * self.idfs.get(termid))
+ for termid, tf in bow if self.idfs.get(termid, 0.0) != 0.0
+ ]
# and finally, normalize the vector either to unit length, or use a
# user-defined normalization function
diff --git a/gensim/models/word2vec.py b/gensim/models/word2vec.py
index aaa15660a1..85aeefe173 100644
--- a/gensim/models/word2vec.py
+++ b/gensim/models/word2vec.py
@@ -205,7 +205,6 @@ def score_sentence_sg(model, sentence, work=None):
will use the optimized version from word2vec_inner instead.
"""
-
log_prob_sentence = 0.0
if model.negative:
raise RuntimeError("scoring is only available for HS=True")
@@ -483,7 +482,6 @@ def __init__(
logger.warning("The rule, if given, is only used to prune vocabulary during build_vocab() and is not stored as part of the model. ")
logger.warning("Model initialized without sentences. trim_rule provided, if any, will be ignored." )
-
def initialize_word_vectors(self):
self.wv = KeyedVectors()
@@ -1208,7 +1206,6 @@ def most_similar(self, positive=[], negative=[], topn=10, restrict_vocab=None, i
Deprecated. Use self.wv.most_similar() instead.
Refer to the documentation for `gensim.models.KeyedVectors.most_similar`
"""
-
return self.wv.most_similar(positive, negative, topn, restrict_vocab, indexer)
def wmdistance(self, document1, document2):
@@ -1216,7 +1213,6 @@ def wmdistance(self, document1, document2):
Deprecated. Use self.wv.wmdistance() instead.
Refer to the documentation for `gensim.models.KeyedVectors.wmdistance`
"""
-
return self.wv.wmdistance(document1, document2)
def most_similar_cosmul(self, positive=[], negative=[], topn=10):
@@ -1224,7 +1220,6 @@ def most_similar_cosmul(self, positive=[], negative=[], topn=10):
Deprecated. Use self.wv.most_similar_cosmul() instead.
Refer to the documentation for `gensim.models.KeyedVectors.most_similar_cosmul`
"""
-
return self.wv.most_similar_cosmul(positive, negative, topn)
def similar_by_word(self, word, topn=10, restrict_vocab=None):
@@ -1232,7 +1227,6 @@ def similar_by_word(self, word, topn=10, restrict_vocab=None):
Deprecated. Use self.wv.similar_by_word() instead.
Refer to the documentation for `gensim.models.KeyedVectors.similar_by_word`
"""
-
return self.wv.similar_by_word(word, topn, restrict_vocab)
def similar_by_vector(self, vector, topn=10, restrict_vocab=None):
@@ -1240,7 +1234,6 @@ def similar_by_vector(self, vector, topn=10, restrict_vocab=None):
Deprecated. Use self.wv.similar_by_vector() instead.
Refer to the documentation for `gensim.models.KeyedVectors.similar_by_vector`
"""
-
return self.wv.similar_by_vector(vector, topn, restrict_vocab)
def doesnt_match(self, words):
@@ -1248,7 +1241,6 @@ def doesnt_match(self, words):
Deprecated. Use self.wv.doesnt_match() instead.
Refer to the documentation for `gensim.models.KeyedVectors.doesnt_match`
"""
-
return self.wv.doesnt_match(words)
def __getitem__(self, words):
@@ -1256,7 +1248,6 @@ def __getitem__(self, words):
Deprecated. Use self.wv.__getitem__() instead.
Refer to the documentation for `gensim.models.KeyedVectors.__getitem__`
"""
-
return self.wv.__getitem__(words)
def __contains__(self, word):
@@ -1264,7 +1255,6 @@ def __contains__(self, word):
Deprecated. Use self.wv.__contains__() instead.
Refer to the documentation for `gensim.models.KeyedVectors.__contains__`
"""
-
return self.wv.__contains__(word)
def similarity(self, w1, w2):
@@ -1272,7 +1262,6 @@ def similarity(self, w1, w2):
Deprecated. Use self.wv.similarity() instead.
Refer to the documentation for `gensim.models.KeyedVectors.similarity`
"""
-
return self.wv.similarity(w1, w2)
def n_similarity(self, ws1, ws2):
@@ -1280,7 +1269,6 @@ def n_similarity(self, ws1, ws2):
Deprecated. Use self.wv.n_similarity() instead.
Refer to the documentation for `gensim.models.KeyedVectors.n_similarity`
"""
-
return self.wv.n_similarity(ws1, ws2)
def predict_output_word(self, context_words_list, topn=10):
@@ -1347,7 +1335,6 @@ def log_evaluate_word_pairs(pearson, spearman, oov, pairs):
Deprecated. Use self.wv.log_evaluate_word_pairs() instead.
Refer to the documentation for `gensim.models.KeyedVectors.log_evaluate_word_pairs`
"""
-
return KeyedVectors.log_evaluate_word_pairs(pearson, spearman, oov, pairs)
def evaluate_word_pairs(self, pairs, delimiter='\t', restrict_vocab=300000, case_insensitive=True, dummy4unknown=False):
@@ -1355,7 +1342,6 @@ def evaluate_word_pairs(self, pairs, delimiter='\t', restrict_vocab=300000, case
Deprecated. Use self.wv.evaluate_word_pairs() instead.
Refer to the documentation for `gensim.models.KeyedVectors.evaluate_word_pairs`
"""
-
return self.wv.evaluate_word_pairs(pairs, delimiter, restrict_vocab, case_insensitive, dummy4unknown)
def __str__(self):
diff --git a/gensim/models/wrappers/dtmmodel.py b/gensim/models/wrappers/dtmmodel.py
index a953ce858a..94a2e5eb1a 100644
--- a/gensim/models/wrappers/dtmmodel.py
+++ b/gensim/models/wrappers/dtmmodel.py
@@ -22,6 +22,7 @@
import logging
import random
+import warnings
import tempfile
import os
from subprocess import PIPE
@@ -93,7 +94,7 @@ def __init__(
lencorpus = sum(1 for _ in corpus)
if lencorpus == 0:
raise ValueError("cannot compute DTM over an empty corpus")
- if model == "fixed" and any([i == 0 for i in [len(text) for text in corpus.get_texts()]]):
+ if model == "fixed" and any(not text for text in corpus):
raise ValueError("""There is a text without words in the input corpus.
This breaks method='fixed' (The DIM model).""")
if lencorpus != sum(time_slices):
@@ -283,12 +284,17 @@ def show_topics(self, num_topics=10, times=5, num_words=10, log=False, formatted
# topic))
return shown
- def show_topic(self, topicid, time, num_words=50):
+ def show_topic(self, topicid, time, topn=50, num_words=None):
"""
Return `num_words` most probable words for the given `topicid`, as a list of
`(word_probability, word)` 2-tuples.
"""
+ if num_words is not None: # deprecated num_words is used
+ logger.warning("The parameter num_words for show_topic() would be deprecated in the updated version.")
+ logger.warning("Please use topn instead.")
+ topn = num_words
+
topics = self.lambda_[:, :, time]
topic = topics[topicid]
# liklihood to probability
@@ -296,13 +302,17 @@ def show_topic(self, topicid, time, num_words=50):
# normalize to probability dist
topic = topic / topic.sum()
# sort according to prob
- bestn = matutils.argsort(topic, num_words, reverse=True)
+ bestn = matutils.argsort(topic, topn, reverse=True)
beststr = [(topic[id], self.id2word[id]) for id in bestn]
return beststr
- def print_topic(self, topicid, time, num_words=10):
+ def print_topic(self, topicid, time, topn=10, num_words=None):
"""Return the given topic, formatted as a string."""
- return ' + '.join(['%.3f*%s' % v for v in self.show_topic(topicid, time, num_words)])
+ if num_words is not None: # deprecated num_words is used
+ warnings.warn("The parameter num_words for print_topic() would be deprecated in the updated version. Please use topn instead.")
+ topn = num_words
+
+ return ' + '.join(['%.3f*%s' % v for v in self.show_topic(topicid, time, topn)])
def dtm_vis(self, corpus, time):
"""
diff --git a/gensim/models/wrappers/fasttext.py b/gensim/models/wrappers/fasttext.py
index 771c09b3de..926e994eaf 100644
--- a/gensim/models/wrappers/fasttext.py
+++ b/gensim/models/wrappers/fasttext.py
@@ -42,6 +42,8 @@
logger = logging.getLogger(__name__)
+FASTTEXT_FILEFORMAT_MAGIC = 793712314
+
class FastTextKeyedVectors(KeyedVectors):
"""
@@ -257,7 +259,15 @@ def load_binary_data(self, model_binary_file, encoding='utf8'):
self.load_vectors(f)
def load_model_params(self, file_handle):
- (dim, ws, epoch, minCount, neg, _, loss, model, bucket, minn, maxn, _, t) = self.struct_unpack(file_handle, '@12i1d')
+ magic, version = self.struct_unpack(file_handle, '@2i')
+ if magic == FASTTEXT_FILEFORMAT_MAGIC: # newer format
+ self.new_format = True
+ dim, ws, epoch, minCount, neg, _, loss, model, bucket, minn, maxn, _, t = self.struct_unpack(file_handle, '@12i1d')
+ else: # older format
+ self.new_format = False
+ dim = magic
+ ws = version
+ epoch, minCount, neg, _, loss, model, bucket, minn, maxn, _, t = self.struct_unpack(file_handle, '@10i1d')
# Parameters stored by [Args::save](https://github.com/facebookresearch/fastText/blob/master/src/args.cc)
self.vector_size = dim
self.window = ws
@@ -272,11 +282,13 @@ def load_model_params(self, file_handle):
self.sample = t
def load_dict(self, file_handle, encoding='utf8'):
- (vocab_size, nwords, _) = self.struct_unpack(file_handle, '@3i')
+ vocab_size, nwords, _ = self.struct_unpack(file_handle, '@3i')
# Vocab stored by [Dictionary::save](https://github.com/facebookresearch/fastText/blob/master/src/dictionary.cc)
assert len(self.wv.vocab) == nwords, 'mismatch between vocab sizes'
assert len(self.wv.vocab) == vocab_size, 'mismatch between vocab sizes'
- ntokens, = self.struct_unpack(file_handle, '@q')
+ self.struct_unpack(file_handle, '@1q') # number of tokens
+ if self.new_format:
+ pruneidx_size, = self.struct_unpack(file_handle, '@q')
for i in range(nwords):
word_bytes = b''
char_byte = file_handle.read(1)
@@ -285,12 +297,17 @@ def load_dict(self, file_handle, encoding='utf8'):
word_bytes += char_byte
char_byte = file_handle.read(1)
word = word_bytes.decode(encoding)
- count, _ = self.struct_unpack(file_handle, '@ib')
- _ = self.struct_unpack(file_handle, '@i')
+ count, _ = self.struct_unpack(file_handle, '@qb')
assert self.wv.vocab[word].index == i, 'mismatch between gensim word index and fastText word index'
self.wv.vocab[word].count = count
+ if self.new_format:
+ for j in range(pruneidx_size):
+ self.struct_unpack(file_handle, '@2i')
+
def load_vectors(self, file_handle):
+ if self.new_format:
+ self.struct_unpack(file_handle, '@?') # bool quant_input in fasttext.cc
num_vectors, dim = self.struct_unpack(file_handle, '@2q')
# Vectors stored by [Matrix::save](https://github.com/facebookresearch/fastText/blob/master/src/matrix.cc)
assert self.vector_size == dim, 'mismatch between model sizes'
diff --git a/gensim/models/wrappers/ldamallet.py b/gensim/models/wrappers/ldamallet.py
index fb9ae1e31d..640cf11dd8 100644
--- a/gensim/models/wrappers/ldamallet.py
+++ b/gensim/models/wrappers/ldamallet.py
@@ -43,7 +43,7 @@
from smart_open import smart_open
from gensim import utils, matutils
-from gensim.utils import check_output
+from gensim.utils import check_output, revdict
from gensim.models.ldamodel import LdaModel
from gensim.models import basemodel
@@ -190,7 +190,7 @@ def load_word_topics(self):
if hasattr(self.id2word, 'token2id'):
word2id = self.id2word.token2id
else:
- word2id = dict((v, k) for k, v in iteritems(self.id2word))
+ word2id = revdict(self.id2word)
with utils.smart_open(self.fstate()) as fin:
_ = next(fin) # header
@@ -240,14 +240,19 @@ def show_topics(self, num_topics=10, num_words=10, log=False, formatted=True):
logger.info("topic #%i (%.3f): %s", i, self.alpha[i], topic)
return shown
- def show_topic(self, topicid, num_words=10):
+ def show_topic(self, topicid, topn=10, num_words=None):
+ if num_words is not None: # deprecated num_words is used
+ logger.warning("The parameter num_words for show_topic() would be deprecated in the updated version.")
+ logger.warning("Please use topn instead.")
+ topn = num_words
+
if self.word_topics is None:
logger.warning(
"Run train or load_word_topics before showing topics."
)
topic = self.word_topics[topicid]
topic = topic / topic.sum() # normalize to probability dist
- bestn = matutils.argsort(topic, num_words, reverse=True)
+ bestn = matutils.argsort(topic, topn, reverse=True)
beststr = [(self.id2word[id], topic[id]) for id in bestn]
return beststr
diff --git a/gensim/models/wrappers/wordrank.py b/gensim/models/wrappers/wordrank.py
index e691a6b757..efeb020199 100644
--- a/gensim/models/wrappers/wordrank.py
+++ b/gensim/models/wrappers/wordrank.py
@@ -8,7 +8,7 @@
`Word2Vec` for that.
Example:
->>> model = gensim.models.wrappers.Wordrank('/Users/dummy/wordrank', corpus_file='text8', out_path='wr_model')
+>>> model = gensim.models.wrappers.Wordrank('/Users/dummy/wordrank', corpus_file='text8', out_name='wr_model')
>>> print model[word] # prints vector for given words
.. [1] https://bitbucket.org/shihaoji/wordrank/
@@ -45,14 +45,14 @@ class Wordrank(KeyedVectors):
"""
@classmethod
- def train(cls, wr_path, corpus_file, out_path, size=100, window=15, symmetric=1, min_count=5, max_vocab_size=0,
+ def train(cls, wr_path, corpus_file, out_name, size=100, window=15, symmetric=1, min_count=5, max_vocab_size=0,
sgd_num=100, lrate=0.001, period=10, iter=90, epsilon=0.75, dump_period=10, reg=0, alpha=100,
beta=99, loss='hinge', memory=4.0, cleanup_files=True, sorted_vocab=1, ensemble=0):
"""
`wr_path` is the path to the Wordrank directory.
`corpus_file` is the filename of the text file to be used for training the Wordrank model.
Expects file to contain space-separated tokens in a single line
- `out_path` is the path to directory which will be created to save embeddings and training data.
+ `out_name` is name of the directory which will be created (in wordrank folder) to save embeddings and training data.
`size` is the dimensionality of the feature vectors.
`window` is the number of context words to the left (and to the right, if symmetric = 1).
`symmetric` if 0, only use left context words, else use left and right both.
@@ -82,7 +82,7 @@ def train(cls, wr_path, corpus_file, out_path, size=100, window=15, symmetric=1,
meta_file = 'meta'
# prepare training data (cooccurrence matrix and vocab)
- model_dir = os.path.join(wr_path, out_path)
+ model_dir = os.path.join(wr_path, out_name)
meta_dir = os.path.join(model_dir, 'meta')
os.makedirs(meta_dir)
logger.info("Dumped data will be stored in '%s'", model_dir)
@@ -95,14 +95,16 @@ def train(cls, wr_path, corpus_file, out_path, size=100, window=15, symmetric=1,
cmd_del_vocab_freq = ['cut', '-d', " ", '-f', '1', temp_vocab_file]
commands = [cmd_vocab_count, cmd_cooccurence_count, cmd_shuffle_cooccurences]
- logger.info("Prepare training data using glove code '%s'", commands)
input_fnames = [corpus_file.split('/')[-1], corpus_file.split('/')[-1], cooccurrence_file]
output_fnames = [temp_vocab_file, cooccurrence_file, cooccurrence_shuf_file]
+ logger.info("Prepare training data (%s) using glove code", ", ".join(input_fnames))
for command, input_fname, output_fname in zip(commands, input_fnames, output_fnames):
with smart_open(input_fname, 'rb') as r:
with smart_open(output_fname, 'wb') as w:
utils.check_output(w, args=command, stdin=r)
+
+ logger.info("Deleting frequencies from vocab file")
with smart_open(vocab_file, 'wb') as w:
utils.check_output(w, args=cmd_del_vocab_freq)
@@ -117,7 +119,12 @@ def train(cls, wr_path, corpus_file, out_path, size=100, window=15, symmetric=1,
if iter % dump_period == 0:
iter += 1
else:
- logger.warning('Resultant embedding would be from %d iteration', iter - iter % dump_period)
+ logger.warning(
+ 'Resultant embedding will be from %d iterations rather than the input %d iterations, '
+ 'as wordrank dumps the embedding only at dump_period intervals. '
+ 'Input an appropriate combination of parameters (iter, dump_period) such that '
+ '"iter mod dump_period" is zero.', iter - (iter % dump_period), iter
+ )
wr_args = {
'path': 'meta',
@@ -142,11 +149,11 @@ def train(cls, wr_path, corpus_file, out_path, size=100, window=15, symmetric=1,
for option, value in wr_args.items():
cmd.append('--%s' % option)
cmd.append(str(value))
- logger.info("Running wordrank binary '%s'", cmd)
+ logger.info("Running wordrank binary")
output = utils.check_output(args=cmd)
# use embeddings from max. iteration's dump
- max_iter_dump = iter - iter % dump_period
+ max_iter_dump = iter - (iter % dump_period)
copyfile('model_word_%d.txt' % max_iter_dump, 'wordrank.words')
copyfile('model_context_%d.txt' % max_iter_dump, 'wordrank.contexts')
model = cls.load_wordrank_model('wordrank.words', os.path.join('meta', vocab_file), 'wordrank.contexts', sorted_vocab, ensemble)
diff --git a/gensim/sklearn_integration/sklearn_wrapper_gensim_ldamodel.py b/gensim/sklearn_integration/sklearn_wrapper_gensim_ldamodel.py
index 003d313e6d..3eae9d0265 100644
--- a/gensim/sklearn_integration/sklearn_wrapper_gensim_ldamodel.py
+++ b/gensim/sklearn_integration/sklearn_wrapper_gensim_ldamodel.py
@@ -16,8 +16,6 @@
from sklearn.base import TransformerMixin, BaseEstimator
-
-
class SklearnWrapperLdaModel(models.LdaModel, TransformerMixin, BaseEstimator):
"""
Base LDA module
@@ -68,7 +66,6 @@ def get_params(self, deep=True):
"gamma_threshold": self.gamma_threshold, "minimum_probability": self.minimum_probability,
"random_state": self.random_state}
-
def set_params(self, **parameters):
"""
Set all parameters.
@@ -81,7 +78,7 @@ def fit(self, X, y=None):
"""
For fitting corpus into the class object.
Calls gensim.model.LdaModel:
- >>>gensim.models.LdaModel(corpus=corpus,num_topics=num_topics,id2word=id2word,passes=passes,update_every=update_every,alpha=alpha,iterations=iterations,eta=eta,random_state=random_state)
+ >>> gensim.models.LdaModel(corpus=corpus, num_topics=num_topics, id2word=id2word, passes=passes, update_every=update_every, alpha=alpha, iterations=iterations, eta=eta, random_state=random_state)
"""
if sparse.issparse(X):
self.corpus = matutils.Sparse2Corpus(X)
@@ -106,16 +103,15 @@ def transform(self, docs, minimum_probability=None):
# The input as array of array
check = lambda x: [x] if isinstance(x[0], tuple) else x
docs = check(docs)
- X = [[] for i in range(0,len(docs))];
- for k,v in enumerate(docs):
+ X = [[] for _ in range(0, len(docs))]
+ for k, v in enumerate(docs):
doc_topics = self.get_document_topics(v, minimum_probability=minimum_probability)
probs_docs = list(map(lambda x: x[1], doc_topics))
# Everything should be equal in length
if len(probs_docs) != self.num_topics:
probs_docs.extend([1e-12]*(self.num_topics - len(probs_docs)))
X[k] = probs_docs
- probs_docs = []
return np.reshape(np.array(X), (len(docs), self.num_topics))
def get_topic_dist(self, bow, minimum_probability=None, minimum_phi_value=None, per_word_topics=False):
@@ -134,4 +130,4 @@ def partial_fit(self, X):
if sparse.issparse(X):
X = matutils.Sparse2Corpus(X)
- self.update(corpus=X)
\ No newline at end of file
+ self.update(corpus=X)
diff --git a/gensim/test/test_data/bgwiki-latest-pages-articles-shortened.xml.bz2 b/gensim/test/test_data/bgwiki-latest-pages-articles-shortened.xml.bz2
new file mode 100644
index 0000000000..11f3d795c3
Binary files /dev/null and b/gensim/test/test_data/bgwiki-latest-pages-articles-shortened.xml.bz2 differ
diff --git a/gensim/test/test_data/lee_fasttext_new.bin b/gensim/test/test_data/lee_fasttext_new.bin
new file mode 100644
index 0000000000..e9a8f07061
Binary files /dev/null and b/gensim/test/test_data/lee_fasttext_new.bin differ
diff --git a/gensim/test/test_data/lee_fasttext_new.vec b/gensim/test/test_data/lee_fasttext_new.vec
new file mode 100644
index 0000000000..9258025186
--- /dev/null
+++ b/gensim/test/test_data/lee_fasttext_new.vec
@@ -0,0 +1,1764 @@
+1763 10
+the -0.33022 -0.31812 0.10051 -1.0401 0.087806 -0.76704 0.39969 -0.19609 -0.13398 0.30554
+to -0.31987 0.1434 -0.091811 -0.843 -0.88571 -0.61017 0.48257 0.044776 -0.37158 0.36873
+of -0.40452 -0.17903 0.16196 -0.77842 -0.10352 -0.83879 0.57681 -0.027571 -0.21793 0.32038
+in -0.48886 -0.26498 0.044217 -0.9284 0.77722 -1.2153 0.58476 0.069224 -0.022603 -0.33941
+and -0.18378 -0.10715 0.0021384 -0.90693 -0.11685 -0.97686 0.28666 0.118 -0.26708 0.088453
+a -0.33519 -0.29483 -0.035565 -0.8615 0.041433 -1.0078 0.46877 0.016823 -0.047888 0.2902
+is -0.15199 0.30374 0.28534 -0.91076 -0.98414 -0.44665 0.32348 0.23459 -0.52628 0.62149
+for -0.22979 -0.13901 -0.015729 -0.8383 0.16051 -1.1942 0.38129 0.14235 -0.19877 -0.16216
+The -0.203 0.11324 0.099092 -0.86532 0.10771 -0.9537 0.51709 -0.013929 -0.35589 -0.091968
+on -0.20967 -0.19373 0.048881 -0.78054 -0.066953 -1.0141 0.45248 0.12231 -0.13282 0.20771
+he -0.65305 0.05148 -0.15596 -0.99271 -0.90892 -0.69412 0.28682 -0.37462 -0.16706 0.41628
+has -0.1483 -0.32801 0.34937 -1.0249 0.1493 -0.82411 0.37246 0.20367 -0.012695 0.29427
+says -0.20093 -0.28266 0.3047 -0.93661 -0.40017 -0.75988 0.24951 0.12185 -0.23146 0.37119
+was -0.049137 -0.05114 0.22581 -0.99179 0.1735 -0.9129 0.36463 0.097124 -0.19442 0.059312
+have -0.29204 0.045827 0.13382 -1.0406 -0.014144 -0.76245 0.3981 0.045299 -0.25308 0.10293
+that -0.37507 -0.22137 0.23466 -1.0205 -1.0564 -0.34831 0.25409 -0.13757 -0.32117 0.77917
+be -0.74752 -0.034953 0.10615 -1.0751 0.18641 -0.79894 0.42422 -0.5107 -0.15459 -0.0026315
+are 0.1874 -0.36622 0.54514 -1.2728 -0.014307 -0.43794 0.17535 0.059977 -0.25114 0.45772
+will -0.50062 -0.1738 0.050607 -0.8865 -0.19088 -0.78701 0.4637 0.030464 -0.27489 0.33913
+with 0.034973 0.026743 -0.015003 -0.97088 0.21784 -0.9557 0.45493 0.12413 -0.31648 -0.044328
+Mr -0.21121 -0.12407 0.15874 -0.76779 0.10648 -1.1075 0.3834 0.17831 -0.10013 0.20325
+said. -0.46169 -0.25106 0.25167 -0.96973 -0.40777 -0.68279 0.30918 -0.16919 -0.09165 0.50847
+ -0.20078 -0.10337 0.060461 -0.91415 0.1093 -0.90508 0.5017 0.12717 -0.31144 0.15824
+at -0.019143 -0.017879 0.25241 -1.2375 0.81273 -1.2265 0.16146 -0.14443 0.21601 -0.25264
+from -0.09197 -0.012729 -0.0060296 -0.9388 -0.10932 -0.83526 0.49825 0.030485 -0.26831 0.23417
+by -0.33604 -0.2464 0.29963 -0.94804 0.098661 -0.92171 0.32177 -0.0036779 -0.071185 0.048951
+been -0.25494 -0.074667 0.19737 -0.96355 0.0032124 -0.87794 0.34561 -0.10455 -0.16332 -0.050892
+not -0.19936 -0.24242 0.52563 -1.0341 -0.88076 -0.56225 0.13903 -0.015902 -0.11776 0.59915
+as -0.21367 -0.22472 0.12267 -1.0727 0.18493 -0.81337 0.44497 -0.045123 -0.14549 0.2321
+his -0.48603 0.025464 0.12305 -0.91551 -0.22244 -0.86314 0.35775 -0.092713 -0.23615 0.039325
+an -0.48816 -0.23777 0.21673 -0.89387 -0.085778 -0.95051 0.22186 0.051467 -0.054749 0.14937
+it 0.088151 0.06808 0.18461 -0.99546 -1.0999 -0.44344 0.18492 0.12933 -0.57249 0.66995
+were -0.3996 -0.35245 0.11363 -1.0223 -0.30912 -0.69173 0.16747 0.107 -0.19961 0.47033
+had -0.28349 -0.12388 0.22009 -1.0286 0.10782 -0.84988 0.45542 0.0085547 -0.24302 -0.0012204
+after -0.34786 0.045636 -0.023854 -0.82463 0.16736 -1.013 0.56963 0.027898 -0.20699 0.033479
+but -0.20207 -0.2405 0.19077 -0.8589 -0.5069 -0.81368 0.31668 0.39135 -0.29938 0.41487
+they -0.1298 -0.31476 0.38336 -1.1176 -0.15058 -0.56061 0.28551 -0.20703 -0.29367 0.34014
+said -0.39474 -0.36102 0.3086 -0.93279 -0.26217 -0.81679 0.25929 -0.16438 0.027083 0.45336
+this -0.092812 -0.10551 0.35124 -0.96899 -0.40498 -0.68296 0.30468 0.0917 -0.38239 0.35388
+who -0.20105 -0.095276 0.32085 -0.91372 -0.16938 -0.79356 0.38023 0.0038929 -0.37117 0.027668
+Australian -0.28387 -0.070388 0.13034 -0.80433 0.13356 -0.9253 0.67532 0.22014 -0.4056 -0.10834
+we -0.43378 -0.024427 -0.047516 -1.1312 -0.57399 -0.59072 0.2059 -0.40156 -0.31484 0.4411
+Palestinian -0.11929 -0.0022232 -0.3722 -1.021 1.0886 -1.3694 0.71574 -0.21566 0.077381 -0.28098
+their -0.18141 -0.17626 0.16614 -0.92324 -0.40464 -0.69394 0.26769 -0.055062 -0.30801 0.40075
+which -0.19957 -0.11257 0.22569 -0.88656 -0.070622 -0.89995 0.36561 -0.0093768 -0.28029 0.11368
+people 0.040337 -0.13309 0.12374 -0.90381 0.11218 -0.88308 0.48656 0.18849 -0.36103 0.23864
+two -0.10355 0.0031566 0.1133 -1.0841 0.44068 -0.84927 0.50567 -0.067127 -0.30814 -0.0035397
+up -0.25718 -0.099621 0.096818 -0.97874 0.13418 -0.89428 0.35766 -0.09086 -0.20867 0.14024
+there -0.20937 -0.37056 0.45481 -1.1153 -0.99885 -0.32444 0.14491 0.17278 -0.36451 0.87714
+about -0.28995 -0.19368 0.29839 -0.85869 0.072993 -0.97997 0.39695 0.28764 -0.28723 0.023871
+also -0.064228 -0.073123 0.15672 -0.89139 0.23014 -0.98076 0.5608 0.14038 -0.23721 0.099119
+its -0.094794 -0.0089572 -0.061089 -0.92513 0.15797 -0.89387 0.53339 0.099649 -0.25874 0.08474
+South 0.29405 0.3306 0.29858 -0.98334 0.3158 -0.82364 0.64126 0.40047 -0.65075 -0.17298
+out -0.48574 -0.13048 0.083028 -0.80023 -0.57943 -0.77788 0.37554 0.28506 -0.37562 0.32881
+into -0.36937 -0.17606 -0.13391 -0.94078 0.34168 -0.99665 0.64295 -0.11696 -0.1669 -0.0054359
+would -0.28577 -0.23793 0.18882 -0.95394 -0.58257 -0.59028 0.37692 0.12538 -0.2183 0.64066
+US -0.24584 -0.41715 0.18146 -0.92122 0.72152 -1.1306 0.47029 0.012934 0.053693 -0.0086742
+when -0.4617 -0.29113 0.1645 -1.0117 -0.043739 -0.84131 0.34338 -0.1758 -0.0047467 0.28804
+against -0.24538 0.078647 -0.056364 -0.85907 0.29025 -1.0199 0.58317 0.11893 -0.27324 -0.1253
+more 0.31973 -0.055314 0.46665 -0.98215 0.10186 -0.77057 0.36613 0.61176 -0.47173 0.26166
+I -0.3498 -0.20172 -0.020818 -1.0454 -1.1019 -0.45334 0.42848 -0.012756 -0.25491 0.65215
+last -0.18243 -0.29531 0.23062 -0.99804 0.62399 -1.0047 0.51393 0.31862 -0.26751 -0.041609
+first -0.13614 0.093762 0.13683 -0.96666 0.43209 -1.0278 0.54946 0.12433 -0.24603 -0.069502
+New 0.25543 0.27318 0.18001 -1.058 0.24951 -0.82687 0.54174 0.3602 -0.52094 -0.14335
+A -0.31209 0.011631 0.22446 -0.92899 0.067243 -0.93878 0.49181 0.0034494 -0.14883 -0.012253
+He -0.75237 -0.67949 0.049378 -0.93605 -0.078393 -0.99741 0.23349 0.020627 0.027144 0.28656
+Israeli -0.26379 -0.10594 -0.41009 -1.0794 1.2779 -1.3504 0.68809 -0.53596 0.089812 -0.21142
+Australia -0.21157 0.022314 0.07604 -0.78834 0.15132 -0.90568 0.72975 0.28225 -0.51305 -0.1799
+one 0.1261 -0.099178 0.15185 -1.06 0.17422 -0.92321 0.3207 0.11182 -0.13747 0.0563
+if -0.48178 -0.23325 0.025306 -0.92117 -0.29508 -0.82477 0.49644 -0.049607 -0.15429 0.30857
+United -0.42223 -0.26833 0.26968 -0.98168 0.49726 -1.0389 0.36316 -0.083795 -0.01649 -0.10937
+over -0.26667 -0.11149 0.37049 -0.90953 0.12751 -0.97556 0.35257 0.17014 -0.22273 0.17052
+Government -0.29815 -0.43747 0.24994 -0.82199 0.34598 -1.0416 0.54893 0.34044 -0.087492 0.035292
+or -0.27575 -0.096525 0.20245 -0.89668 -0.91115 -0.61982 0.16335 0.0376 -0.42491 0.53103
+than -0.33 -0.21229 0.19697 -1.0063 0.00041447 -0.6629 0.4611 0.093545 -0.33235 0.3079
+all -0.26628 -0.14896 0.24025 -0.89538 -0.16143 -0.8923 0.30807 0.12912 -0.084696 0.32484
+no -0.40013 -0.16031 0.15533 -0.84239 -0.20989 -0.87639 0.52505 0.050859 -0.22787 0.15883
+could -0.49695 -0.36794 0.25997 -0.88439 -0.49008 -0.65626 0.40343 0.15186 -0.16664 0.51918
+before -0.3407 -0.058254 0.0596 -0.95199 0.08027 -0.8974 0.41102 0.020095 -0.24425 0.012549
+three -0.26678 -0.015288 -0.047869 -0.95902 0.50998 -1.122 0.43916 -0.010073 -0.22936 -0.22466
+say -0.032002 -0.26703 0.19972 -0.94938 0.079993 -1.0376 0.34252 -0.020494 -0.10124 0.080242
+told -0.42438 -0.20704 -0.0056567 -0.90714 -0.24315 -0.75983 0.41709 -0.090443 -0.13259 0.37537
+new -0.54038 -0.21066 0.05754 -0.91031 0.2118 -0.8609 0.47213 -0.12175 -0.28987 0.020162
+some -0.20853 -0.07386 0.15236 -0.97983 -0.019563 -0.69457 0.50208 0.087262 -0.38281 0.18132
+any -0.22956 -0.25756 0.27368 -0.85082 -0.871 -0.73584 0.17203 0.35347 -0.39229 0.48237
+"We -0.24137 -0.11195 0.16273 -1.0519 0.15123 -0.91253 0.33623 -0.34671 -0.076989 0.071385
+bin -1.1369 -0.19739 0.43403 -1.2298 1.1812 -1.1997 0.2993 -0.53082 0.32635 -0.62403
+attacks -0.3315 -0.046235 0.0059848 -1.189 1.2997 -1.355 0.46851 -0.31042 0.00050552 -0.73975
+very -0.24525 -0.18 0.24736 -1.0176 -1.009 -0.44592 0.27645 -0.0046416 -0.3104 0.80213
+still -0.48843 -0.23529 0.18375 -0.91555 0.030518 -0.81222 0.49447 -0.10733 -0.12523 0.2623
+now -0.22925 -0.28336 0.41243 -0.96635 -0.11287 -0.77443 0.3012 0.037946 -0.20964 0.2162
+just -0.26101 -0.088722 0.26512 -0.96292 -0.06783 -0.72943 0.47772 0.24115 -0.38367 0.21506
+security -0.30075 -0.16487 -0.15123 -0.92988 0.50471 -1.1198 0.58145 -0.1661 -0.028197 0.014528
+police -0.1127 -0.1825 0.039113 -0.96865 -0.027693 -0.90584 0.39815 0.076821 -0.062821 0.37022
+our -0.14584 -0.04477 -0.14099 -0.80127 -0.90416 -0.79821 0.27668 -0.15629 -0.43345 0.38823
+killed -0.18154 -0.19708 0.10268 -0.97047 0.42065 -1.1094 0.34835 -0.076757 0.01084 -0.098382
+Arafat -0.52927 -0.03356 -0.17259 -0.96235 0.71803 -1.2615 0.49958 -0.55362 0.26507 -0.15125
+"I -0.39736 0.13625 -0.023159 -0.98296 -1.4491 -0.42356 0.37467 -0.19728 -0.26311 0.77596
+them -0.16723 -0.29989 0.2555 -1.0339 -0.59839 -0.57582 0.26505 -0.13065 -0.33663 0.47831
+being -0.29334 -0.13715 0.11571 -0.90582 -0.42662 -0.83744 0.29407 -0.0085566 -0.33293 -0.079894
+Minister -0.77857 -0.041647 0.055757 -0.78251 0.35742 -1.1563 0.62952 -0.24813 0.18156 0.0074797
+forces -0.31611 -0.26023 0.13476 -0.95077 0.68984 -1.1688 0.46259 0.055986 -0.04458 -0.18726
+States -0.21613 -0.22527 0.2876 -0.8926 0.5769 -1.0471 0.57627 0.048655 -0.192 -0.17265
+But -0.04944 -0.13106 0.14496 -0.89226 -0.10263 -0.82114 0.51183 0.072984 -0.2451 0.28737
+fire 0.4593 -0.2163 0.63263 -1.1762 0.35455 -0.69784 0.27687 0.50648 -0.33031 0.23918
+other -0.29501 -0.42237 0.33144 -1.0263 0.19114 -0.78719 0.38055 -0.11142 -0.1256 0.24444
+what -0.28618 -0.23335 0.4351 -1.0115 -0.54005 -0.5525 0.27984 -0.11731 -0.2719 0.50848
+man -0.2559 0.24893 0.065557 -0.8362 0.024539 -1.0033 0.51622 0.11164 -0.35693 -0.20309
+around -0.31284 -0.16331 0.21767 -1.0888 0.29451 -0.89781 0.24964 -0.052161 -0.10537 -0.017295
+where -0.3501 -0.35727 0.47376 -1.063 -0.22067 -0.68804 0.12911 0.058005 -0.1445 0.34572
+can -0.055922 0.030927 0.058761 -1.0212 0.14008 -0.88066 0.367 -0.07038 -0.22231 0.073602
+think -0.1731 -0.1926 0.32029 -1.0495 -0.84012 -0.48107 0.28456 0.059164 -0.4043 0.52989
+per -0.35812 0.11215 -0.18332 -0.80901 0.1944 -1.0827 0.57297 -0.26525 -0.10867 0.11525
+day 0.012601 0.1896 0.095068 -0.86566 0.34808 -1.0604 0.52837 0.19571 -0.52749 -0.26127
+next -0.31571 -0.060529 -0.0075701 -0.86395 0.11404 -0.95719 0.53605 -0.029668 -0.22698 0.14434
+Al -1.3353 -0.092465 0.39203 -1.0874 0.63041 -1.0789 0.35236 -0.71515 0.2539 -0.46528
+company -0.19165 -0.28963 0.25913 -0.84085 -0.2211 -0.83681 0.36584 0.31188 -0.30592 0.29555
+It -0.284 -0.11323 0.36562 -0.93056 -0.60506 -0.65385 0.23822 0.0019475 -0.25188 0.54004
+four 0.037041 0.084169 -0.089042 -0.91625 0.064523 -1.0705 0.28069 -0.05684 -0.33851 -0.033145
+Qaeda -0.91306 -0.15476 0.19559 -1.0222 0.77476 -1.1154 0.47116 -0.39111 0.024283 -0.52145
+"The -0.38686 -0.023665 0.048808 -0.95457 0.36467 -1.0604 0.50549 -0.097076 -0.02866 0.014342
+take -0.4219 -0.03483 0.0012477 -0.72079 -0.21566 -1.0017 0.52492 0.13763 -0.22839 0.072995
+you -0.087751 -0.16713 0.16542 -1.0399 -1.0658 -0.64585 0.15991 0.045301 -0.22861 0.5694
+officials -0.32051 -0.21077 0.037937 -0.95437 0.60363 -1.1237 0.57537 -0.21206 0.067605 -0.1028
+suicide -0.28618 -0.10063 -0.22642 -0.9794 0.79137 -1.1419 0.51294 -0.19923 -0.065382 -0.14762
+so -0.13635 -0.30147 -0.081736 -1.0996 0.067797 -0.51032 0.70453 -0.16876 -0.39209 0.58863
+Afghan -1.1706 -0.3155 0.067356 -1.0196 1.0935 -1.2534 0.6789 -0.43071 0.18712 -0.46391
+under -0.48811 -0.067802 0.0057148 -0.87174 -0.12285 -1.0058 0.40118 -0.17195 -0.19471 0.070437
+President -0.32441 -0.2851 0.16182 -0.85866 0.1776 -1.0136 0.41405 0.19381 -0.16785 -0.043423
+Federal -0.25055 -0.4163 0.36037 -0.87956 0.020906 -0.89772 0.38427 0.23415 -0.11237 0.12181
+In -0.56293 0.22817 0.088029 -0.71931 0.010781 -1.0672 0.64169 -0.25677 -0.24875 -0.038812
+time -0.52833 -0.056916 0.043438 -0.85557 -0.29898 -0.82812 0.40698 -0.15416 -0.31113 0.056874
+Taliban -0.74078 -0.31884 0.16068 -0.96942 0.53735 -1.053 0.48127 -0.18015 0.10536 -0.23858
+made -0.37549 0.23968 0.0083349 -0.85004 -0.51914 -0.76553 0.45113 0.21039 -0.44265 0.10979
+number -0.12057 -0.15317 0.18722 -0.90501 0.45588 -1.1197 0.42543 0.019829 -0.21294 -0.10256
+days -0.072485 0.027079 0.1092 -0.93477 0.11575 -0.9536 0.45806 0.2372 -0.4375 0.059927
+Laden -0.90037 -0.18422 0.36881 -1.1109 0.78784 -1.0523 0.24214 -0.38333 -0.059544 -0.4559
+down -0.072217 -0.12778 0.054756 -1.1118 0.34834 -0.82644 0.47349 -0.12034 -0.22186 0.052777
+through -0.25441 -0.033393 0.18056 -1.0011 0.15407 -0.97571 0.37446 0.020637 -0.20062 -0.072035
+those -0.21808 -0.19675 0.40513 -0.99374 -0.42841 -0.66267 0.22048 0.09974 -0.20575 0.43765
+meeting -0.45894 -0.0097337 -0.18976 -0.83278 0.23521 -1.156 0.46863 -0.13151 -0.10006 -0.19093
+including -0.22074 -0.15843 0.069876 -0.88368 0.28899 -1.0397 0.517 0.10314 -0.33756 -0.23928
+Hamas -0.18186 -0.22878 0.11349 -0.90034 0.53257 -1.0775 0.49148 0.10502 -0.13666 -0.042208
+Gaza 0.097114 0.057388 -0.021218 -0.92321 0.58338 -1.0238 0.54903 -0.006679 -0.42196 -0.066572
+workers -0.53701 -0.18398 0.14693 -0.95833 -0.065438 -0.76921 0.48484 0.010138 -0.10631 0.090272
+Sydney 0.42337 0.17773 0.41603 -1.1125 0.11717 -0.67133 0.60738 0.33638 -0.49206 0.19372
+she -0.74477 -0.083015 0.31348 -0.95676 0.12699 -0.94792 0.40537 -0.48402 0.0428 0.17063
+military -0.55691 -0.17432 0.019327 -0.91594 0.56586 -1.1349 0.52516 -0.18514 -0.013428 -0.18002
+should -0.44619 -0.28481 0.20766 -0.89841 -0.39622 -0.70786 0.38093 0.17609 -0.25709 0.38403
+called -0.18626 -0.18218 0.13878 -0.91435 0.2147 -1.1208 0.34494 0.16412 0.0060132 0.0092137
+since -0.36396 -0.0063352 -0.12582 -0.94967 0.46389 -0.94729 0.57313 -0.069767 -0.20223 -0.085856
+cent -0.1494 -0.15155 0.26716 -0.73245 -0.010617 -1.0085 0.59429 0.60725 -0.18189 0.039498
+second -0.22516 0.037361 -0.049318 -0.94383 -0.014107 -0.87578 0.54721 0.015513 -0.31377 0.16859
+Test -0.033719 0.34239 -0.17324 -0.83923 0.14746 -0.98619 0.72496 0.045624 -0.65275 -0.21728
+Wales 0.053656 0.28324 0.055285 -0.96671 0.22785 -0.81622 0.68125 0.21636 -0.27924 0.10288
+Islamic -0.22614 -0.19205 -0.016994 -0.91469 0.66012 -1.0559 0.57151 0.075808 -0.17611 -0.018144
+today -0.11086 -0.080891 0.070837 -0.83169 0.40189 -1.0907 0.53966 0.13669 -0.29179 -0.083567
+get -0.26654 0.11655 -0.0089127 -0.88745 -0.87471 -0.70857 0.35412 -0.061572 -0.26976 0.58908
+World -0.19331 -0.044139 0.28748 -0.94014 -0.073753 -0.8746 0.34228 0.15011 -0.2833 0.16732
+between -0.49888 -0.17084 0.071969 -0.98188 0.49201 -1.0247 0.51424 -0.14237 -0.085301 -0.17244
+September -0.18975 -0.064982 0.19584 -0.94499 0.34528 -0.98927 0.44114 -0.020988 -0.18947 -0.095848
+back -0.27283 -0.059433 0.13938 -0.94367 -0.15244 -0.87978 0.41035 0.014452 -0.19059 0.13653
+because -0.43831 -0.10384 0.093664 -1.0154 -0.2803 -0.67246 0.3828 -0.14649 -0.17899 0.40723
+members -0.30101 -0.18809 0.17142 -0.9703 0.73091 -1.1413 0.5201 -0.084871 -0.074367 -0.25823
+while -0.20608 -0.041047 0.18854 -0.90223 0.20482 -0.97908 0.46448 0.00026892 -0.23846 0.022349
+- -0.3817 -0.435 0.20958 -1.1556 -0.43852 -0.68643 0.37675 -0.36924 0.10648 0.5545
+Bank -0.19756 -0.18417 0.055352 -1.025 0.69146 -1.0088 0.56631 -0.15637 -0.20305 -0.22032
+staff -0.25389 -0.26152 0.32648 -0.85727 -0.22542 -0.81737 0.31717 0.039845 -0.20685 0.30589
+report -0.20298 -0.29043 0.13439 -0.86815 -0.22312 -0.77922 0.42075 0.2256 -0.22305 0.39632
+near 0.063087 -0.13746 0.04948 -0.97593 0.57767 -1.0817 0.54532 0.017656 -0.20763 -0.052598
+going -0.34285 -0.30352 0.10319 -0.94297 -0.59234 -0.6944 0.36893 -0.084732 -0.22855 0.39174
+further -0.21702 -0.11672 0.28065 -1.0878 -0.19926 -0.5478 0.38098 -0.15962 -0.2667 0.40168
+world -0.34496 -0.2073 0.31545 -0.97096 0.096099 -0.83532 0.37357 0.10556 -0.23433 0.10518
+him -0.68234 -0.058697 -0.16187 -0.83721 0.057589 -1.017 0.49891 -0.32563 0.050758 0.069657
+local -0.38225 -0.29285 0.21606 -0.9377 0.07638 -0.939 0.34291 0.045473 -0.044331 0.19069
+former -0.42283 -0.13981 0.18794 -0.81396 -0.11717 -0.9734 0.34292 0.087346 -0.16579 0.048752
+Australia's -0.25338 -0.06371 0.16415 -0.81694 0.052229 -0.90036 0.64271 0.22495 -0.4157 -0.061322
+end -0.15889 0.24558 -0.055311 -0.83453 -0.043634 -0.96825 0.43326 0.15381 -0.34587 -0.053832
+attack -0.17126 -0.013473 0.045089 -1.1729 1.0639 -1.2619 0.4447 -0.2385 -0.085905 -0.58779
+Israel -0.35389 -0.13604 -0.40936 -1.0228 1.3653 -1.4265 0.73706 -0.53007 0.19975 -0.26019
+West 0.23361 0.22398 -0.15368 -0.9687 0.64233 -0.98961 0.6875 -0.029865 -0.54572 -0.15117
+hours -0.043814 -0.041619 0.30682 -0.97939 -0.025469 -0.8232 0.25673 0.19288 -0.27147 0.15032
+government -0.30304 -0.35389 0.19802 -0.82989 0.28777 -1.0482 0.59923 0.34018 -0.12011 0.092391
+international -0.48173 -0.17748 0.049294 -0.75205 0.53984 -1.1959 0.64301 0.21821 -0.16932 -0.14384
+Afghanistan -1.1966 -0.44437 0.15186 -1.0763 1.0313 -1.1843 0.59815 -0.42975 0.27204 -0.37176
+leader -0.77413 -0.10514 -0.06776 -0.8829 0.21361 -1.121 0.39473 -0.36839 0.12408 -0.001276
+like -0.50088 -0.28038 0.22604 -0.99886 0.02199 -0.83443 0.38421 0.014473 -0.057847 0.098137
+only -0.23106 0.07436 -0.044206 -0.79498 -0.98372 -0.62288 0.41626 0.080259 -0.44068 0.59421
+do -0.36461 -0.098158 -0.041925 -1.0584 -0.58619 -0.52612 0.47479 -0.20727 -0.26433 0.47595
+off -0.35971 -0.31598 0.18306 -1.0005 0.46072 -1.01 0.54209 -0.072324 0.022646 0.030194
+make -0.30113 0.049585 -0.073341 -0.891 -0.5538 -0.77719 0.42122 -0.023962 -0.29509 0.41399
+claims -0.36885 -0.26892 0.27832 -0.92341 0.031652 -0.83457 0.3885 -0.014591 -0.22142 0.17689
+another -0.20017 -0.37784 0.34455 -0.99432 -0.22374 -0.69401 0.30434 -0.0451 -0.16091 0.41697
+expected -0.26527 -0.079034 0.15499 -0.94017 -0.033431 -0.92758 0.37691 0.090954 -0.22083 0.12863
+it's -0.14742 -0.21869 0.0124 -1.1749 -0.38998 -0.56266 0.50493 -0.11677 -0.22547 0.50131
+many 0.057404 0.085441 0.17244 -0.87545 -0.25314 -0.8043 0.35708 0.27594 -0.41464 0.18643
+spokesman -0.10824 -0.084093 0.24338 -1.0036 0.37949 -1.0592 0.44724 0.10272 -0.095293 -0.069468
+given -0.43047 0.071514 0.0050702 -0.82036 -0.12134 -1.0035 0.50737 0.094394 -0.21531 -0.032985
+five -0.16969 0.029744 0.15184 -0.97348 0.25694 -0.95445 0.43115 0.071704 -0.24367 -0.043415
+go 0.078374 -0.19796 0.088563 -1.1733 -0.30378 -0.71145 0.24977 0.049791 -0.22449 0.48565
+good -0.15264 -0.054147 0.08512 -0.93805 -0.23591 -0.88373 0.4757 -0.030068 -0.34181 0.23762
+looking -0.34264 -0.11517 0.12333 -0.91539 -0.50244 -0.80002 0.422 0.03678 -0.32918 0.2594
+Osama -0.72648 -0.056445 0.38 -0.94188 0.057955 -0.91816 0.26583 -0.19597 -0.10273 -0.14943
+left -0.35945 -0.33987 0.4253 -1.0454 0.20377 -0.83701 0.36679 0.091791 -0.042632 0.10899
+group -0.3377 -0.11551 0.16781 -0.90786 0.28475 -0.99937 0.36675 0.016636 -0.304 -0.17077
+saying -0.13771 -0.21288 0.22513 -0.85351 -0.53943 -0.84151 0.33804 0.1424 -0.37602 0.18033
+Tora -0.9356 -0.25858 0.31026 -1.0856 0.4355 -0.96089 0.36838 -0.35437 0.1891 -0.12649
+Qantas -0.65939 -0.30487 0.28702 -0.902 -0.1617 -0.80426 0.37087 0.018263 -0.0082757 0.099808
+work -0.47285 -0.073762 0.35523 -1.0526 -0.31577 -0.55673 0.26701 0.084506 -0.16215 0.26032
+Prime -0.46482 0.1016 0.027213 -0.77844 -0.26784 -0.9513 0.5403 -0.21129 -0.1866 0.029596
+put -0.1248 -0.16775 0.41894 -1.0333 -0.583 -0.44898 0.27078 0.20875 -0.49565 0.57485
+know -0.59676 -0.37725 0.34874 -1.1539 -0.45241 -0.53131 0.22397 -0.29578 0.012399 0.39678
+during -0.21109 -0.10078 0.039143 -0.76322 0.25825 -1.179 0.54879 0.27295 -0.27751 -0.26122
+most 0.13793 0.04456 0.22526 -0.98488 -0.36391 -0.65328 0.40291 0.42988 -0.60882 0.25216
+air 0.26741 -0.074068 0.29465 -1.0677 0.15293 -0.85312 0.1924 -0.0063716 -0.39769 -0.0064085
+action -0.75958 -0.37306 0.11814 -0.88526 -0.14207 -0.83496 0.34372 0.20376 -0.090856 0.1885
+Indian -0.52359 0.00096827 0.19522 -0.90832 0.41361 -1.1591 0.5906 -0.22471 0.025619 -0.19738
+these -0.40536 -0.37146 0.31893 -1.0205 -0.58433 -0.45952 0.29835 -0.0078722 -0.31221 0.56102
+way -0.043924 -0.17663 0.024529 -1.0232 -0.037238 -0.90182 0.28442 0.025983 -0.18049 0.087988
+Yasser -0.55153 0.097803 -0.13613 -0.90587 0.54645 -1.1794 0.63655 -0.20578 0.077585 -0.14809
+found -0.29833 -0.023452 0.13144 -1.0503 0.34796 -0.98793 0.32726 -0.12611 -0.15783 -0.18491
+support -0.3308 -0.052304 0.051654 -0.93353 0.072362 -0.84319 0.49866 0.040403 -0.26077 0.12345
+died -0.29559 -0.13104 -0.025482 -0.9058 0.30137 -1.1359 0.33124 0.014805 -0.038687 -0.14306
+whether -0.4517 -0.20994 0.25801 -0.89994 -0.60469 -0.66229 0.33923 -0.060431 -0.19682 0.57263
+years -0.11243 -0.15501 0.057288 -1.009 0.33752 -0.83704 0.64005 0.12612 -0.21893 0.1085
+national -0.32425 -0.18239 0.0956 -0.66453 0.46247 -1.1941 0.63184 0.46011 -0.27304 -0.11904
+metres -0.0024676 -0.017324 0.34124 -0.92656 0.13016 -1.0326 0.44193 0.43597 -0.22321 0.11957
+Afghanistan. -1.0575 -0.4603 0.2064 -1.0681 0.91871 -1.0909 0.54211 -0.34105 0.23903 -0.29086
+come -0.28172 -0.26435 0.032234 -0.82914 -0.27653 -0.82745 0.48779 0.29357 -0.29068 0.23432
+set -0.54167 -0.095208 -0.075166 -0.82243 -0.09791 -0.8961 0.59911 -0.017329 -0.092535 0.13694
+six -0.49273 -0.098289 -0.13952 -0.86139 0.18714 -1.1441 0.54307 -0.063048 -0.10221 -0.088211
+year. -0.02107 -0.076537 0.025904 -0.88686 0.28219 -0.8883 0.62195 0.1519 -0.30504 0.083323
+interim -0.55091 -0.063612 -0.098585 -0.75635 0.36788 -1.1661 0.75483 0.019867 -0.26509 -0.16353
+team -0.23807 0.076507 0.10539 -0.84693 0.44473 -1.1163 0.69508 0.094907 -0.054019 -0.16245
+power -0.36772 -0.052696 0.19511 -0.89868 -0.34209 -0.84353 0.38587 -0.047361 -0.21216 0.3796
+Foreign -0.42638 -0.33748 0.13848 -0.87259 0.047046 -0.92526 0.46018 -0.0052589 -0.023769 0.1715
+terrorist -0.48168 -0.18911 0.20601 -0.90702 0.50172 -1.0465 0.52332 0.0094234 -0.23281 -0.23079
+how -0.01793 -0.1927 0.35889 -1.0481 -0.16997 -0.64969 0.38255 0.078809 -0.21499 0.39838
+arrested -0.24013 -0.030285 0.077353 -0.99602 0.77287 -1.1554 0.5164 -0.05334 0.0040569 -0.19586
+11 -0.22958 0.060746 0.23076 -1.0176 0.38595 -0.94242 0.46941 0.048733 -0.25515 -0.10744
+trying -0.35715 -0.17342 0.086758 -0.82987 0.054978 -1.0587 0.47409 0.0037893 -0.21448 -0.1188
+don't 0.054365 0.057473 0.0429 -1.0606 -0.8671 -0.48444 0.41769 0.11095 -0.55092 0.47744
+start -0.12029 -0.16384 0.22578 -1.0056 -0.23547 -0.64488 0.49944 0.068158 -0.33384 0.32661
+Africa -0.24422 0.33216 -0.17724 -0.96404 0.15966 -0.93151 0.67632 -0.058485 -0.47762 -0.23633
+official -0.27807 -0.27393 0.050194 -0.95618 0.68695 -1.1649 0.52643 -0.1773 0.10471 -0.11005
+part -0.34295 -0.38945 -0.008875 -0.8628 0.13704 -1.0347 0.53515 0.22521 -0.10832 0.0017831
+Bora -0.761 -0.16929 0.2101 -1.1157 0.74454 -1.1218 0.47463 -0.29481 0.17755 -0.29277
+force -0.43796 -0.28804 -0.086755 -0.91122 0.42819 -1.1881 0.41628 0.079343 -0.11514 -0.073932
+us -0.27594 -0.071172 -0.20932 -0.96433 0.67585 -1.2376 0.37581 -0.30314 -0.14262 -0.33753
+John -0.23649 0.10085 -0.032482 -0.76448 -0.44261 -0.85056 0.49864 0.14196 -0.2002 0.32642
+early -0.27264 0.025757 0.10665 -0.91027 0.36301 -1.0546 0.51371 -0.058836 -0.30955 -0.20738
+groups -0.36844 -0.17499 0.19636 -0.95065 0.53639 -1.0072 0.3913 -0.11098 -0.19514 -0.15957
+third -0.010789 0.10851 0.11942 -0.86795 -0.41811 -0.78002 0.41311 0.31304 -0.47069 0.19831
+week -0.3272 -0.23908 0.076761 -0.98408 0.073715 -0.74294 0.37483 -0.016686 -0.19666 0.20611
+Meanwhile, -0.34793 -0.12484 0.015576 -0.82154 0.21952 -1.0105 0.49337 -0.031845 -0.23596 0.0039735
+several -0.29225 -0.38715 0.37939 -0.87924 -0.10337 -0.82511 0.32297 0.17878 -0.12513 0.21207
+area 0.051532 -0.079689 0.48495 -1.1795 0.40117 -0.7519 0.39812 -0.12376 -0.091419 0.2226
+believe -0.79137 -0.19282 0.30649 -1.0801 -0.20069 -0.65642 0.28836 -0.38646 -0.090604 0.1288
+war -0.072243 -0.19775 0.16679 -0.94965 0.23958 -0.99727 0.35177 0.22279 -0.092642 0.1017
+authorities -0.30937 -0.19796 0.20576 -1.0058 0.10396 -0.80331 0.4773 0.098594 -0.19382 0.21115
+yesterday -0.28142 0.068957 0.0015726 -0.86612 0.43597 -1.0735 0.69558 -0.018298 -0.23959 -0.10578
+50 -0.0055131 0.080877 0.11533 -1.0507 -0.39545 -0.64091 0.43053 0.24701 -0.28416 0.40795
+100 -0.16565 0.06535 -0.018297 -0.87717 -0.23263 -0.92046 0.48097 0.33465 -0.40374 0.057434
+troops -0.36922 -0.29548 0.19338 -0.87142 0.10742 -0.94169 0.4159 0.10247 -0.18104 0.051702
+few -0.16567 0.076985 0.16912 -0.88033 -0.37245 -0.86022 0.42284 0.21798 -0.42656 0.18659
+does -0.13963 0.10813 0.28804 -0.99911 -0.79363 -0.50716 0.35456 0.10862 -0.4552 0.42702
+Defence -0.33275 -0.25229 0.073641 -0.92387 0.25551 -0.8613 0.59302 0.015404 -0.076559 0.048157
+Arafat's -0.4413 -0.043291 -0.081289 -0.97458 0.29104 -0.9891 0.52851 -0.36433 0.084934 0.11324
+Dr 0.16032 0.22758 0.71706 -0.94864 -0.35815 -0.64197 0.41628 0.65847 -0.48944 0.11823
+Minister, -0.70766 -0.14913 0.079326 -0.81266 0.38258 -1.1154 0.59941 -0.16916 0.14987 -0.007826
+peace -0.40619 -0.10977 -0.29056 -0.8501 0.38124 -1.085 0.66782 0.041173 -0.11891 0.071098
+best -0.27665 -0.072405 -0.094569 -1.0096 0.1308 -0.86547 0.53475 -0.20991 -0.53823 -0.085448
+following -0.050905 0.051095 -0.0067287 -0.87015 -0.10041 -0.9497 0.49073 0.17642 -0.31966 0.035212
+areas -0.10445 -0.21987 0.52734 -1.1521 0.12014 -0.59408 0.40009 0.0093 -0.16673 0.35121
+leaders -0.67576 -0.26126 -0.010294 -1.0261 0.66205 -1.0784 0.4177 -0.33904 0.10604 -0.17319
+weather -0.3792 -0.21008 0.34221 -1.0772 -0.26225 -0.60824 0.28774 -0.069025 -0.24615 0.40984
+match -0.117 0.17501 -0.069883 -0.81983 -0.32765 -0.87824 0.53774 0.20668 -0.56704 -0.036571
+militants -0.40722 -0.23625 -0.014737 -0.91665 0.84706 -1.1936 0.56538 -0.074623 -0.067505 -0.24198
+eight -0.062936 -0.11716 0.24904 -0.90097 0.28141 -0.97546 0.4945 0.33458 -0.26289 -0.054645
+want -0.40733 -0.26792 0.10897 -0.84382 -0.76376 -0.80675 0.35178 0.14463 -0.20792 0.12911
+need -0.27294 -0.12376 0.088059 -1.0434 0.055144 -0.95699 0.26591 -0.15744 -0.16685 0.1328
+confirmed -0.45606 -0.34261 0.15985 -1.0169 0.38209 -0.9086 0.42904 -0.030701 0.025065 0.054046
+Christmas -0.42346 -0.20059 0.11211 -0.95636 0.26371 -0.97323 0.45247 -0.045015 -0.040042 0.050131
+close -0.38157 -0.17253 0.1743 -0.9326 0.27512 -0.97731 0.53688 -0.076656 -0.15657 0.030634
+state -0.017015 -0.26476 0.50601 -1.0991 0.40481 -0.83473 0.3761 0.065668 -0.24655 0.11798
+came -0.037252 -0.09478 0.041212 -0.91742 0.24824 -1.0215 0.38391 0.092885 -0.28134 -0.13021
+Pakistan -1.0344 -0.55098 0.11778 -1.0003 0.79482 -1.3063 0.43674 -0.25713 0.10226 -0.45291
+must -0.3736 -0.19977 0.10201 -0.80054 -0.0041245 -0.89754 0.58479 0.23427 -0.44577 0.069285
+months -0.047902 0.014662 0.089159 -0.83993 0.14977 -1.0455 0.43782 0.33566 -0.41078 -0.15639
+agreement -0.42855 -0.30447 0.23283 -0.90504 0.33957 -1.0427 0.55064 0.21444 -0.030315 -0.019234
+Sharon -0.69808 0.10407 -0.08883 -0.79992 0.5812 -1.3704 0.55518 -0.65019 0.26103 -0.22204
+fighters -0.39655 -0.19813 0.39436 -1.0541 0.8874 -1.0215 0.56892 0.041546 -0.087308 -0.36139
+12 -0.061409 0.12732 0.26817 -0.92274 -0.4023 -0.67929 0.46967 0.30867 -0.55092 0.21683
+help -0.21131 -0.05732 0.10325 -0.8906 0.01684 -0.99759 0.36714 0.14182 -0.34479 0.019159
+reports -0.088877 -0.17663 0.053594 -0.84656 0.040128 -0.8556 0.52898 0.23915 -0.24587 0.21702
+East -0.13076 -0.14889 0.23242 -0.90461 0.3774 -0.84839 0.49526 0.35142 -0.45245 -0.023691
+They -0.13025 -0.011063 0.16687 -1.0313 0.29914 -0.87881 0.45788 -0.13691 -0.2807 -0.09322
+brought -0.26909 -0.16135 0.31796 -0.95253 0.0753 -0.92622 0.39879 0.19852 -0.20287 0.016761
+city 0.18714 0.034228 -0.02816 -1.0117 0.45029 -0.99585 0.50922 0.12603 -0.26503 -0.035275
+Peter -0.28762 0.030569 0.17909 -0.83232 0.16319 -0.98571 0.45237 0.033855 -0.16878 0.0051439
+pay -0.19245 -0.22097 0.087784 -0.78108 -0.016373 -1.1001 0.29659 0.24276 -0.19958 -0.0072483
+hit -0.029509 -0.098019 0.2516 -1.0285 -0.30607 -0.86616 0.22168 0.18218 -0.26614 0.24156
+pressure -0.35528 -0.19301 0.12892 -0.89867 -0.22716 -0.77736 0.37761 0.095688 -0.28776 0.3276
+then -0.35176 -0.12039 -0.031151 -0.97461 -0.50257 -0.66456 0.3705 -0.16971 -0.25899 0.4849
+taken -0.57664 -0.13542 0.028271 -0.78419 0.30666 -1.1149 0.5694 -0.046325 -0.076419 -0.10462
+better -0.59077 -0.0051027 0.019789 -1.0218 0.16008 -0.91138 0.47098 -0.33228 -0.1158 0.07229
+believed -0.60145 -0.22611 0.30478 -1.1601 0.07217 -0.74924 0.26245 -0.33365 0.0025145 0.072238
+did -0.3958 -0.38905 0.18655 -1.0382 0.16038 -1.1677 0.057599 -0.20014 0.19879 -0.0073938
+took -0.10839 0.15146 -0.15541 -0.81116 -0.44431 -0.89837 0.53021 0.16957 -0.40483 0.11652
+senior -0.38855 -0.17306 0.11395 -0.94729 0.3511 -0.99365 0.53431 -0.10283 0.043805 0.069495
+held -0.41108 -0.37821 0.22226 -0.99807 0.42861 -0.94562 0.37285 -0.16397 -0.086166 -0.051732
+got -0.11397 -0.15438 0.13865 -0.97647 -0.35867 -0.73659 0.31706 0.14362 -0.24058 0.4386
+talks -0.49931 -0.24516 -0.034256 -0.92318 0.021433 -0.91349 0.44587 -0.16768 -0.11938 0.10184
+British -0.22552 -0.16832 0.169 -0.97782 -0.012919 -0.7888 0.42968 0.070682 -0.23556 0.22442
+her -0.56598 0.055176 0.047555 -0.94216 -0.24176 -0.80909 0.40857 -0.51021 -0.30132 0.29705
+without -0.25886 0.032211 0.1553 -0.93582 -0.25533 -0.85048 0.38057 0.20898 -0.3702 0.15656
+injured 0.00074489 -0.19117 0.18654 -0.98108 0.79047 -1.1871 0.46141 0.30228 -0.15298 -0.24834
+Northern -0.39826 -0.15638 0.18488 -1.0521 0.46062 -0.85188 0.52095 -0.034399 -0.10565 0.060159
+well -0.48799 -0.12225 0.032829 -1.0237 -0.50361 -0.5965 0.29598 -0.28021 -0.44472 0.27521
+maintenance -0.61375 -0.15998 -0.027396 -0.85445 0.032333 -0.9155 0.50017 0.058579 -0.084606 0.031407
+Melbourne -0.066233 0.013403 0.13338 -0.98869 -0.047375 -0.90083 0.39517 0.051843 -0.37869 0.085985
+lot -0.41807 -0.029617 0.20499 -1.028 -1.1215 -0.49001 0.24849 -0.1816 -0.20683 0.79841
+both -0.25871 -0.12474 0.18061 -1.0886 0.4581 -0.94714 0.46536 -0.34784 -0.017741 0.008381
+much -0.23254 -0.44972 0.19097 -0.92214 -0.70927 -0.51109 0.40323 0.2331 -0.29684 0.75299
+south 0.14374 -0.010507 0.36496 -1.127 0.7505 -0.80757 0.644 0.12086 -0.45496 0.0085237
+cut -0.20004 -0.091944 0.12072 -0.73048 -0.74231 -0.68913 0.36247 0.32306 -0.43043 0.42861
+accused -0.33038 -0.23776 0.13033 -0.92194 0.54449 -1.2079 0.43274 -0.019504 0.057967 -0.24631
+earlier -0.41149 -0.010543 0.10353 -0.90578 0.39562 -1.0189 0.52437 -0.15676 -0.17788 -0.17027
+asylum -0.11538 -0.20893 0.10297 -0.86109 -0.50573 -0.62727 0.37326 0.067346 -0.33961 0.6031
+10 -0.49438 -0.050398 -0.16229 -0.73428 0.53319 -1.2559 0.65049 -0.10158 -0.056939 -0.27456
+see -0.45122 -0.12177 -0.070854 -0.87715 0.040913 -0.86575 0.49731 -0.13756 -0.12492 0.24944
+too 0.18287 0.10614 0.18846 -0.96015 -0.44039 -0.75929 0.37792 0.27915 -0.43332 0.24308
+armed -0.45016 -0.20113 0.10785 -1.0627 0.9071 -1.2118 0.48822 -0.21134 0.16624 -0.29639
+across -0.075699 -0.19022 0.3084 -0.9353 0.038318 -0.82468 0.41577 0.25869 -0.37853 0.1539
+family -0.33615 -0.3111 0.19088 -0.97136 -0.15573 -0.72959 0.43323 -0.12522 -0.099185 0.41385
+such -0.26143 -0.21594 -0.064741 -0.88669 0.069956 -0.83751 0.53223 0.10565 -0.3756 0.13515
+Royal -0.27418 -0.14938 0.23725 -0.83529 -0.044903 -0.91108 0.41084 0.21683 -0.15486 0.22923
+court -0.29437 -0.34959 0.23248 -0.80349 -0.42829 -0.74289 0.328 0.19074 -0.2465 0.34028
+children -0.13545 -0.1111 0.18071 -0.91423 -0.18138 -0.891 0.37643 0.22425 -0.2165 0.20958
+shot -0.43351 -0.030061 0.041995 -1.0517 0.018698 -0.88439 0.39441 -0.18968 -0.093336 0.14719
+that's -0.32107 -0.31654 0.20442 -1.0791 -0.60048 -0.50862 0.31307 -0.054204 -0.17481 0.64328
+won -0.24805 -0.090059 0.23784 -1.0392 -0.25679 -0.82697 0.40986 0.015292 -0.13416 0.47716
+Labor -0.28452 -0.016484 0.13024 -0.82394 -0.092419 -0.88592 0.36317 0.10737 -0.36493 0.10804
+lead -0.6846 -0.14887 0.018467 -0.89821 -0.15861 -0.89749 0.43335 -0.21159 -0.091138 0.16415
+There -0.12561 -0.18585 0.26807 -1.0295 -0.14237 -0.65248 0.3227 0.12425 -0.29543 0.31737
+economy -0.13179 -0.041875 0.16667 -0.94119 -0.19573 -0.78074 0.54168 0.26438 -0.43674 0.28742
+change -0.24467 -0.17946 0.32127 -0.92869 -0.58275 -0.69887 0.24735 0.21868 -0.29403 0.5092
+Authority -0.42217 -0.25264 0.022447 -0.9286 0.28218 -0.95497 0.52786 0.025805 -0.13965 0.068731
+despite -0.27543 -0.041766 0.18001 -0.89649 -0.11945 -0.94989 0.47286 0.19412 -0.37398 0.012583
+Commission -0.53551 -0.17218 0.0092348 -0.83712 0.11043 -0.99994 0.50412 0.058624 -0.050097 0.075623
+return -0.36737 0.11555 -0.16405 -0.92815 0.37975 -1.0945 0.55478 -0.1456 -0.05689 -0.03853
+David -0.29403 -0.2399 0.18094 -0.8574 0.5962 -1.0227 0.5455 0.081132 -0.16507 -0.11859
+commission -0.50002 -0.21257 -0.037708 -0.77535 0.21794 -1.0495 0.53008 0.13119 -0.0026982 0.026108
+call -0.43877 -0.13721 0.060423 -0.79857 -0.11072 -1.0173 0.41725 0.10147 -0.11236 0.098438
+statement -0.26195 -0.34525 0.2732 -0.99823 0.61947 -1.0717 0.46618 0.012719 -0.010397 -0.12125
+past -0.28915 -0.24112 0.17668 -0.87909 0.27451 -1.0496 0.47349 0.31347 -0.22181 -0.047715
+information -0.45086 -0.23671 0.13233 -0.80661 -0.14774 -0.91442 0.46415 0.22982 -0.20506 0.13103
+even -0.27058 0.048198 0.1213 -1.0117 0.10055 -0.90405 0.63489 -0.07906 -0.12025 0.057478
+arrest -0.086376 -0.058381 0.046312 -0.99876 0.7556 -1.1305 0.61311 0.063976 -0.18774 -0.18132
+place -0.4674 -0.10882 -0.080005 -0.83937 -0.094726 -0.98055 0.49015 -0.032161 -0.28943 0.15186
+year 0.27039 -0.007233 0.012092 -0.95417 0.52353 -0.99487 0.63841 0.25306 -0.25799 0.010371
+play -0.27862 0.060327 -0.093838 -0.85528 -0.30116 -0.93001 0.49092 -0.0087302 -0.35281 0.2132
+asked -0.35223 -0.14514 -0.012209 -1.0488 0.58157 -1.2223 0.42761 -0.20006 0.052649 -0.1946
+public -0.18334 -0.1192 0.10097 -0.97386 -0.028791 -0.84513 0.43904 0.045472 -0.2538 0.19906
+working -0.4241 -0.18273 0.19895 -0.99109 -0.27194 -0.70348 0.3491 -0.006734 -0.17877 0.17169
+Union -0.70263 -0.24755 0.16509 -0.7781 -0.16795 -0.91712 0.41789 0.088135 -0.013141 -0.040677
+night 0.3338 -0.081474 0.50338 -1.0528 0.50759 -0.95374 0.4198 0.65809 -0.2817 -0.074874
+key -0.20595 0.18618 0.12332 -0.91346 -0.035322 -0.8324 0.40848 0.15421 -0.45605 -0.01563
+north 0.20955 0.088155 0.17457 -1.0621 0.55754 -0.98414 0.57941 0.19041 -0.20828 -0.012402
+continuing -0.32642 -0.11595 0.059364 -0.91338 0.023336 -0.87647 0.44437 0.19161 -0.25225 0.02376
+morning 0.11553 -0.086749 0.34038 -0.89217 -0.11287 -0.90697 0.4043 0.25783 -0.49138 0.012475
+leading -0.40523 -0.14475 0.0038426 -0.85295 0.21737 -1.0282 0.50669 -0.020732 -0.21039 -0.12723
+George -0.26592 -0.074049 0.13953 -0.91944 0.27929 -0.92697 0.43571 0.043364 -0.22461 -0.017088
+Police -0.15652 -0.10478 0.25991 -0.99124 0.11237 -0.88833 0.36961 0.1514 -0.1261 0.27155
+used -0.18958 -0.30279 0.24588 -1.0075 0.61903 -1.1353 0.33997 -0.018343 0.050965 -0.15997
+An -0.73566 -0.74929 -0.083344 -0.85408 1.1541 -1.4149 0.54669 -0.041512 0.28698 -0.33833
+southern -0.19998 -0.22607 0.29311 -1.0376 0.53842 -0.77811 0.56998 0.045338 -0.23899 0.16733
+captured -0.32281 0.002107 0.028081 -0.98009 0.4448 -1.0749 0.43451 -0.10775 -0.11066 -0.18329
+fighting -0.4162 -0.16521 0.11722 -0.96955 0.39053 -1.0457 0.48753 -0.038334 -0.084151 -0.24748
+released -0.22014 -0.23812 0.27713 -1.0124 0.2663 -0.93555 0.38565 0.0074626 -0.11337 -0.023907
+Waugh -0.22164 0.31172 -0.15097 -0.9313 -0.1693 -0.91597 0.56799 0.0009978 -0.40726 -0.1532
+Bush -0.21058 -0.060153 0.13604 -0.93105 -0.019194 -0.84492 0.35936 -0.12177 -0.25703 0.039739
+crew -0.25699 -0.1299 0.11452 -0.96782 0.51338 -1.0019 0.54531 -0.18658 -0.11839 -0.057076
+Pentagon -0.50376 -0.14918 0.27364 -0.93791 0.11913 -0.87984 0.36275 0.073624 -0.18463 0.011025
+At -0.040947 0.050582 0.15598 -0.94083 -0.22792 -0.76664 0.49245 0.17632 -0.43636 0.22072
+possible -0.18651 -0.11057 0.16929 -0.87023 -0.24695 -0.88211 0.36847 0.13555 -0.31755 0.19685
+December -0.30598 -0.10927 0.042717 -0.92013 0.52984 -1.1102 0.5527 -0.040006 -0.15978 -0.17016
+major -0.44952 -0.12188 0.14541 -0.92805 -0.015055 -0.9586 0.41725 0.020641 -0.19495 -0.050842
+economic -0.1941 -0.053602 0.18002 -0.87354 -0.12584 -0.75682 0.49391 0.26234 -0.40212 0.21214
+least -0.4078 -0.10687 0.12923 -1.0402 0.62992 -0.99161 0.53722 -0.11336 -0.11638 -0.14722
+head -0.26216 -0.18496 0.23588 -1.0013 0.25817 -0.88177 0.3235 0.0361 -0.21237 0.069663
+"If -0.31976 -0.32386 0.082281 -0.95444 -0.43486 -0.67138 0.38064 0.081474 -0.24689 0.43899
+eastern -0.68697 -0.16474 0.29965 -1.0627 0.81823 -1.0332 0.50845 -0.20251 0.10151 -0.2734
+American -0.1649 0.027856 0.083113 -0.93408 0.087311 -0.90684 0.49512 0.053134 -0.26319 0.060772
+win -0.32446 0.17796 -0.062549 -0.98462 -0.17157 -0.72011 0.52971 -0.15452 -0.46296 0.22632
+Queensland 0.016527 0.044988 0.066651 -0.98336 0.13892 -0.92871 0.43538 0.28002 -0.28573 0.070304
+winds -0.025473 -0.0088529 0.28802 -1.0519 0.037637 -0.7839 0.40417 0.21201 -0.39116 0.16377
+final -0.17817 0.037405 0.10366 -0.84472 0.24134 -1.0356 0.50595 0.20273 -0.2776 -0.029239
+Australians -0.17072 -0.092185 0.14527 -0.85526 0.19707 -0.95961 0.64503 0.21698 -0.453 -0.11382
+received -0.23806 -0.34423 0.38949 -1.0045 0.07478 -0.92204 0.20017 0.28745 -0.14465 0.1326
+give -0.304 0.094438 0.14191 -0.81612 -0.71065 -0.79482 0.35153 0.23877 -0.35483 0.26816
+Hill -0.59047 -0.42286 0.35883 -1.0262 -0.15735 -0.52297 0.46119 -0.056186 -0.089813 0.56627
+charged -0.11786 -0.16061 0.1471 -0.92866 0.27995 -1.1529 0.23516 0.0059955 -0.097249 -0.033219
+unions -0.51975 -0.38418 0.23387 -0.82299 -0.13937 -0.86892 0.37861 0.32471 -0.13947 0.16578
+behind -0.28614 -0.042677 0.25453 -0.93283 -0.42581 -0.7827 0.30868 -0.013495 -0.30824 0.3429
+within -0.076887 -0.09889 0.20743 -1.003 -0.54968 -0.64844 0.32648 0.1392 -0.52967 0.41043
+use -0.27491 -0.19348 0.06734 -0.99719 -0.31027 -0.82378 0.36823 0.011094 -0.27744 0.27496
+detainees -0.22612 -0.050389 0.22689 -0.88058 0.086684 -0.97112 0.49411 0.20803 -0.16081 0.12214
+fires 0.32144 -0.1585 0.72934 -1.1739 0.13582 -0.74533 0.26257 0.40625 -0.28486 0.26694
+director -0.16241 -0.20705 0.25557 -0.91044 -0.081741 -0.91521 0.2435 0.21604 -0.21635 0.12319
+Afghanistan, -1.0889 -0.40139 0.15554 -1.0654 0.87947 -1.1022 0.55014 -0.39719 0.21156 -0.28073
+Two 0.039483 0.10582 0.0092365 -0.86606 0.25223 -1.0646 0.52589 0.044008 -0.14283 0.068588
+large 0.060472 -0.077479 0.43381 -1.0201 0.17983 -0.93253 0.278 0.23135 -0.31978 0.12265
+your -0.017639 -0.13241 0.012114 -1.0383 -0.67943 -0.81817 0.1693 -0.11422 -0.35786 0.36868
+far -0.064217 -0.017994 0.13792 -1.0614 0.68439 -1.1755 0.46011 0.17765 -0.15849 -0.33652
+Williams -0.40211 -0.099452 0.039003 -0.94621 0.13136 -0.90192 0.449 -0.10279 -0.20152 0.032541
+India -0.48269 0.028407 0.23908 -0.9185 0.78234 -1.3202 0.63529 -0.1195 -0.045859 -0.58352
+damage -0.071572 -0.053924 0.26836 -0.91188 -0.25969 -0.78529 0.3703 0.42384 -0.40989 0.28618
+known -0.34469 -0.33091 0.15857 -1.096 0.087767 -0.74902 0.34905 -0.19035 -0.035211 0.20957
+child -0.022981 -0.24182 0.31703 -0.8507 -0.24837 -0.86491 0.27661 0.40675 -0.14529 0.28148
+million -0.68378 -0.34192 0.10969 -0.83539 0.16323 -1.0869 0.3785 0.039167 -0.029955 -0.029096
+legal -0.3215 -0.2221 0.14686 -0.88873 -0.33615 -0.78393 0.38696 0.15758 -0.22108 0.29789
+able -0.37697 -0.13266 0.22291 -0.95258 -0.20151 -0.85636 0.33762 -0.17219 -0.15911 0.36692
+stop -0.41403 -0.08617 0.19544 -0.98944 0.12766 -0.91673 0.31749 -0.10126 -0.1994 0.037702
+high -0.034964 -0.0018303 0.032594 -0.95032 0.40361 -1.1331 0.54806 0.02092 -0.12699 -0.16178
+may -0.088139 0.17051 -0.11849 -0.83318 -0.21708 -0.98455 0.50248 0.20886 -0.60238 -0.1351
+long -0.58904 -0.29478 0.20529 -1.0233 0.20982 -0.93275 0.38765 -0.3167 -0.12836 -0.037084
+soldiers -0.54631 -0.24664 -0.0013115 -0.98134 0.50507 -0.97717 0.5551 -0.23703 -0.051598 -0.064674
+centre -0.16357 -0.088873 0.23781 -0.91368 0.50806 -1.0944 0.56083 0.43539 -0.15288 -0.12649
+water 0.0163 0.15115 0.20899 -0.90915 -0.17346 -0.85685 0.42673 0.20662 -0.37154 0.3471
+process -0.3912 -0.33273 0.13729 -0.98103 -0.18425 -0.71818 0.32323 -0.064817 -0.20143 0.42002
+interest -0.22889 -0.10378 -0.04432 -0.77426 0.41609 -1.0316 0.66139 0.07026 -0.27459 -0.038257
+remain -0.33546 -0.041843 0.19591 -0.98053 0.27623 -0.93689 0.51675 0.089744 -0.15228 -0.088631
+Cup -0.39222 0.20582 -0.19441 -1.04 0.49042 -1.0834 0.74202 -0.31475 -0.10536 -0.2315
+forced -0.26048 -0.20505 0.033136 -0.93682 0.54523 -1.2521 0.37319 0.11684 -0.073001 -0.23612
+cricket -0.079899 0.23933 0.080047 -0.87499 0.057947 -0.94591 0.57732 0.27945 -0.42145 -0.17575
+Centre -0.098491 -0.098919 0.29968 -0.88787 0.36578 -0.91792 0.519 0.44358 -0.24666 -0.097863
+there's -0.14267 -0.37116 0.43398 -1.1361 -0.58011 -0.43721 0.23176 -0.046849 -0.2551 0.63644
+services -0.027633 -0.11073 0.24148 -0.94517 0.099167 -0.81618 0.48252 0.27186 -0.22543 0.21946
+role -0.25075 -0.14711 -0.016529 -0.84185 -0.30629 -0.91197 0.4177 0.0034674 -0.28862 0.18281
+morning. -0.13115 -0.12871 0.30529 -0.93988 0.13057 -0.9026 0.42958 0.10169 -0.33749 -0.0042525
+seen -0.25503 -0.14886 -0.0071094 -0.96396 0.29621 -0.96718 0.52247 -0.13596 -0.084971 0.042143
+might -0.12136 -0.015283 0.34012 -0.99759 0.18655 -0.90543 0.46259 0.18435 -0.22641 -0.00030777
+radio -0.31587 -0.025496 0.12696 -0.90522 0.19487 -1.0003 0.47502 -0.046322 -0.19526 -0.051484
+15 -0.0027815 -0.010627 0.35139 -0.91984 0.031506 -1.0024 0.38359 0.30927 -0.34202 -0.027485
+failed -0.28408 -0.13997 0.13126 -0.95783 0.23679 -1.0263 0.38851 0.03713 -0.10962 -0.058015
+"It -0.2127 0.0065145 0.15431 -0.9023 -0.72538 -0.57713 0.47729 0.13848 -0.30031 0.56853
+conditions -0.30619 -0.33596 0.28766 -0.97664 0.16578 -0.84737 0.39348 0.24603 -0.24549 0.12226
+heard -0.036849 -0.13805 0.32521 -0.96742 -0.21658 -0.76473 0.27268 0.16032 -0.36014 0.33078
+training -0.33004 -0.10865 0.19393 -0.90618 0.15863 -1.0472 0.41182 0.043335 -0.24102 -0.12324
+Palestinians -0.084665 -0.042344 -0.32423 -1.0157 0.99253 -1.327 0.65852 -0.17126 0.014889 -0.21945
+already -0.23222 -0.17401 0.1536 -0.9142 0.0067545 -0.88513 0.39916 0.076321 -0.14774 0.19729
+taking -0.34111 -0.083927 -0.076089 -0.8193 -0.3044 -0.88678 0.43857 0.010839 -0.37356 -0.063635
+towards -0.24141 -0.067541 0.1225 -0.93239 0.29631 -1.0728 0.46656 0.034587 -0.15042 -0.10345
+dead -0.30904 -0.2196 0.0068343 -1.0166 0.67039 -1.0183 0.52865 -0.24931 0.014482 0.069926
+same -0.28721 -0.05681 0.067372 -0.89551 -0.75304 -0.62097 0.23121 -0.10415 -0.427 0.42835
+Lee -0.57065 0.287 -0.31648 -0.79795 0.079652 -1.0968 0.48908 -0.042647 -0.30256 -0.36131
+board -0.085416 -0.21346 0.17409 -0.90036 0.13884 -0.87745 0.48846 0.12395 -0.19114 0.17559
+latest -0.1553 -0.098918 0.045245 -0.85515 0.51157 -1.0497 0.60468 0.062235 -0.3672 -0.13693
+However, -0.35929 -0.23019 0.23856 -1.0014 0.3696 -1.105 0.36722 -0.010449 0.081887 -0.039713
+due 0.26217 0.15237 0.08374 -0.82533 0.73949 -1.1983 0.7406 0.44587 -0.39938 -0.33752
+rates -0.23545 -0.052488 0.23878 -0.89587 0.30736 -0.91888 0.59585 0.16932 -0.24906 0.027702
+thought -0.34475 -0.14509 0.32866 -0.99768 -0.054778 -0.80094 0.37854 0.053205 -0.15685 0.18667
+Alliance -0.74028 -0.15661 -0.045948 -1.0052 0.34901 -0.97857 0.50019 -0.30252 -0.01314 -0.063821
+canyoning 0.00096151 -0.12593 0.14691 -0.81281 0.069324 -1.097 0.39818 0.10198 -0.35588 -0.11744
+offer -0.50201 -0.24419 -0.022542 -0.82268 0.15344 -1.0637 0.50403 -0.11537 -0.020506 -0.019694
+strikes -0.13813 -0.10681 0.082023 -0.95404 0.37004 -0.9642 0.49136 -0.0023333 -0.20633 0.023301
+half -0.29326 -0.17483 0.10851 -0.99788 0.093941 -0.83061 0.39508 -0.13702 -0.15929 0.11612
+Shane -0.40518 0.13598 -0.035674 -0.97505 0.06794 -1.0409 0.40337 -0.10238 -0.069763 -0.0046095
+storm 0.022854 -0.01158 0.37061 -1.0505 0.087099 -0.8141 0.37323 0.10991 -0.26812 0.23018
+I'm -0.21456 -0.013578 -0.085031 -0.91216 -0.49315 -0.83478 0.57767 -0.052232 -0.30253 0.2994
+aircraft 0.042737 -0.070513 0.23712 -1.0473 0.20835 -0.94714 0.36037 0.081297 -0.24822 0.043258
+bowler -0.34089 0.33838 -0.0016996 -0.93128 0.025725 -0.98494 0.49746 -0.047297 -0.2425 -0.10725
+Adelaide -0.044939 0.079867 0.074706 -0.9117 -0.040047 -0.9078 0.47254 0.1713 -0.47185 -0.026522
+great -0.071226 0.030578 0.21264 -1.0573 -0.11088 -0.70531 0.4732 0.026426 -0.30536 0.34767
+army -0.42827 -0.206 -0.03653 -1.0522 0.77898 -1.1822 0.52938 -0.39546 0.2095 -0.16596
+position -0.57825 -0.26886 0.043969 -0.81203 0.15898 -1.0362 0.46842 0.12703 -0.12563 0.07192
+administration -0.58092 -0.22505 0.080669 -0.81091 0.26554 -1.035 0.53144 0.077576 -0.095358 0.10347
+control -0.050748 -0.17503 0.22343 -0.97335 0.094469 -0.84336 0.42982 0.27929 -0.24484 0.24974
+violence -0.2323 -0.070436 -0.0079771 -0.94813 0.36988 -0.97238 0.48493 0.013139 -0.16171 -0.021263
+continue -0.28833 -0.014519 0.15639 -0.97033 0.23828 -0.90154 0.47412 0.29717 -0.22186 0.0051712
+news -0.59215 -0.15621 0.017262 -0.94796 0.27223 -0.91985 0.42573 -0.12415 -0.18765 -0.061016
+After -0.33253 0.14233 0.075992 -0.78174 -0.0085916 -0.97334 0.50921 0.097841 -0.27502 0.032031
+series -0.2891 -0.085626 0.17297 -0.90951 -0.092798 -0.70652 0.57757 -0.022069 -0.20133 0.29959
+York -0.23988 0.20609 0.11751 -1.0175 0.17168 -0.82336 0.44681 0.045135 -0.14793 -0.023503
+ago -0.37093 -0.35865 0.10456 -1.0275 -0.18844 -0.77547 0.29485 0.23651 -0.19287 0.21604
+strong -0.12487 -0.19973 0.31163 -0.93742 -0.26036 -0.73653 0.34761 0.14708 -0.36251 0.34623
+likely -0.5693 -0.3563 0.11266 -1.0417 0.012559 -0.80848 0.34499 -0.17244 -0.031366 0.24764
+later -0.54452 -0.075866 0.11857 -0.81861 0.2648 -1.0883 0.51375 -0.080365 -0.077602 0.030132
+today. -0.17828 -0.08966 0.19215 -0.86792 0.20829 -0.98452 0.53873 0.094942 -0.31039 0.004548
+Australia, -0.19928 -0.043556 0.17708 -0.80673 0.045817 -0.89914 0.66936 0.28328 -0.47921 -0.097498
+along -0.35875 -0.23487 0.13841 -0.99281 0.28109 -0.96264 0.46829 -0.088655 -0.25587 -0.18575
+Blue -0.32921 0.17929 0.14827 -0.96856 0.19427 -0.97128 0.51554 0.15698 -0.37641 -0.024242
+line -0.25023 -0.10539 0.15418 -0.97211 0.24883 -1.0799 0.38552 0.039534 0.052945 -0.069131
+right -0.10617 -0.11315 0.23144 -0.97867 0.33221 -1.0104 0.52065 0.12385 -0.14311 -0.133
+claimed -0.35025 -0.17717 0.27294 -0.95935 0.34933 -0.97263 0.38423 -0.025992 -0.20119 -0.17129
+Nations -0.46612 -0.40396 0.13747 -0.84393 0.28359 -0.93601 0.51385 0.24839 -0.26527 0.10986
+risk -0.16904 -0.17755 0.30726 -0.98054 -0.065309 -0.8264 0.33338 0.11302 -0.1219 0.21404
+own -0.014097 -0.26602 0.11106 -0.89549 0.46691 -0.89507 0.45256 -0.030322 -0.22677 0.04712
+buildings -0.061143 -0.0047946 -0.025975 -0.89984 0.095714 -0.994 0.49 0.15945 -0.31889 0.022768
+hospital -0.037349 -0.12998 0.18955 -0.95268 0.21243 -0.91339 0.46714 0.13491 -0.21672 0.073965
+chief -0.36788 -0.1658 0.11414 -0.84898 -0.033546 -0.87442 0.46921 0.15726 -0.25923 0.1799
+matter -0.54564 0.006823 -0.029043 -0.86175 0.077958 -1.0526 0.52024 -0.086473 -0.3017 -0.070833
+concerned -0.1732 -0.20173 0.32926 -1.0133 0.084236 -0.86609 0.36841 0.20493 -0.10091 0.17977
+campaign -0.40699 -0.27161 0.025515 -0.89658 0.24313 -0.98682 0.44068 0.035115 -0.091242 0.032817
+show -0.21416 -0.27701 0.3153 -1.0321 -0.15971 -0.67999 0.3832 0.09236 -0.17343 0.30864
+Adventure -0.040517 -0.13532 0.353 -0.95168 0.023476 -0.93465 0.30689 0.23992 -0.28893 0.11019
+guilty -0.20216 -0.043522 0.083572 -0.92687 0.20071 -0.92099 0.43184 -0.067119 -0.3179 0.00091516
+African -0.2024 0.35056 -0.066211 -0.93961 0.14744 -0.97384 0.62526 -0.010557 -0.41916 -0.24435
+envoy -0.67473 -0.14806 -0.046447 -0.81502 0.33356 -1.1331 0.44533 -0.16175 0.05812 -0.085223
+homes 0.14088 -0.034475 0.39976 -1.0534 0.10088 -0.71782 0.52129 0.29373 -0.34711 0.14863
+boat 0.0086287 -0.10716 0.22524 -0.98937 -0.10005 -0.61566 0.41495 -0.068118 -0.2481 0.32673
+rate -0.39919 -0.069286 0.0034546 -0.84483 0.2335 -0.92006 0.62933 0.1589 -0.31521 0.08722
+month -0.091638 -0.071536 0.12925 -0.9252 0.43046 -1.0922 0.45127 0.1197 -0.20081 -0.17437
+west 0.0094504 0.022666 -0.043279 -1.0138 0.60269 -0.97813 0.59047 0.02209 -0.39855 -0.031355
+launched -0.28076 0.0035345 0.09612 -1.0289 0.55087 -1.0934 0.46398 -0.030732 -0.08573 -0.20215
+Ms -0.1693 -0.024214 0.2036 -1.065 -0.48284 -0.8086 0.24235 -0.10775 -0.33241 0.33664
+move -0.12252 -0.17686 0.078811 -0.85205 -0.16308 -0.942 0.38466 0.059054 -0.22594 0.2758
+industrial -0.51341 -0.1521 0.060852 -0.82769 -0.37555 -0.86717 0.37908 0.13903 -0.1444 0.093291
+special -0.24727 -0.25317 0.05622 -0.88789 0.0026893 -0.82646 0.35959 0.20746 -0.21304 0.23176
+Downer -0.58852 -0.22938 0.35931 -0.94535 -0.29586 -0.76081 0.32812 -0.16499 -0.045356 0.26698
+Kandahar -0.29682 -0.11128 0.13414 -0.92091 0.58496 -1.0662 0.5407 0.041656 -0.14686 -0.16771
+plans -0.26949 -0.17732 0.129 -0.91765 0.26921 -0.93932 0.51331 0.029577 -0.18829 0.081238
+officers -0.28559 -0.22919 0.1258 -0.93854 0.61275 -1.151 0.53171 -0.063958 0.03405 -0.1115
+town -0.035656 -0.22186 0.089919 -1.04 0.84213 -1.14 0.45628 -0.166 -0.035811 -0.095347
+firefighters -0.12692 -0.19933 0.46779 -1.0823 0.31873 -0.84548 0.45963 0.24173 -0.21691 0.042936
+decision -0.48339 -0.020654 0.030877 -0.85514 0.19915 -1.0503 0.51939 0.055609 -0.1193 0.00057066
+flight -0.17949 -0.1495 0.18775 -0.88991 0.39984 -1.0605 0.47715 0.15983 -0.10361 -0.20439
+death 0.057547 -0.13383 0.095832 -0.95011 0.12673 -0.86911 0.37013 0.14583 -0.21993 0.17032
+Swiss -0.25035 -0.17656 0.15292 -0.85423 -0.50274 -0.69927 0.32663 0.21845 -0.31788 0.43299
+me -0.35403 -0.10642 -0.16637 -0.77878 -0.60913 -0.94614 0.47597 -0.023491 -0.16753 0.2014
+Trade -0.21383 0.019332 0.046942 -0.93263 0.30361 -1.0082 0.44523 0.032281 -0.26381 -0.049947
+men -0.08405 -0.17308 -0.12009 -0.97632 0.90007 -1.2422 0.57854 -0.26449 0.020887 -0.25508
+today, -0.13313 -0.11522 0.17201 -0.90986 0.29172 -0.96278 0.53729 0.11652 -0.26396 0.079324
+captain -0.33279 0.11211 0.05412 -0.91826 0.17814 -0.98857 0.53571 -0.015231 -0.25209 -0.11477
+really -0.35473 -0.097594 0.19306 -0.91792 -0.78484 -0.678 0.38567 0.20476 -0.3916 0.50253
+planning -0.24482 -0.083709 0.097347 -0.85534 -0.10172 -0.92347 0.36516 0.0084243 -0.38041 -0.081162
+jobs -0.14945 -0.047958 0.27719 -0.84695 -0.48319 -0.70103 0.48378 0.16802 -0.35257 0.43338
+Laden's -0.84876 -0.22134 0.40069 -1.1128 0.73137 -1.0188 0.2719 -0.36349 0.013102 -0.33474
+event -0.024604 -0.072418 0.23616 -0.96716 0.32212 -0.99293 0.64446 0.38017 -0.13658 0.0059757
+enough -0.32023 0.098302 0.0096032 -0.85613 -0.16518 -0.99403 0.48051 -0.030379 -0.2096 0.080321
+bus -0.71213 -0.079262 -0.25133 -0.92337 1.0806 -1.4162 0.78796 -0.48473 0.078979 -0.33458
+UN -0.54816 -0.048432 0.12611 -0.86137 0.27431 -1.0655 0.59413 -0.031415 -0.13637 -0.003523
+Zinni -0.50128 0.017488 -0.1918 -0.93632 0.22123 -0.94346 0.60073 -0.18784 0.0024587 0.066372
+important -0.22451 -0.2042 0.18431 -0.9583 0.047671 -0.85765 0.44231 0.11295 -0.27287 0.10683
+health -0.10558 -0.17499 0.2288 -0.94858 0.22865 -0.97927 0.37151 0.10052 -0.10578 0.02896
+others -0.28787 -0.395 0.34208 -1.0667 0.40654 -0.84971 0.45401 0.0066727 -0.10223 0.083959
+Industrial -0.57372 -0.09693 0.02127 -0.83164 -0.49892 -0.89496 0.39075 0.063725 -0.16263 0.13486
+Mark -0.36877 0.27568 -0.03104 -0.90449 -0.48173 -0.91208 0.40797 0.147 -0.29862 0.023239
+union -0.6202 -0.28923 0.21896 -0.77973 -0.28947 -0.94769 0.37978 0.32314 -0.027591 0.07152
+"He -0.45909 -0.21562 0.058993 -0.84483 -0.099631 -0.87833 0.43008 0.078521 -0.29326 0.10338
+late -0.3475 -0.25341 0.25738 -0.9397 0.35775 -1.0721 0.51631 0.10118 -0.22909 -0.032748
+sure -0.14035 -0.058039 0.11566 -0.92932 -0.49738 -0.66551 0.38233 0.16903 -0.45261 0.49004
+side -0.31318 -0.055371 -0.13075 -0.93074 -0.095578 -0.98997 0.42557 -0.00036938 -0.31781 0.0050635
+weapons -0.4718 -0.2076 0.10377 -0.92709 0.22634 -0.9232 0.38295 -0.13678 -0.20357 -0.015071
+Service 0.10996 -0.18879 0.25716 -1.0271 -0.25646 -0.64395 0.33704 0.37956 -0.33758 0.45461
+jail -0.37638 0.026194 -0.025368 -0.92421 0.23885 -0.87398 0.55937 -0.048547 -0.18372 0.054618
+Zealand 0.0043498 0.13487 0.09079 -0.8991 -0.030375 -0.8348 0.50432 0.36604 -0.35552 0.095627
+International -0.3996 -0.14946 0.11295 -0.73793 0.39268 -1.1519 0.57884 0.2536 -0.20171 -0.08369
+probably -0.33818 -0.12145 0.18426 -0.95942 -0.076147 -0.75996 0.4724 0.059822 -0.3102 0.21379
+network -0.76982 -0.14867 0.29491 -1.0421 0.31272 -0.82967 0.35551 -0.3277 -0.070193 -0.056567
+Australia. -0.22091 -0.017653 0.187 -0.80241 0.0045872 -0.84669 0.65882 0.30617 -0.49004 -0.096869
+find -0.3074 0.13148 0.22503 -0.90008 -0.27705 -0.77319 0.38533 -0.092883 -0.28932 0.13818
+my -0.55451 0.12144 -0.0093719 -0.97671 -0.62483 -0.66736 0.48245 -0.25872 -0.50878 0.077701
+station -0.53447 -0.29584 0.13297 -0.85269 0.32235 -1.0057 0.42896 0.17143 -0.1653 0.042014
+Bichel -0.40509 -0.044277 -0.0088616 -0.88798 0.24663 -1.1217 0.52461 0.015741 -0.17843 -0.15178
+1999 -0.18455 0.069741 0.18764 -0.8856 0.53729 -1.1298 0.46995 0.0017765 -0.30787 -0.26738
+life -0.24612 -0.33524 0.18092 -0.93094 0.15349 -0.92696 0.32574 0.04259 -0.15932 0.10387
+National -0.32576 -0.18403 0.13924 -0.69377 0.30312 -1.1257 0.63131 0.3921 -0.2447 0.010923
+prepared -0.21991 -0.20488 0.12785 -0.885 0.14089 -0.97711 0.3799 0.14702 -0.18668 0.038318
+home -0.17758 -0.073843 -0.056814 -0.9827 0.25884 -0.88551 0.56779 -0.03591 -0.26311 -0.022736
+Sydney, 0.38089 0.10582 0.37774 -1.1327 0.37215 -0.75697 0.59491 0.24816 -0.40122 0.19168
+political -0.23895 -0.17101 0.093477 -0.8838 0.17914 -0.98695 0.50693 0.25728 -0.17116 0.046091
+14 -0.31575 -0.076226 0.018773 -0.82727 0.83066 -1.2612 0.6004 -0.37257 0.009522 -0.22477
+helicopters -0.23121 -0.12721 0.081126 -0.9841 0.62077 -1.0337 0.47125 0.0061639 -0.20141 -0.12253
+wants -0.39181 -0.28013 0.16434 -0.88982 0.34146 -1.0773 0.43535 0.0763 -0.12478 -0.13743
+General -0.33731 -0.36169 0.34625 -0.91321 -0.25595 -0.83866 0.32095 0.27438 -0.14076 0.26922
+carrying -0.37832 -0.025182 -0.028287 -0.89457 0.22042 -1.0354 0.52053 -0.11677 -0.2781 -0.29793
+Middle -0.40746 -0.1289 -0.089367 -0.91371 0.2377 -0.95979 0.52702 -0.015126 -0.079519 0.1346
+using -0.26039 -0.19974 0.0027378 -0.85818 0.07639 -0.91421 0.45785 -0.044387 -0.19986 0.036548
+northern -0.1314 -0.10187 0.16731 -1.0625 0.51044 -0.84401 0.53194 0.10286 -0.16342 0.10458
+operations -0.4037 -0.3316 0.025667 -0.89908 0.47275 -1.0024 0.56941 0.22835 -0.24424 -0.031465
+defence -0.34835 -0.2211 0.050224 -0.95564 0.17038 -0.8682 0.51416 -0.0026649 -0.068871 0.15956
+carried -0.26137 -0.13546 -0.019057 -0.94378 0.33946 -1.0995 0.42136 0.014035 -0.021979 -0.016227
+Hollingworth -0.030212 0.07015 0.23553 -0.96126 -0.022018 -0.89547 0.47997 0.25544 -0.2351 0.065736
+comes -0.12515 -0.21734 0.30096 -0.96863 0.091718 -0.82099 0.52801 0.28297 -0.21324 0.20068
+person -0.31443 -0.073895 -0.07208 -0.99595 0.39315 -1.0454 0.47549 -0.15756 -0.21276 -0.042487
+Unions -0.58393 -0.34103 0.16019 -0.85448 -0.010625 -0.86669 0.40667 0.13006 -0.11536 0.059159
+Jihad -0.25044 -0.1311 0.11354 -0.91889 0.56741 -1.0475 0.54933 -0.036782 -0.07521 -0.13064
+every -0.23024 -0.18855 0.20065 -0.97747 -0.54483 -0.62561 0.38552 0.046008 -0.21172 0.57272
+Israelis -0.3279 -0.092185 -0.3384 -1.0311 0.98386 -1.3114 0.63479 -0.45063 0.025191 -0.21609
+years. -0.071905 -0.16077 0.028975 -0.98995 0.16042 -0.80394 0.58575 0.14118 -0.29376 0.15611
+Relations -0.47999 -0.22928 0.007061 -0.82066 0.38031 -1.0734 0.60641 0.2529 -0.34925 -0.099694
+abuse -0.41617 -0.17554 0.11074 -0.97323 -0.09883 -0.84307 0.39191 0.030254 -0.13567 0.22793
+kilometres -0.23393 -0.077966 0.26269 -0.95865 0.21062 -0.98748 0.44364 0.1892 -0.14065 0.021218
+until -0.36532 -0.015918 0.015624 -0.91358 0.06594 -0.95147 0.50103 -0.04358 -0.14284 0.088419
+tried -0.3707 -0.23133 0.098954 -0.93306 0.34043 -1.1061 0.2772 0.037984 0.050213 -0.051578
+become -0.30943 -0.1512 0.041733 -0.94765 -0.16194 -0.72357 0.51597 0.0077977 -0.36744 0.21203
+Fire 0.44447 -0.19245 0.64022 -1.2496 -0.43456 -0.4926 0.11751 0.48495 -0.56266 0.60643
+alleged -0.13987 -0.14386 0.23963 -0.96288 0.2887 -1.1934 0.18076 0.18774 0.023957 -0.033033
+policy -0.23414 -0.14868 0.037421 -0.92876 -0.17332 -0.80174 0.47739 0.083862 -0.1634 0.30527
+job -0.29817 0.15863 -0.011452 -0.73014 -0.44408 -0.80389 0.59752 0.047052 -0.40874 0.17
+race -0.23623 0.12272 -0.12789 -0.87828 0.12137 -1.0007 0.55153 0.084974 -0.30908 -0.056355
+raids -0.4405 0.055256 0.062423 -0.97207 0.59969 -1.1253 0.52018 -0.20184 -0.073733 -0.19791
+Security -0.25179 -0.26211 0.012311 -0.99779 0.54335 -1.0733 0.49651 -0.13276 -0.044672 0.03434
+each -0.25856 0.074421 0.05647 -0.83658 -0.083764 -0.9344 0.56955 0.26072 -0.35617 0.015302
+said, -0.25902 -0.23099 0.28328 -1.0505 -0.28032 -0.65906 0.29252 -0.11605 -0.18789 0.41448
+deal -0.17886 -0.34216 0.20539 -0.84713 -0.13152 -0.9213 0.3724 0.21342 -0.14918 0.36828
+making -0.28803 -0.08816 0.0096906 -0.82431 -0.64894 -0.738 0.39803 0.1298 -0.52578 0.13695
+emergency -0.26612 -0.1099 0.065147 -0.91568 0.41001 -1.0744 0.52717 0.037593 -0.18953 -0.10363
+sent -0.19357 -0.27141 0.29263 -0.77914 -0.046579 -0.91439 0.52573 0.31398 -0.057896 0.13753
+plane -0.34638 -0.1023 0.082522 -0.96101 0.2565 -1.0639 0.36914 -0.16762 -0.031945 -0.043187
+McGrath -0.17043 -0.051306 -0.020217 -0.9169 0.27615 -1.0228 0.48797 0.020804 -0.22426 -0.031734
+seekers -0.48644 -0.44184 0.13376 -0.96734 0.12767 -0.66738 0.49504 -0.098521 -0.095709 0.32882
+immediately -0.36355 -0.227 0.043711 -0.96031 0.26305 -1.0306 0.46492 0.039272 -0.097878 0.042057
+opening -0.17873 0.038679 0.038125 -0.85227 0.24574 -1.1265 0.48156 -0.01966 -0.35637 -0.31265
+financial -0.30118 -0.27108 0.12209 -0.83559 0.11444 -1.009 0.40422 0.124 -0.10492 0.10687
+opposition -0.52953 -0.28352 0.024003 -0.82079 0.18808 -1.0265 0.48246 0.12408 -0.16336 0.077984
+beat -0.44952 0.047498 -0.059399 -1.057 -0.36293 -0.67571 0.42371 -0.24185 -0.28456 0.22016
+HIH -0.47638 -0.30372 0.11696 -0.78749 -0.012724 -0.80612 0.40415 0.14367 -0.23078 0.29014
+am -0.46544 0.14943 -0.027456 -0.98103 0.27046 -0.88687 0.48696 -0.33832 -0.11751 -0.088589
+proposed -0.22032 -0.21796 0.32289 -0.90467 -0.45117 -0.7144 0.30345 0.28281 -0.27091 0.30881
+evidence -0.40336 -0.16851 0.057301 -0.99487 0.063628 -0.8195 0.45084 -0.0023912 -0.15993 0.12882
+issue -0.20286 0.14266 0.12136 -0.92091 -0.2528 -0.76598 0.39263 0.30105 -0.31183 0.19294
+community -0.23492 -0.37318 0.25533 -0.91363 0.059018 -0.88329 0.39885 0.32284 -0.1588 0.23015
+suspected -0.34525 -0.10702 0.18567 -1.0473 0.27441 -0.84641 0.39509 -0.14546 -0.14767 -0.013768
+bombing -0.27627 -0.12407 -0.025868 -0.8967 0.11571 -0.88214 0.44439 -0.13446 -0.24008 -0.0067358
+deaths -0.080647 -0.10233 0.13786 -0.92447 0.033617 -0.85785 0.36213 0.16064 -0.30454 0.10584
+radical -0.14642 -0.02874 0.0063376 -0.8976 0.41309 -1.0202 0.49822 0.15056 -0.23713 -0.13541
+laws -0.21059 -0.044915 0.19047 -0.93203 0.11675 -0.97785 0.42655 0.20573 -0.3067 -0.03825
+went -0.45038 -0.11098 0.18192 -0.94314 -0.014208 -0.95047 0.4455 0.14278 -0.11996 0.082557
+allow -0.090794 -0.074926 0.1755 -0.97726 0.088112 -0.94784 0.37043 0.15258 -0.090754 0.18805
+result -0.29583 -0.18522 0.2357 -0.91104 0.0054902 -0.92295 0.34697 0.17311 -0.19238 0.17041
+"It's -0.37833 -0.11473 -0.013531 -1.0353 0.011208 -0.81535 0.59501 -0.086298 -0.14879 0.24644
+Senator -0.36911 -0.24955 0.33815 -1.0253 0.043189 -0.81357 0.30949 0.046325 -0.17431 0.24162
+Department -0.30033 -0.34507 0.26669 -0.84222 -0.049876 -0.91746 0.46195 0.35211 -0.11222 0.14093
+warplanes -0.29239 -0.09469 0.18425 -0.94719 0.32186 -0.9657 0.37826 -0.04488 -0.047268 0.035734
+Council -0.35524 -0.3422 0.12564 -0.89316 0.2501 -0.97434 0.483 0.20005 -0.14954 0.0051255
+Ariel -0.48719 0.037788 0.031469 -0.74902 -0.10285 -0.97876 0.61608 -0.13939 -0.040404 0.20097
+different -0.098991 -0.16926 0.22102 -0.98309 0.02349 -0.91166 0.49676 0.16403 -0.161 0.13375
+"There -0.19236 -0.17333 0.25456 -1.0477 -0.0088454 -0.76699 0.321 0.086953 -0.15734 0.26294
+rejected -0.28531 -0.049575 0.055564 -0.89875 0.026118 -0.95553 0.38317 0.037622 -0.12241 0.067581
+reported -0.19702 -0.19608 0.18906 -0.91063 -0.021168 -0.88541 0.3912 0.13244 -0.17061 0.17465
+One -0.1883 -0.021232 0.32823 -1.0043 0.34116 -1.0373 0.30109 -0.0078587 0.017532 -0.097623
+details -0.2459 -0.0043803 0.17589 -0.8699 -0.030403 -0.96819 0.41229 0.11994 -0.21326 0.03971
+hundreds -0.1179 -0.098527 0.14361 -0.99988 0.28367 -0.97639 0.47041 0.074146 -0.18637 -0.08792
+Secretary -0.35576 -0.20606 0.075651 -0.93688 0.191 -0.90759 0.51749 -0.11536 -0.036384 0.024759
+full -0.12038 0.14044 0.20748 -0.92319 -0.29811 -0.66853 0.4504 0.078529 -0.33246 0.34165
+calls -0.29975 -0.15709 0.15753 -0.84773 -0.14706 -0.86039 0.36401 0.27873 -0.19795 0.14575
+drop -0.42159 -0.13677 0.080038 -0.94191 0.11131 -1.0293 0.42309 -0.16101 -0.19606 0.01346
+growth -0.07259 -0.28981 0.27615 -1.0298 0.29384 -0.78796 0.46195 0.11602 -0.19833 0.19302
+hard -0.20498 -0.02465 0.1661 -0.9054 0.0013864 -0.94509 0.44938 -0.043615 -0.14109 0.10793
+fight -0.25409 -0.15276 0.3869 -1.0736 0.57205 -0.94709 0.49104 0.060624 -0.061899 -0.23552
+Woomera -0.24158 -0.055035 0.19377 -0.91819 -0.016829 -0.87466 0.50685 0.29162 -0.17826 0.2253
+allegations -0.30052 -0.30396 0.21668 -0.86851 -0.016164 -0.94555 0.37399 0.33265 -0.20512 0.17885
+caught -0.28616 -0.038503 0.10752 -0.95922 0.27824 -1.0011 0.51716 0.058671 -0.05291 -0.085257
+opened -0.16128 -0.099327 0.041465 -0.97883 0.51974 -1.1129 0.40655 -0.14017 -0.0966 -0.24245
+getting -0.60234 -0.078298 -0.16695 -0.90073 0.0048955 -0.99791 0.46585 -0.19268 -0.1735 -0.077584
+bombings -0.18144 -0.045621 -0.08839 -0.93942 0.35837 -0.94121 0.53292 -0.071254 -0.24936 -0.024677
+although -0.39983 -0.086445 0.2458 -1.0253 -0.13943 -0.83683 0.36993 -0.12281 -0.076385 0.21606
+building -0.17313 -0.06911 0.02497 -0.86335 -0.10145 -0.91135 0.46626 0.11933 -0.34054 -0.01262
+always -0.099338 -0.16392 0.23017 -0.99304 -0.091784 -0.71582 0.35264 0.086487 -0.2496 0.3216
+2 -0.15545 0.035836 -0.0077972 -0.98733 0.25396 -1.0182 0.48962 -0.015344 -0.22089 -0.24752
+look -0.20214 -0.12049 0.13222 -0.91602 -0.7254 -0.70241 0.40645 0.13836 -0.3255 0.44921
+Jewish -0.056948 -0.041342 0.04173 -0.94601 0.38495 -0.97229 0.52959 0.0022186 -0.2429 -0.076027
+source -0.33497 -0.087328 -0.1113 -0.9172 0.34065 -0.9776 0.57745 -0.1448 -0.21374 -0.012633
+flights -0.22701 -0.10587 0.22915 -0.90944 0.37374 -1.0377 0.47777 0.1837 -0.14216 -0.19452
+quite -0.25175 -0.22422 0.27164 -1.0326 -0.18216 -0.74256 0.33936 0.06437 -0.28912 0.29318
+killing -0.25686 -0.14649 -0.00014256 -0.90274 -0.12336 -0.98858 0.44911 0.0068644 -0.20173 -0.040376
+Strip -0.1356 -0.26696 0.18214 -0.98423 0.82974 -1.1219 0.42037 0.036817 -0.078735 -0.12709
+bid -0.37077 -0.23103 0.19757 -1.0146 0.81412 -1.2049 0.28057 -0.12465 0.22361 -0.28075
+understand -0.37883 -0.15315 0.098055 -0.94035 -0.0089921 -0.86651 0.40035 0.027029 -0.22866 0.048836
+year's -0.0078248 -0.09619 0.17557 -0.97446 0.1173 -0.83811 0.59289 0.18105 -0.29963 0.17022
+innings -0.17982 0.076362 -0.15785 -0.92241 0.49047 -1.13 0.59651 0.023168 -0.32511 -0.32717
+access -0.35938 -0.23449 0.048085 -0.88403 0.11329 -0.97133 0.5078 -0.035407 -0.18539 0.068704
+ago. -0.16745 -0.11987 0.047255 -0.98502 0.091684 -0.94066 0.48054 0.059722 -0.26523 0.11259
+young -0.052772 -0.1533 0.21176 -0.98047 -0.1703 -0.86867 0.28618 0.16079 -0.2614 0.1791
+himself -0.41513 -0.01386 -0.013133 -0.96159 -0.03035 -0.83948 0.4415 -0.073896 -0.18098 0.13272
+meet -0.6892 0.12278 -0.059613 -0.82415 0.1907 -1.1716 0.49659 -0.23193 -0.034196 -0.17992
+On -0.24926 0.2477 -0.054523 -0.80452 -0.51011 -0.76588 0.50566 0.037485 -0.20762 0.23825
+Commonwealth -0.23119 -0.14886 0.14519 -0.90128 -0.0035436 -0.92767 0.44947 0.10889 -0.23743 0.18265
+Bureau -0.16077 -0.021117 0.24038 -0.91542 0.17771 -0.8947 0.5087 -0.02436 -0.29051 0.085372
+targets -0.14882 -0.11944 0.053442 -0.94995 0.34197 -0.88971 0.48677 -0.026809 -0.26734 0.032219
+"We're -0.086301 -0.012208 0.073553 -0.97498 0.04867 -0.88478 0.40897 0.088269 -0.24305 0.13491
+militant -0.4391 -0.29585 0.026277 -0.95421 0.69054 -1.1851 0.54031 -0.11661 -0.024484 -0.18447
+running -0.079897 0.05711 0.038853 -0.91501 -0.13324 -1.0015 0.39354 0.18559 -0.45126 -0.18506
+caves -0.28691 0.11883 0.066076 -1.0442 0.57614 -0.99247 0.55009 -0.031571 -0.032725 -0.20788
+declared -0.052889 -0.05141 0.15401 -0.96843 0.46864 -1.0297 0.46719 0.10203 -0.21326 -0.12957
+reached -0.44212 -0.16723 0.12568 -0.91301 0.024306 -0.90236 0.39833 0.062853 -0.08777 0.046893
+18 0.054921 -0.11309 0.11605 -0.77062 -0.060609 -0.92333 0.35447 0.30628 -0.45244 0.18787
+20 -0.22533 -0.1106 0.14139 -0.8831 0.4986 -1.1438 0.49541 0.11371 -0.20783 0.0040105
+among -0.49796 -0.27051 0.11415 -0.94687 0.47517 -0.96776 0.47731 -0.12582 -0.13888 -0.090727
+based -0.25069 -0.22415 0.25071 -0.98506 0.13426 -0.90901 0.29891 0.037888 -0.048462 0.078416
+Howard -0.094855 0.025978 0.21725 -1.0421 -0.22107 -0.83696 0.3628 0.026865 -0.2639 0.11659
+try -0.46085 -0.26442 0.20995 -0.99564 0.10121 -0.90349 0.35863 -0.057345 -0.12138 0.15032
+believes -0.67137 -0.17036 0.26685 -1.0768 0.14739 -0.72407 0.35888 -0.36828 -0.040496 -0.0035927
+July -0.32955 -0.36062 0.39482 -0.94008 -0.14972 -0.71303 0.32794 0.19853 -0.2126 0.25866
+actually -0.2289 -0.073151 0.14274 -0.93537 -0.43631 -0.73256 0.33608 0.085956 -0.31936 0.35602
+currently -0.30254 -0.11033 0.17194 -0.84159 -0.084001 -0.8061 0.50987 0.17284 -0.28415 0.13813
+announced -0.42028 -0.18456 0.18197 -0.9778 0.14524 -0.92541 0.39784 -0.054199 -0.044202 0.042121
+clear -0.01766 -0.14938 0.25252 -1.0235 -0.07958 -0.72622 0.41753 -0.029375 -0.27612 0.2745
+State -0.37626 -0.24598 0.14162 -0.87221 0.56295 -1.087 0.61413 0.022581 -0.19923 -0.25108
+Parliament -0.46191 -0.25638 0.23833 -0.90053 0.43435 -0.99233 0.53144 0.1489 -0.10922 -0.11228
+here -0.29754 -0.28154 0.27744 -1.0449 -0.3836 -0.61764 0.23703 0.12542 -0.27986 0.43399
+Britain -0.41509 -0.16547 0.19246 -0.93156 0.4166 -1.0847 0.5131 0.12528 -0.23074 -0.25184
+year, -0.10651 -0.013289 0.075046 -0.9704 0.31971 -0.93009 0.60224 0.21259 -0.30398 0.0022467
+executive -0.15144 0.032854 0.10856 -0.85138 -0.23817 -0.86353 0.43772 0.14153 -0.40333 0.21818
+surrender -0.39416 -0.1037 0.12542 -0.92244 0.022295 -0.92797 0.50559 0.03057 -0.17871 0.13987
+Alexander -0.39375 -0.070985 0.019329 -0.82865 0.25753 -1.1342 0.5418 -0.13856 -0.035291 -0.0044781
+flying -0.12358 -0.085658 0.0039751 -0.90072 -0.052444 -0.98421 0.44649 0.076044 -0.35365 -0.11421
+weekend -0.27191 -0.20531 0.14953 -1.0588 0.21455 -0.8166 0.39648 -0.052563 -0.24843 0.1731
+time. -0.27678 -0.035373 0.015089 -0.85877 -0.17774 -0.82998 0.40956 0.0045983 -0.29376 0.14535
+human -0.325 0.040529 0.1051 -0.91902 0.32687 -1.0524 0.5324 0.015324 -0.20708 -0.086047
+Immigration -0.62166 -0.32346 0.16311 -0.83192 0.17072 -0.95213 0.47813 0.17228 -0.11817 0.084171
+days. -0.21962 -0.072669 0.16506 -0.92774 0.24268 -1.0086 0.4454 0.040181 -0.32095 -0.066109
+airline -0.23745 -0.10759 0.25576 -1.0061 -0.017262 -0.93116 0.33664 0.060348 -0.14106 0.0597
+river -0.31547 -0.14686 0.24461 -0.79813 -0.53261 -0.88609 0.33665 0.082408 -0.27995 0.35627
+annual -0.07151 -0.0082519 0.10891 -0.90364 0.13375 -0.95719 0.39925 0.14119 -0.27644 0.024862
+yet 0.032871 0.011395 0.30045 -0.93906 -0.21429 -0.78424 0.52674 0.21043 -0.27387 0.25205
+we're -0.11724 -0.2325 0.16908 -1.0447 -0.12363 -0.72712 0.359 0.076173 -0.26036 0.31823
+travel -0.24645 -0.12484 0.29633 -0.85444 -0.19344 -0.84535 0.40432 0.08839 -0.29996 0.27319
+sex -0.12885 -0.17146 0.17803 -0.94059 0.29546 -0.91931 0.50233 0.047885 -0.077329 0.21403
+expect -0.24927 -0.091875 0.0239 -0.89745 -0.25451 -0.82458 0.39189 0.11725 -0.31161 0.24229
+outside -0.29979 -0.087116 0.094266 -0.8943 0.00045409 -0.93309 0.41995 0.077897 -0.32089 0.0948
+gave -0.20511 0.10134 0.27444 -1.0108 -0.50872 -0.68088 0.28404 0.065595 -0.31698 0.37159
+future -0.13391 -0.038217 0.22378 -1.0037 0.1114 -0.96022 0.29762 0.099481 -0.26009 0.013145
+people, -0.12077 -0.13285 0.064225 -0.89586 0.20543 -0.92167 0.52999 0.11773 -0.3363 0.041737
+Kallis -0.22297 0.042648 0.024925 -0.89704 -0.25563 -0.93736 0.43078 0.055985 -0.2876 0.061194
+arrived -0.20971 -0.11303 0.15902 -1.0245 0.44714 -1.0242 0.40287 -0.048344 -0.0052894 -0.069317
+responsibility -0.10669 -0.19514 0.24343 -1.0107 0.27539 -0.92534 0.34749 0.13125 -0.22071 -0.023336
+Chief -0.21446 -0.028111 0.10836 -0.85874 0.050764 -0.97388 0.51436 0.1159 -0.30729 0.038645
+sources -0.26177 -0.10926 0.082926 -0.91144 0.37326 -0.96432 0.54344 -0.004703 -0.21427 -0.05484
+expressed -0.29158 -0.1309 0.084792 -0.89884 0.067026 -0.92867 0.46335 -0.021058 -0.17247 0.10516
+again -0.28054 0.10503 -0.031662 -0.87632 0.13049 -1.0395 0.5817 0.1995 -0.31702 -0.13758
+needs -0.69947 -0.16154 0.12364 -1.0156 0.20539 -0.97623 0.36055 -0.27137 -0.10422 -0.095865
+times -0.19059 -0.082645 0.23329 -0.95971 -0.10218 -0.79772 0.4028 0.085123 -0.26922 0.19396
+leader, -0.63614 -0.15872 -0.021229 -0.97841 0.23224 -1.0678 0.37307 -0.24234 0.069236 0.059875
+media -0.36359 -0.33511 0.11944 -0.96789 0.53827 -1.1395 0.51612 0.081919 -0.025864 -0.17633
+overnight 0.0026782 -0.16337 0.5379 -1.0077 0.30795 -0.95688 0.40591 0.4137 -0.20355 0.067067
+caused -0.32583 -0.15928 0.24236 -1.1669 0.65764 -1.1243 0.40977 -0.038673 0.019246 -0.15685
+investigation -0.24167 -0.16591 0.095002 -0.84671 0.093706 -0.96468 0.47883 0.28363 -0.23212 0.12051
+victory -0.14237 -0.031346 -0.040868 -0.86656 0.10501 -0.99638 0.47517 0.052506 -0.37078 0.020271
+cost -0.33531 -0.31573 0.21847 -0.93661 0.098379 -0.70413 0.42365 0.12404 -0.28671 0.21083
+means -0.21139 -0.21608 0.035407 -0.95129 0.066174 -0.84813 0.46002 0.059957 -0.26126 0.21452
+guides -0.195 -0.11011 0.1319 -0.89564 0.31203 -1.038 0.43995 0.011187 -0.29771 -0.16936
+Afghanistan's -1.0852 -0.42476 0.11174 -1.0582 1.0038 -1.1647 0.58459 -0.38897 0.24901 -0.34677
+Test. -0.35461 0.17268 -0.081656 -0.87091 -0.061653 -0.94691 0.54565 -0.035941 -0.46732 -0.15517
+parties -0.25651 -0.23092 0.10005 -0.87591 -0.013637 -0.92388 0.46985 0.22713 -0.19787 0.16205
+November -0.15432 -0.02002 -0.035073 -0.88768 0.66978 -1.2153 0.60292 -0.031687 -0.15572 -0.21273
+away -0.026386 -0.011804 0.03243 -0.95371 0.010804 -0.85442 0.58628 0.049947 -0.38244 0.091771
+Glenn -0.2025 0.10252 -0.18489 -0.82056 -0.064589 -0.94112 0.5735 0.08154 -0.16768 0.098507
+night. 0.14726 -0.13411 0.43348 -0.98423 0.48031 -1.0034 0.48321 0.56319 -0.26636 -0.052575
+less -0.34774 -0.087851 0.020444 -0.975 0.24762 -0.89823 0.53689 -0.19108 -0.17694 0.10416
+gives -0.30648 0.00095754 0.12032 -0.87802 -0.081254 -0.92676 0.50007 0.22389 -0.18439 -0.0023295
+refused -0.20994 -0.12008 0.043993 -0.95577 0.43249 -1.1323 0.47532 0.082949 -0.042781 -0.082605
+decided -0.25164 -0.15128 0.14698 -0.93189 0.42 -1.1056 0.4041 0.03367 -0.13229 -0.021897
+wage -0.15918 -0.18509 0.15199 -0.94062 0.094331 -0.86955 0.37411 0.32563 -0.27524 0.11126
+certainly -0.36459 -0.25731 0.20807 -0.96518 -0.3249 -0.76573 0.36741 0.088155 -0.2325 0.37688
+face -0.41217 -0.10768 0.014726 -0.9521 0.52086 -1.1 0.50755 0.03151 -0.021106 -0.17162
+having -0.21479 -0.094558 0.02323 -0.92666 -0.26909 -0.80156 0.35772 -0.052212 -0.30304 0.14079
+bombers -0.25448 -0.24683 -0.018675 -1.0088 0.98883 -1.1495 0.57514 -0.24854 0.0085853 -0.30157
+13 -0.2514 -0.17806 0.23568 -0.97128 0.20978 -0.84804 0.46292 0.032578 -0.44878 -0.12467
+More -0.17663 -0.063707 0.21824 -0.97878 0.1765 -0.85505 0.40461 0.18004 -0.35206 0.12186
+Musharraf -0.31525 -0.17799 0.15549 -0.96105 0.086704 -0.96223 0.36748 -0.010628 -0.14574 0.11553
+Sir -0.079121 -0.16903 0.152 -0.92032 -0.10546 -0.7863 0.37201 0.1975 -0.27092 0.33351
+Western -0.35848 -0.0024335 0.15828 -1.0712 0.76845 -1.0325 0.61841 -0.21278 -0.080128 -0.20051
+Warne -0.17762 0.088479 0.051883 -0.98411 0.27026 -1.0481 0.45843 0.04756 -0.20877 -0.23601
+we've -0.27202 -0.062883 0.079416 -1.1073 -0.22395 -0.74813 0.32714 -0.062417 -0.18222 0.27504
+returned -0.25684 -0.014633 -0.013888 -0.9261 0.34609 -1.0596 0.40779 -0.1468 -0.081473 -0.062052
+house -0.25438 -0.19207 0.098137 -1.08 0.3123 -0.92787 0.35562 0.0027978 -0.076825 0.14032
+figures -0.24194 -0.11312 0.34844 -0.95948 -0.0096087 -0.81971 0.48248 0.17377 -0.29287 0.066865
+soon -0.50325 -0.24627 0.077906 -0.85691 -0.057127 -0.89574 0.49108 -0.072185 -0.13845 0.16944
+Opposition -0.55172 -0.28795 0.025411 -0.82279 0.15422 -1.0296 0.43879 0.087711 -0.12644 0.081617
+Energy -0.52219 0.0083993 0.13363 -0.99076 0.077371 -0.91272 0.46538 -0.10207 -0.15002 -0.045709
+appeared -0.07688 -0.12422 0.15721 -0.98941 0.32351 -1.022 0.33754 0.039885 -0.14885 -0.028518
+"What -0.11813 -0.12509 0.11716 -0.96938 0.18886 -0.96622 0.43089 -0.085791 -0.11728 0.12231
+parts -0.15738 -0.13451 -0.028285 -0.92109 0.36979 -1.0102 0.59461 0.23022 -0.24569 -0.019808
+point -0.31625 -0.15522 0.11486 -0.85343 -0.091388 -0.86706 0.49356 0.15778 -0.21028 0.15196
+weeks -0.31093 -0.12954 0.15842 -1.0454 0.14362 -0.78349 0.3397 -0.15216 -0.18535 0.086455
+step -0.24769 -0.085087 0.14128 -1.0074 -0.053275 -0.74336 0.48037 -0.027143 -0.17706 0.29505
+Hicks -0.7295 -0.27189 -0.031627 -1.0161 0.38705 -1.0018 0.4731 -0.20284 0.079142 -0.086215
+ended -0.46385 -0.18155 -0.094221 -0.89408 0.8742 -1.3229 0.50061 -0.12044 0.0057563 -0.33374
+big -0.10401 0.029808 -0.11517 -1.132 0.90247 -1.2428 0.42649 -0.23731 -0.041625 -0.43859
+run -0.026621 0.058854 0.23927 -0.88576 0.22445 -0.97455 0.42719 0.25457 -0.43438 -0.13246
+Robert -0.18885 -0.24728 0.16502 -0.98001 0.23693 -0.83256 0.53339 0.0054403 -0.14761 0.18343
+rather -0.3921 -0.18474 0.26465 -1.0403 -0.1311 -0.66115 0.3769 -0.095472 -0.27836 0.36664
+dispute -0.38151 -0.13673 -0.055228 -0.96796 0.28504 -0.99155 0.46686 -0.097563 -0.17168 -0.066103
+thousands -0.20815 -0.052841 0.13491 -1.0094 0.16653 -0.96091 0.44757 0.058202 -0.20876 0.060942
+countries -0.30045 -0.16149 0.20757 -0.9137 0.050078 -0.93056 0.44059 0.26011 -0.20321 0.065365
+Reserve -0.12533 -0.037669 0.16344 -0.95301 0.48337 -0.99076 0.5605 0.14818 -0.23771 -0.18154
+biggest -0.17696 -0.16626 0.0047131 -0.98654 0.71474 -1.1002 0.51817 -0.038413 -0.11861 -0.14601
+can't -0.14316 -0.038243 0.029441 -0.95136 -0.084281 -0.90605 0.38663 0.036244 -0.27861 0.048795
+region -0.67664 -0.11581 0.1894 -0.92088 0.14594 -1.015 0.44602 0.078378 0.019968 -0.07271
+issues -0.18967 0.2033 0.15594 -0.96702 -0.22947 -0.86135 0.43302 0.28347 -0.29927 0.10877
+beyond -0.31479 -0.26768 0.24727 -1.0479 -0.071206 -0.80629 0.36612 -0.041367 -0.24782 0.18122
+huge -0.081257 0.32418 0.050619 -0.83504 -0.26709 -0.95097 0.45568 0.19193 -0.39068 0.031293
+them. -0.22048 -0.31103 0.13912 -0.97638 -0.10318 -0.75878 0.42885 -0.033493 -0.25774 0.35585
+break -0.034984 -0.16235 0.10839 -0.87804 -0.10658 -0.79174 0.56585 0.22224 -0.2667 0.28082
+ensure -0.13504 -0.10694 0.22563 -0.9484 -0.052151 -0.80939 0.35235 0.16253 -0.32167 0.21016
+ground -0.23677 -0.18518 0.27125 -1.0726 0.19723 -0.85167 0.26207 -0.030968 -0.17042 0.04772
+tourists -0.36032 -0.11573 0.067654 -0.80911 0.14706 -1.072 0.548 0.14094 -0.24755 -0.11331
+shortly -0.22155 -0.017607 0.093779 -1.0259 0.31995 -0.92463 0.50257 0.045857 -0.19887 0.016661
+something -0.21869 -0.1265 0.17115 -0.92387 -0.29442 -0.71776 0.42465 0.14374 -0.37656 0.23146
+terms -0.45899 -0.1411 0.16535 -0.99854 -0.086348 -0.81406 0.37841 -0.10845 -0.26085 0.059
+top -0.52027 0.15306 -0.28095 -0.71283 -0.35815 -1.0341 0.46039 -0.13882 -0.28352 0.043035
+safety -0.1443 -0.1671 0.089085 -0.86247 0.047306 -0.9441 0.42246 0.14677 -0.2587 0.1754
+whose -0.15884 -0.16834 0.28541 -0.90295 0.061624 -0.948 0.33678 0.10392 -0.17018 0.067711
+order -0.61937 -0.18925 -0.087087 -0.87335 0.32193 -1.1745 0.53799 -0.12251 -0.097608 -0.11936
+21 -0.35252 -0.3032 0.19308 -0.73447 0.1381 -1.053 0.5186 0.49776 -0.37451 -0.02863
+seven -0.32449 -0.090214 0.01174 -0.86455 0.13791 -0.91959 0.56921 -0.12242 -0.083617 0.06326
+worst -0.093513 0.13321 0.21029 -0.89142 -0.16556 -0.82448 0.47646 0.23528 -0.4839 0.10489
+200 -0.07257 0.050468 0.089691 -0.95671 -0.28409 -0.75853 0.45668 0.16083 -0.44957 0.24796
+changes -0.23516 -0.19769 0.32922 -0.90195 -0.31492 -0.78549 0.30952 0.20265 -0.23301 0.35419
+Mountains -0.52557 -0.0070327 0.1814 -1.1009 0.42383 -0.93624 0.50681 0.024818 -0.13433 -0.11177
+1,000 -0.13408 -0.0031082 -0.0071555 -1.0484 0.062027 -0.81847 0.46957 0.075143 -0.34528 0.0069611
+attempt -0.21777 -0.13646 0.037072 -0.99718 0.42526 -1.057 0.46832 -0.007266 -0.13694 -0.073133
+wave -0.1237 0.06939 0.13382 -0.97782 -0.31499 -0.75068 0.31747 0.15389 -0.31235 0.26425
+She -0.57889 0.024528 0.29046 -0.75118 -0.40824 -0.97542 0.2939 -0.14587 -0.20861 0.24839
+heavy -0.25764 -0.17552 0.18984 -0.94983 0.12568 -0.86752 0.3063 -0.017535 -0.095908 0.13068
+banks -0.21874 -0.1408 0.14122 -0.9399 0.16793 -0.85874 0.41583 0.11357 -0.21469 0.05204
+struck -0.21501 -0.12046 0.25122 -0.90835 0.044797 -0.91014 0.40803 0.11327 -0.32215 0.14165
+bill -0.51599 -0.271 -0.00073813 -0.9119 0.64295 -1.1705 0.52009 0.043544 0.089397 -0.25129
+massive -0.32483 -0.06258 -0.013938 -0.83339 -0.061484 -0.86455 0.4527 0.091288 -0.25824 0.15906
+foreign -0.4108 -0.35808 0.051144 -0.92501 0.25962 -1.0232 0.3764 0.024047 -0.0070203 -0.0068743
+Monday -0.30755 0.021482 0.19795 -0.85557 0.11402 -0.9588 0.44421 0.17013 -0.24844 -0.019671
+residents -0.18282 -0.17343 0.16833 -0.89066 0.055371 -0.95199 0.4216 0.31978 -0.299 0.0037635
+Detention -0.39476 -0.14458 0.15996 -0.82447 0.041516 -0.93888 0.51081 0.35091 -0.20424 0.057306
+protect -0.15792 -0.18836 0.083067 -0.88514 -0.20558 -0.83509 0.45219 0.19162 -0.32459 0.25114
+crash -0.37378 -0.15719 0.27305 -0.95702 0.087902 -0.82382 0.34335 0.042488 -0.15859 0.054763
+Kabul -0.26443 -0.16481 0.3114 -0.94373 -0.19575 -0.79068 0.40535 0.15031 -0.24102 0.28756
+Jacques -0.16548 0.058804 0.152 -0.97539 0.056449 -0.92379 0.45371 0.11116 -0.24996 -0.015329
+gunmen -0.063602 -0.13022 0.0073489 -0.93044 0.24926 -1.0779 0.53094 -0.0072605 -0.221 -0.029566
+River -0.34144 -0.19561 0.22203 -0.75358 -0.077981 -1.023 0.34737 0.094786 -0.17132 0.1026
+denied -0.38527 -0.30648 0.16933 -0.94346 0.72253 -1.2464 0.36061 -0.068193 0.098559 -0.25226
+Governor-General -0.32618 -0.34426 0.26649 -0.91605 0.058059 -0.92269 0.40493 0.17553 -0.14275 0.17862
+act -0.50422 -0.27666 0.10494 -1.0576 0.42009 -1.0005 0.3776 -0.14642 -0.01395 -0.10325
+Safety 0.075012 -0.13791 0.25712 -0.92599 0.1798 -0.91217 0.42602 0.34453 -0.34069 0.074462
+he's -0.76149 -0.25187 0.16369 -1.0071 -0.25816 -0.7671 0.30205 -0.32931 0.036876 0.21559
+general -0.34393 -0.29248 0.27743 -0.82782 -0.18493 -0.88201 0.34027 0.26915 -0.13627 0.16395
+inside -0.18642 -0.14043 0.18562 -0.95672 -0.01729 -0.88924 0.36219 0.24348 -0.22756 0.077796
+"In -0.37246 0.014417 -0.04582 -0.72871 -0.92338 -0.67055 0.44785 0.30004 -0.49557 0.20112
+feel -0.18015 -0.17165 0.10289 -1.0476 -0.39224 -0.67754 0.40273 0.046544 -0.27883 0.51188
+beginning -0.31562 -0.040915 0.044385 -0.94139 0.12054 -0.97017 0.45147 -0.047391 -0.3183 -0.19903
+it, -0.058407 0.022432 0.35018 -0.94972 -0.4629 -0.56738 0.37619 0.096636 -0.4813 0.30966
+Israel, -0.31752 -0.079865 -0.38344 -0.98436 0.97659 -1.2919 0.65736 -0.39128 0.072302 -0.12737
+Pakistani -0.88321 -0.48353 0.10139 -0.98664 0.71346 -1.2353 0.44264 -0.25487 0.059342 -0.40104
+decide -0.18269 -0.13465 0.062756 -0.90553 0.10744 -0.99522 0.40557 0.15254 -0.20989 0.15111
+though -0.43559 -0.026187 0.16191 -0.99179 -0.087738 -0.82109 0.37879 -0.17426 -0.10786 0.15234
+Russian -0.41079 -0.094812 -0.0072505 -0.90948 0.27982 -1.0214 0.55015 -0.10899 -0.13083 -0.074736
+trees -0.045673 -0.030206 0.30045 -0.91687 -0.1976 -0.84275 0.45741 0.4433 -0.34606 0.24949
+giving -0.11861 0.065388 -0.08762 -0.88097 0.031239 -1.0487 0.46707 -0.0079952 -0.36886 -0.17169
+attacks. -0.25392 -0.032236 0.016311 -1.1569 1.1794 -1.3267 0.5168 -0.23359 -0.076471 -0.64156
+commanders -0.46258 -0.24119 0.10232 -0.96894 0.4164 -1.0378 0.4784 -0.054836 -0.077072 -0.14536
+president -0.33677 -0.30219 0.11797 -0.84533 0.16699 -1.0107 0.424 0.19408 -0.20011 0.0065484
+witnesses -0.14075 -0.027639 0.11048 -0.9081 0.07699 -0.87359 0.48174 0.018874 -0.28779 0.079268
+"They -0.2338 -0.139 0.11731 -1.0643 0.39083 -0.97842 0.42528 -0.15528 -0.069783 -0.059073
+fact -0.60335 -0.19652 0.11474 -1.006 -0.10808 -0.7693 0.35506 -0.13315 -0.030147 0.2337
+longer -0.55887 -0.24209 0.29306 -0.99047 0.18211 -0.94771 0.3721 -0.19865 -0.02897 0.17724
+Powell -0.51358 -0.10002 0.075846 -1.0233 -0.023909 -0.83873 0.41738 -0.22141 -0.25485 0.096454
+collapse -0.19438 -0.084022 0.080058 -0.90043 0.00063324 -0.83945 0.49146 0.062255 -0.23982 0.1335
+boy -0.16384 -0.11246 0.014311 -1.1305 0.25015 -0.68509 0.4744 -0.26999 -0.076922 0.1922
+involved -0.23396 -0.1966 0.18392 -1.0082 0.052488 -0.92838 0.3691 0.11502 -0.10534 0.17506
+forward -0.23214 -0.14703 0.15094 -0.98371 -0.23519 -0.98797 0.36629 0.099242 -0.22711 0.1875
+militia -0.43077 -0.22653 -0.067671 -0.92817 0.68308 -1.1518 0.55044 -0.1347 -0.026065 -0.12228
+situation -0.5339 -0.14488 0.032761 -0.79896 0.25281 -1.0406 0.50343 0.18194 -0.12588 0.0018157
+ASIO -0.23567 -0.30546 0.39225 -0.89077 -0.5876 -0.67046 0.27892 0.33283 -0.21739 0.49707
+response -0.1571 -0.087453 0.26491 -0.95863 -0.086937 -0.78355 0.31053 0.20016 -0.26188 0.041097
+As -0.34266 -0.024876 0.19252 -0.9934 0.56732 -1.1094 0.45626 0.15147 -0.27764 -0.42731
+disease -0.030411 0.047909 0.062412 -0.95125 0.2199 -0.95773 0.42018 0.080438 -0.22037 0.027476
+placed -0.31413 -0.078587 0.059193 -0.94205 -0.040332 -0.99733 0.36912 -0.054451 -0.2258 0.095615
+chance -0.52828 -0.12485 0.095101 -1.0493 -0.17369 -0.78443 0.39704 -0.036474 -0.14919 0.25015
+address -0.21817 0.010137 0.092077 -0.89644 -0.15827 -0.90724 0.42719 0.10645 -0.30291 0.22053
+States. -0.16341 -0.15763 0.15576 -0.92377 0.64676 -1.1334 0.57526 -0.03691 -0.25793 -0.2704
+party -0.35005 -0.18344 -0.040608 -0.96678 0.11393 -0.97366 0.46049 -0.010412 -0.17804 0.11221
+entered -0.19469 -0.22309 0.15619 -0.87189 0.35959 -0.97711 0.5195 0.14584 -0.16542 -0.056635
+Day -0.26448 -0.067392 -0.0062925 -0.9677 0.62361 -1.1215 0.65933 -0.16353 -0.13543 -0.27834
+short -0.29861 -0.096258 0.051112 -0.99715 0.2501 -0.87024 0.50111 0.013569 -0.15373 0.040428
+Boxing -0.17312 0.25159 -0.16338 -0.84506 -0.1219 -1.0209 0.54002 0.042375 -0.49181 -0.31169
+Martin -0.23051 0.14617 -0.1226 -0.81136 -0.39254 -0.929 0.4845 0.17899 -0.31655 0.10688
+Donald -0.3668 -0.31585 0.18055 -0.88702 -0.082501 -0.83075 0.43068 0.057527 -0.060287 0.3466
+Local -0.4166 -0.31576 0.13843 -0.79674 -0.091945 -0.97806 0.37698 0.074839 -0.084672 0.20961
+followed -0.068441 0.044316 0.091602 -0.93525 0.32473 -1.0785 0.4884 0.17029 -0.1689 -0.057822
+warned -0.27548 -0.099088 0.13718 -0.94938 0.18932 -0.9183 0.2788 -0.048953 -0.1141 -0.11389
+48 -0.84419 -0.18653 0.29293 -0.83778 0.3217 -0.97718 0.43714 -0.17774 0.010283 -0.20241
+serious -0.32479 -0.19952 0.01172 -0.87772 0.61493 -1.0947 0.58385 -0.10002 -0.019106 -0.014508
+inquiry -0.29588 -0.25489 0.087912 -0.87935 0.29723 -1.0276 0.47503 -0.048841 -0.10658 -0.055423
+sort -0.051322 -0.15173 0.14363 -0.88693 -0.18115 -0.69567 0.41564 0.012624 -0.30785 0.38157
+prevent -0.17982 -0.1031 0.17494 -0.90397 -0.22325 -0.77348 0.48222 0.2477 -0.2021 0.29359
+strike -0.26249 -0.10207 0.077732 -0.93117 0.2996 -0.95785 0.51303 0.083574 -0.1658 0.014509
+Anglican -0.1808 -0.056134 0.25429 -1.002 -0.16747 -0.79432 0.39045 0.19596 -0.15251 0.22658
+cancer -0.2918 -0.14501 0.19404 -0.87255 0.014863 -0.88379 0.3752 0.15281 -0.1911 0.11058
+bring -0.082268 0.0039916 -0.0066104 -0.83081 -0.13348 -0.92132 0.54214 0.17418 -0.4141 -0.034829
+available -0.32264 -0.075275 0.033305 -0.98118 0.086585 -0.90595 0.42473 -0.20275 -0.16315 0.13557
+morning, 0.071681 -0.084214 0.37602 -0.92478 -0.07168 -0.85084 0.41671 0.25985 -0.39865 0.13377
+Brett -0.39988 0.12227 -0.14695 -0.84352 0.26524 -1.0828 0.5565 -0.031347 -0.19666 -0.13479
+money 0.10693 -0.24461 0.33205 -0.98638 -0.26491 -0.85389 0.28839 0.2983 -0.35361 0.30637
+Muslim -0.37519 -0.1731 0.21203 -0.97612 -0.01014 -0.85301 0.41571 0.0089575 -0.18362 0.096238
+mountains -0.52235 -0.0313 0.19906 -1.0696 0.45976 -0.95238 0.44859 -0.0063259 -0.10718 -0.17352
+main -0.4307 -0.096279 0.18151 -0.87066 -0.3373 -0.87146 0.37781 0.24209 -0.3128 0.046284
+overnight. -0.05978 -0.17425 0.50098 -0.97825 0.32724 -0.98814 0.44684 0.38269 -0.20557 0.056648
+border -0.61028 -0.14578 0.017971 -0.95586 0.41711 -1.0875 0.48467 -0.21194 0.0062121 -0.20204
+current -0.15874 -0.1446 0.20205 -0.79474 -0.091161 -0.85191 0.54808 0.33502 -0.32097 0.15464
+AFP -0.43069 -0.06174 0.041412 -1.035 0.36056 -0.91176 0.55137 -0.32412 0.032452 -0.011461
+Daryl -0.49073 -0.11078 0.0059247 -0.9298 0.20923 -0.97749 0.41212 -0.31057 -0.061639 -0.028531
+level -0.23517 -0.16461 0.18845 -1.0032 0.53353 -0.97729 0.52615 -0.16901 -0.1271 0.019749
+never -0.47821 -0.085864 0.14045 -0.78639 0.063474 -1.027 0.46941 -0.1867 -0.1437 0.026296
+cannot -0.32269 -0.13929 0.24225 -1.0206 -0.03028 -0.9009 0.3505 -0.049398 -0.092678 0.139
+royal -0.54 -0.23168 0.16237 -0.86746 0.082618 -0.9545 0.37876 -0.044377 -0.038106 0.020396
+calling -0.27627 -0.10037 0.003318 -0.8251 -0.072697 -1.0253 0.38615 0.12013 -0.18733 -0.10744
+Anthony -0.29676 -0.056315 -0.14688 -0.86255 0.1869 -0.96764 0.48321 0.04497 -0.19118 0.084065
+lives -0.33202 -0.28284 0.25269 -1.0548 0.70137 -1.1561 0.44864 -0.04875 0.078248 -0.23948
+according -0.36104 -0.20245 0.088021 -0.90923 0.24059 -1.0262 0.48457 0.03779 -0.21451 -0.22357
+Geoff -0.18856 -0.29532 0.25044 -1.0282 -0.25594 -0.67297 0.33488 0.17971 -0.23647 0.43254
+state's -0.18054 -0.27202 0.37814 -1.1113 0.30717 -0.89957 0.35238 -0.019664 -0.18023 0.1016
+"This -0.42224 -0.05965 0.14074 -0.88887 -0.14413 -0.90186 0.42267 -0.049023 -0.23884 0.11355
+movement -0.15536 -0.36437 0.16587 -0.91736 0.45287 -1.0791 0.48025 0.20168 -0.027551 -0.036841
+Justice -0.16172 -0.08131 0.01455 -0.8468 -0.25169 -0.91213 0.50762 0.17318 -0.24023 0.29564
+Vaughan -0.52576 -0.039893 0.10583 -0.93523 0.36761 -1.0281 0.5194 0.0068841 -0.16615 -0.14374
+deadly -0.20142 -0.12306 -0.062067 -0.97255 0.40423 -0.8879 0.50728 -0.052048 -0.19292 0.084739
+ruled -0.071221 -0.070779 0.19589 -0.96566 0.27633 -1.0952 0.38484 0.27097 -0.084554 -0.013572
+fast -0.37751 0.026316 -0.011812 -0.89545 0.23886 -1.0078 0.54184 0.22559 -0.32299 -0.14724
+led -0.35603 -0.40632 0.26926 -1.1032 0.86261 -1.1361 0.38627 -0.12178 0.28368 -0.1525
+insurance -0.28391 -0.13389 0.054707 -0.92632 0.23121 -0.92442 0.46559 0.15454 -0.19599 0.050628
+burning -0.16641 -0.098378 0.33981 -0.95832 -0.2186 -0.84922 0.37866 0.13416 -0.35948 0.076853
+fired 0.24878 -0.16657 0.38399 -1.1783 0.50918 -0.91939 0.31095 0.27848 -0.22079 -0.022189
+anything -0.18968 -0.25379 0.13284 -0.87075 -0.4172 -0.75594 0.36244 0.088717 -0.29159 0.26666
+study 0.0016092 -0.019816 0.10753 -0.98043 0.24135 -0.87145 0.48581 0.057884 -0.26077 0.080121
+"These -0.45929 -0.19157 0.10329 -0.98596 0.10077 -0.87076 0.40393 -0.10262 -0.096417 0.082231
+trip -0.25091 -0.34149 0.097458 -0.86118 0.71392 -1.1633 0.39403 0.025984 -0.16656 -0.15394
+Workers -0.55923 -0.13892 0.13249 -0.86279 -0.048219 -0.84642 0.50666 -0.00838 -0.098121 0.067299
+speaking -0.37817 -0.098246 0.004995 -0.87793 -0.15698 -0.88181 0.432 0.067994 -0.26448 -0.022681
+White -0.60335 -0.1973 0.2229 -0.93662 0.36819 -0.92215 0.37535 -0.068352 -0.12827 -0.17589
+cent. -0.11218 -0.078002 0.27892 -0.91788 0.18222 -0.98627 0.5622 0.40077 -0.24635 0.040361
+difficult -0.25006 -0.2152 0.12371 -0.9856 0.1282 -0.83924 0.4174 0.041246 -0.22196 0.079786
+rule -0.2409 -0.055641 0.12339 -0.86898 0.12055 -1.0115 0.45835 0.22055 -0.09693 0.096909
+Allan -0.40569 0.11774 0.033441 -0.902 0.1465 -0.98182 0.5628 -0.15242 -0.18927 -0.010615
+costs -0.1733 -0.14964 0.12231 -0.90499 0.17765 -0.84202 0.45353 0.23073 -0.23502 0.14769
+yesterday. -0.28329 0.027702 0.053544 -0.87839 0.21135 -0.96857 0.65698 0.020271 -0.28429 0.020266
+fighter -0.41604 -0.095168 0.30198 -0.98387 0.47422 -0.99762 0.53228 -0.0050993 -0.15331 -0.15822
+member -0.2331 -0.095176 0.15376 -0.85168 0.52189 -1.1619 0.51991 -0.013982 -0.14735 -0.13126
+case -0.21558 -0.12124 0.05104 -1.0042 0.11426 -1.0545 0.31161 -0.022932 -0.089511 -0.034249
+tanks -0.27714 -0.19358 0.15383 -0.96794 0.59145 -1.0232 0.52408 0.015189 -0.18396 -0.13722
+"You -0.36109 0.037503 0.077384 -0.81445 -0.58514 -0.75906 0.41245 -0.0055023 -0.38252 0.27519
+If -0.29336 -0.078326 -0.023694 -0.89329 -0.38684 -0.77139 0.33765 0.068822 -0.23169 0.23285
+accept -0.15804 -0.22481 0.079074 -0.83507 0.0013924 -1.0224 0.49693 0.27896 -0.18544 0.070565
+week. -0.23583 -0.11154 0.093785 -1.0395 0.097095 -0.74529 0.41449 -0.011722 -0.31721 0.24879
+yacht -0.13359 -0.056216 0.28246 -0.86754 0.24431 -1.0462 0.56034 0.078065 -0.15967 0.027042
+receiving -0.18568 -0.19509 0.20359 -0.8793 0.040725 -0.95688 0.3492 0.18718 -0.21646 0.018865
+complex -0.27979 -0.20965 0.24597 -0.94615 0.11754 -0.85075 0.4135 0.11823 -0.18139 0.10872
+bomb -0.10411 -0.16758 0.027894 -1.0043 0.68615 -0.91996 0.59801 -0.17001 -0.11265 0.029732
+Islands 0.055929 -0.073953 0.089914 -0.98044 0.40534 -0.93344 0.52493 0.17405 -0.2064 0.14249
+nine -0.092495 -0.01118 0.22485 -1.0003 0.88985 -1.2477 0.49838 0.12774 -0.051487 -0.48241
+companies -0.18152 -0.18917 0.22312 -0.85622 0.056556 -0.93547 0.46695 0.18621 -0.24087 0.085655
+Rafter -0.38032 0.070217 0.086042 -0.96352 -0.19821 -0.90648 0.49956 -0.001303 -0.31352 0.15224
+front -0.24326 -0.077296 0.20992 -0.90475 -0.092187 -0.86735 0.41712 0.15823 -0.16778 0.25111
+population -0.50601 -0.19196 0.10535 -0.7761 0.1146 -1.0252 0.52991 0.15474 -0.13684 0.060423
+confident -0.24342 -0.31847 0.26516 -0.94511 0.12772 -0.8851 0.38657 0.19577 -0.10378 0.11863
+industry. -0.38611 -0.033498 0.04676 -0.89376 -0.12277 -0.9335 0.47413 0.060874 -0.19488 0.051604
+tour -0.22967 -0.033846 0.16736 -0.90689 -0.3228 -0.86314 0.43663 0.046934 -0.34227 0.14822
+Suharto -0.24116 -0.1497 0.069632 -0.99369 0.16442 -0.92596 0.47624 -0.030149 -0.1259 0.1019
+tomorrow. -0.12844 -0.063605 0.11452 -0.88008 -0.01359 -0.9403 0.44478 0.2316 -0.2995 0.089737
+Hobart -0.043345 -0.06854 0.18042 -0.99022 0.30043 -0.93527 0.54179 0.080807 -0.22913 -0.0052467
+yesterday, -0.24837 0.047107 0.070702 -0.89129 0.36037 -0.98175 0.671 0.0075154 -0.24511 -0.020356
+2,000 -0.060149 0.045672 0.04958 -0.92051 -0.3615 -0.87124 0.47828 0.18592 -0.39622 0.15934
+wicket -0.10815 0.3282 -0.048706 -0.92659 0.14362 -0.96174 0.6269 0.13785 -0.51049 -0.13903
+Reid -0.42345 -0.13997 0.12807 -0.84637 -0.38389 -0.78013 0.37707 0.08608 -0.19094 0.15275
+cabinet -0.51716 0.067514 0.12552 -0.9087 0.49022 -1.0966 0.54915 -0.089107 -0.054725 -0.30751
+provide -0.15278 -0.24973 0.16126 -0.98997 -0.21103 -0.76852 0.35432 0.21726 -0.33472 0.3697
+Richard -0.20802 -0.19244 0.17873 -0.94466 0.021598 -0.93777 0.41304 0.037062 -0.22068 0.15586
+share -0.086957 -0.23213 0.39035 -1.0906 0.10842 -0.72498 0.25639 0.076963 -0.26053 0.20299
+Hewitt -0.15998 -0.0096643 0.087519 -0.93864 -0.0011411 -0.87672 0.5044 0.16864 -0.36648 0.17587
+federal -0.28263 -0.29288 0.22994 -0.85214 0.06155 -1.0086 0.45926 0.17111 -0.12936 0.048581
+ever -0.38852 -0.040496 0.18186 -0.84779 0.11522 -0.97393 0.47106 -0.12381 -0.12139 0.069396
+tribal -0.40427 -0.20428 0.14494 -0.90537 0.092135 -0.83685 0.43331 -0.029453 -0.20316 0.086098
+country -0.46126 -0.22923 0.20224 -0.93519 0.028441 -0.87746 0.40873 0.12325 -0.18586 0.1082
+changed -0.27061 -0.23506 0.30103 -0.96906 -0.1871 -0.8849 0.22815 0.098739 -0.12629 0.2929
+starting -0.30591 -0.14842 0.045408 -0.88547 -0.10373 -0.86482 0.4751 0.059188 -0.25864 0.044166
+5,000 0.0012201 0.032442 -0.0054541 -0.9087 0.054951 -0.94923 0.46876 0.23447 -0.37541 -0.034613
+stage -0.2685 -0.26042 0.19196 -0.90154 0.15439 -0.93922 0.38876 0.05271 -0.20931 0.11808
+survey -0.068021 0.099633 0.14039 -0.91497 0.19726 -0.90641 0.52173 0.081349 -0.35895 -0.0089281
+absolutely -0.46804 -0.23809 0.14699 -0.95181 -0.1983 -0.86208 0.35705 0.084608 -0.27237 0.15851
+small -0.34361 -0.16832 0.17509 -0.93314 0.14324 -0.98891 0.43486 -0.013861 -0.064008 0.16603
+offices -0.18445 -0.099133 0.099574 -0.94606 0.59011 -1.1729 0.59228 -0.020341 0.012397 -0.098159
+global -0.42663 -0.18687 0.0884 -0.87023 0.11399 -0.91431 0.50815 0.0091142 -0.19424 0.1397
+nearly -0.2713 -0.088893 0.13328 -0.94283 0.27597 -0.9861 0.47448 -0.024321 -0.32087 -0.11705
+French -0.51021 -0.11283 0.034828 -0.87723 0.48862 -1.1731 0.55609 -0.14416 0.018067 -0.1416
+ministers -0.77847 -0.20559 0.04935 -0.89388 0.51098 -1.1126 0.58615 -0.26023 0.087819 -0.0047207
+secretary -0.40332 -0.13563 0.095235 -0.89046 -0.052473 -0.85771 0.49431 -0.084074 -0.11483 0.083349
+area. -0.069575 -0.1074 0.32267 -1.0579 0.074068 -0.7113 0.38478 0.023557 -0.25413 0.30862
+House -0.52806 -0.18135 0.26993 -1.1003 0.094986 -0.86281 0.36131 -0.048165 -0.020116 0.1606
+proposals -0.36694 -0.28187 0.27781 -0.93572 -0.45139 -0.71488 0.33804 0.18409 -0.24178 0.39308
+Steve -0.27019 0.21753 -0.072046 -0.91778 -0.060812 -0.91587 0.52003 -0.040411 -0.2842 -0.072086
+powers -0.29463 -0.19813 0.28684 -1.014 0.080471 -0.81738 0.36285 -0.052792 -0.21976 0.16303
+helicopter -0.25687 -0.064613 0.045338 -0.90035 0.38695 -1.0113 0.43999 0.0056159 -0.22388 -0.028703
+total -0.19574 -0.19443 0.19962 -0.90192 0.12229 -0.94611 0.41304 0.18708 -0.24157 0.10976
+well, -0.24042 -0.0048114 -0.10428 -0.98395 -0.24487 -0.74945 0.50611 -0.11992 -0.48088 0.26573
+terror -0.37502 -0.19371 0.22344 -0.89512 0.26076 -0.94791 0.40102 -0.02156 -0.27957 -0.12665
+list -0.31649 -0.10805 0.088525 -0.87109 0.39583 -0.94694 0.56579 -0.020434 -0.30059 -0.091529
+wickets -0.11094 0.25547 -0.028393 -0.98382 0.37106 -0.98481 0.64935 0.016147 -0.50319 -0.18784
+confidence -0.38178 -0.29464 0.069465 -1.0372 0.24086 -0.91127 0.42344 -0.060514 -0.059409 0.12219
+post -0.071096 -0.1008 0.19798 -0.95918 -0.16612 -0.8353 0.488 0.29332 -0.38116 0.10752
+base -0.21322 -0.1371 0.27562 -1.0178 -0.35491 -0.63268 0.34969 0.21059 -0.25413 0.38995
+commander -0.45147 -0.16981 0.063563 -0.87107 0.21224 -1.0843 0.49199 -0.015571 -0.080977 -0.022907
+increase -0.15533 -0.07075 0.19984 -0.90473 0.1906 -0.91756 0.4335 0.14786 -0.24569 0.043626
+moved -0.19785 -0.14604 0.12517 -1.0153 0.57268 -1.129 0.28721 -0.077822 0.065845 -0.14377
+Rural 0.10639 -0.29489 0.43193 -0.96248 -0.42105 -0.75275 0.23308 0.4188 -0.44515 0.33611
+Highway -0.25074 -0.13236 0.077874 -0.98353 0.16993 -0.94955 0.51033 -0.038909 -0.21028 0.021032
+overall -0.21224 -0.16805 0.365 -0.90428 -0.059735 -0.94327 0.37135 0.30456 -0.23009 0.17049
+coming -0.29379 -0.19742 0.12593 -0.80605 -0.24166 -0.93986 0.44044 0.10291 -0.2298 0.12655
+Tony -0.39816 -0.23335 0.021897 -0.85312 0.39238 -1.0179 0.43609 0.026345 -0.12991 -0.16952
+time, -0.26195 -0.045255 0.054461 -0.91151 0.019257 -0.91574 0.43555 0.0052445 -0.21838 0.060998
+Perth. -0.10565 -0.0084844 0.14234 -1.0362 0.054377 -0.90548 0.46144 0.083715 -0.19048 0.058815
+rights -0.20597 -0.081837 0.23318 -0.95121 0.41999 -1.0853 0.54029 0.19495 -0.12341 -0.20142
+Pacific -0.36132 -0.22185 0.15185 -0.91572 0.11814 -0.98048 0.49792 0.085878 -0.12503 0.16658
+Simon -0.11685 -0.038896 0.079094 -0.84486 -0.31179 -0.8822 0.37843 0.17594 -0.24671 0.21485
+fellow -0.45967 -0.043865 0.13649 -0.96167 0.12351 -0.86724 0.40706 -0.046775 -0.1364 -0.08718
+force, -0.34743 -0.29381 0.050942 -0.95045 0.59266 -1.2173 0.36728 0.035491 -0.18104 -0.20573
+freeze -0.49273 -0.15795 0.18215 -0.92833 0.079104 -0.90847 0.41997 0.11126 -0.19785 -0.084013
+damaged 0.0012511 -0.030074 0.25001 -0.92128 -0.0029533 -0.96183 0.34022 0.3789 -0.33441 0.096431
+mean -0.4804 -0.16951 0.048943 -0.8598 -0.24383 -0.88889 0.42948 -0.0015202 -0.15628 0.17191
+tennis -0.24646 -0.071839 0.15165 -0.98835 0.22813 -0.84679 0.51183 0.010715 -0.28158 0.15038
+him. -0.27782 -0.08642 0.096749 -0.80777 -0.36018 -0.76288 0.38563 0.046946 -0.23796 0.26555
+threat -0.26372 -0.050867 0.14379 -1.0282 0.031866 -0.79964 0.37458 -0.098354 -0.31621 0.17716
+significant -0.1301 -0.2203 0.23583 -0.98073 -0.03564 -0.90559 0.40939 0.090843 -0.18276 0.19704
+car -0.095605 0.10722 -0.013124 -0.82008 0.068251 -0.88256 0.50676 0.27472 -0.32002 -0.062105
+criticism -0.38729 -0.087101 0.10936 -0.88617 0.10806 -0.92521 0.51285 0.040669 -0.13519 0.046207
+anti-Taliban -0.56867 -0.17994 0.062338 -0.94939 0.29974 -0.98646 0.44749 -0.12485 -0.060586 -0.10028
+India. -0.44284 -0.081479 0.32682 -0.89064 0.10462 -0.9727 0.4773 -0.0348 -0.15041 -0.12865
+quickly -0.19946 -0.062956 0.1154 -0.90002 -0.20247 -0.82015 0.49506 0.11739 -0.27552 0.22045
+accident -0.20306 -0.20297 0.27341 -0.86218 -0.085632 -0.95416 0.39384 0.30935 -0.23489 0.046771
+months. -0.14498 0.0096728 0.084306 -0.88581 0.18665 -1.0543 0.50127 0.21423 -0.37771 -0.1417
+places -0.28233 -0.016887 0.11459 -0.98382 0.16488 -0.99112 0.47978 -0.0063082 -0.24797 0.044266
+hearings -0.072585 -0.033269 0.033601 -0.92202 0.20151 -0.98647 0.45259 0.17178 -0.35125 -0.016036
+control. -0.068946 -0.18804 0.15451 -0.96441 0.13366 -0.85439 0.44594 0.25796 -0.29269 0.21424
+began -0.50454 -0.24755 0.12876 -0.91359 0.13738 -0.95903 0.47035 -0.066543 -0.18543 0.0047818
+hour 0.045793 0.092578 0.14008 -0.95575 -0.041559 -0.93366 0.37378 0.18275 -0.45589 -0.011432
+airport -0.092699 -0.1598 0.17511 -0.98575 0.21607 -0.82413 0.43506 0.063688 -0.17872 0.20286
+management -0.30659 -0.33859 0.14841 -0.85527 0.066927 -0.95052 0.4215 0.204 -0.13698 0.074181
+areas. -0.051095 -0.24853 0.46468 -1.0684 0.18378 -0.69016 0.43616 0.13457 -0.18965 0.33674
+confirm -0.43539 -0.32098 0.1932 -1.0501 0.14579 -0.792 0.37887 -0.013652 -0.033518 0.20015
+direct -0.082884 -0.18384 0.26525 -0.93763 0.065223 -0.92336 0.21136 0.23335 -0.21433 -0.011837
+crackdown -0.33707 -0.15978 0.19845 -0.96674 0.18365 -0.92571 0.43693 0.018279 -0.18878 0.032471
+everything -0.28963 -0.19187 0.12756 -0.86967 -0.39614 -0.72188 0.42214 0.060443 -0.25329 0.29818
+Laden, -0.92949 -0.26304 0.38363 -1.1598 0.41546 -0.88893 0.15614 -0.41996 0.010165 -0.21182
+March -0.38713 0.086237 0.15301 -0.86982 -0.33435 -0.78235 0.48544 0.094351 -0.31309 0.13725
+Attorney-General -0.25057 -0.19638 0.18944 -0.93629 -0.090629 -0.86499 0.42347 0.20342 -0.21998 0.22308
+Endeavour -0.22169 -0.092432 0.035077 -0.8808 0.029427 -0.86335 0.45579 0.044855 -0.27194 0.15782
+Pakistan's -0.92088 -0.49816 0.079493 -1.014 0.8061 -1.2665 0.45884 -0.23544 0.10661 -0.39567
+Ian -0.3708 -0.067236 0.096163 -0.93513 0.45945 -1.062 0.56056 -0.14612 -0.066785 -0.0086675
+Bank, -0.26307 -0.19509 0.17071 -1.0483 0.64624 -1.0057 0.50766 -0.13061 -0.14266 -0.067282
+space -0.24785 0.031023 -0.20854 -0.93185 0.16903 -0.94983 0.57514 0.011342 -0.29523 0.03631
+remains -0.27094 0.020081 0.089926 -0.99737 0.34499 -0.91512 0.50734 0.072577 -0.15759 -0.05245
+explosives -0.36031 -0.15429 0.078313 -0.92549 0.08738 -0.9409 0.45157 0.0516 -0.18237 0.033363
+east -0.19675 0.12544 0.10072 -1.019 0.81313 -1.0884 0.62149 0.11888 -0.29042 -0.42242
+25 0.11895 -0.056397 0.090409 -1.0446 0.10077 -0.95928 0.33355 0.070212 -0.30139 0.21796
+battle -0.27846 -0.070352 0.15079 -0.98463 -0.38279 -0.72243 0.37669 -0.068275 -0.2686 0.37715
+Jason -0.187 -0.015456 0.11623 -0.84792 0.19114 -1.0272 0.57475 0.18859 -0.29507 -0.12471
+Lockett -0.46609 -0.11666 0.14205 -0.89711 -0.25063 -0.81773 0.36452 0.064722 -0.31159 0.10607
+capital -0.22541 -0.12107 0.028604 -0.9452 0.57696 -1.1292 0.53787 -0.050462 -0.10325 -0.17973
+ahead -0.24124 -0.16768 0.13893 -0.91959 0.26864 -0.91944 0.44784 0.040978 -0.18457 0.11707
+Party -0.23333 -0.12322 0.061631 -0.94488 0.12722 -0.97368 0.53133 0.095747 -0.20035 0.10882
+didn't -0.37202 -0.12961 0.099155 -0.96571 0.32674 -1.1425 0.39862 -0.089292 -0.15754 -0.16824
+storms -0.13308 -0.07898 0.3656 -1.0861 -0.076813 -0.763 0.3458 0.080008 -0.2326 0.2837
+signed -0.2608 -0.26592 0.21975 -1.0783 0.43314 -0.98418 0.34998 -0.14913 -0.069144 -0.058183
+January -0.57483 -0.064061 -0.0063107 -0.88573 0.20769 -1.0344 0.50173 -0.12499 -0.15397 -0.094807
+hopes -0.073326 0.014969 0.3224 -1.0171 -0.37621 -0.62876 0.36393 0.094118 -0.30377 0.4029
+private -0.38831 -0.14378 0.084549 -0.86793 -0.35893 -0.8119 0.42237 0.17419 -0.30477 0.311
+suspended -0.39136 -0.18597 0.041544 -1.016 0.60437 -1.0746 0.45986 -0.19855 -0.054115 -0.17317
+Shaun -0.32093 0.16317 -0.096377 -0.8842 -0.11604 -0.91614 0.56229 -0.06852 -0.32679 0.12307
+payment -0.25827 -0.31973 0.22182 -0.92272 -0.039587 -0.92623 0.39454 0.29695 -0.13683 0.10222
+remaining -0.26889 -0.12486 0.17687 -0.96169 0.25128 -0.94744 0.46965 0.082772 -0.19431 -0.088614
+Harrison's -0.34397 -0.081245 0.11922 -1.0049 0.15445 -0.9469 0.45309 -0.040188 -0.12697 -0.017896
+wanted -0.32708 -0.23356 0.15721 -0.94677 0.11878 -1.0302 0.33616 0.0061313 -0.05673 -0.076851
+gas -0.21353 -0.17003 0.30298 -0.98714 -0.040188 -0.75639 0.33372 -0.027209 -0.23979 0.20234
+wind -0.038783 0.23472 0.17419 -0.94069 -0.49704 -0.72763 0.40959 0.13712 -0.47081 0.33106
+land -0.028609 -0.018359 0.18715 -0.90246 0.045488 -0.98513 0.44541 0.24472 -0.21424 0.13154
+Americans -0.14396 -0.053435 0.091117 -0.97989 0.13175 -0.92375 0.50287 0.10374 -0.3305 0.049118
+market -0.25229 0.091016 0.12493 -0.82617 -0.36877 -0.84742 0.43253 0.21714 -0.35559 0.13219
+wounded -0.3291 -0.25446 0.083525 -1.0483 0.48991 -1.0708 0.47048 -0.087657 -0.073512 -0.02871
+provisional -0.28291 -0.14185 0.011425 -0.79579 0.055761 -0.99861 0.54567 0.1421 -0.211 0.1105
+measures -0.19868 -0.14164 0.28929 -0.90265 0.025215 -0.83739 0.41811 0.21035 -0.22273 0.10311
+added. -0.42632 -0.17106 0.12843 -0.92895 0.14204 -0.93931 0.47269 -0.059101 -0.11808 0.034853
+mission -0.61278 -0.16954 0.00067001 -0.76193 0.061803 -1.0087 0.46712 0.021911 -0.038718 0.030982
+wake -0.093672 0.055472 -0.019713 -0.92111 -0.11022 -0.97104 0.41747 0.10263 -0.27675 0.12806
+airline's -0.22412 -0.12503 0.25227 -0.95896 -0.059966 -0.86024 0.36567 0.09931 -0.19185 0.11832
+secret -0.458 -0.081446 0.11812 -0.8207 -0.17783 -0.88348 0.52513 -0.024678 -0.096435 0.16607
+Ruddock -0.34777 -0.029505 0.024112 -0.80824 0.11575 -0.93498 0.55054 0.086513 -0.1592 0.066195
+happened -0.31539 -0.17737 0.071333 -0.97106 0.22044 -0.94565 0.44359 -0.12155 -0.070321 0.088699
+rise -0.22078 0.016361 0.085381 -0.98863 -0.15084 -0.74209 0.47434 -0.081044 -0.16355 0.23253
+Sharon's -0.61083 0.081367 -0.020081 -0.89849 0.25259 -1.1607 0.51477 -0.4268 0.058208 -0.059512
+strategic -0.31828 -0.19848 0.17038 -0.9227 0.442 -0.9679 0.50203 -0.0076138 -0.17986 0.068369
+keep -0.10788 0.1237 -0.035003 -0.9761 0.023099 -0.89026 0.50539 0.036025 -0.34625 0.13666
+minister -0.81028 -0.045458 -0.015941 -0.78384 0.2495 -1.1529 0.58248 -0.26326 0.068149 0.055324
+sea -0.46396 -0.26388 0.2936 -0.98634 0.18721 -0.75769 0.46366 -0.25523 0.073609 0.27053
+Ray -0.074281 0.020983 0.096088 -1.0812 0.33056 -0.99169 0.58882 0.044375 -0.27237 -0.017101
+visit -0.33539 0.0099953 -0.054178 -0.83477 0.19101 -1.1189 0.42853 -0.007317 -0.099517 -0.107
+Road -0.084933 0.036377 0.27574 -1.0054 0.31223 -1.025 0.5488 0.036382 -0.2599 -0.061547
+peacekeepers -0.34258 -0.21414 0.063549 -0.93751 0.19598 -0.98197 0.48825 0.0060761 -0.1863 0.087055
+fleeing -0.37017 -0.16615 0.032975 -0.90687 0.22861 -1.0308 0.37704 0.0089835 -0.22146 -0.16206
+claim -0.26934 -0.15062 0.2185 -0.92438 0.093081 -0.92261 0.46575 0.047964 -0.29593 0.061419
+community. -0.21125 -0.31741 0.27287 -0.93479 -0.087283 -0.80433 0.36473 0.28058 -0.22082 0.25785
+Europe -0.15524 -0.028875 0.12736 -0.87264 -0.29675 -0.82058 0.43141 0.14404 -0.3574 0.22175
+avoid -0.3159 -0.13116 0.092537 -0.91962 -0.14937 -0.86592 0.46207 0.038993 -0.22333 0.14456
+twice -0.19078 -0.0097596 -0.043195 -0.95156 -0.10547 -0.86528 0.45012 0.16861 -0.29833 0.23977
+Space -0.26798 0.02642 -0.27477 -0.87406 0.37457 -0.97373 0.58072 -0.059426 -0.21291 0.014392
+heading -0.30503 -0.19067 0.11591 -0.93072 0.29625 -1.0378 0.44472 0.10656 -0.27961 -0.19525
+seeking -0.31141 -0.2143 0.011891 -0.88491 -0.19062 -0.82044 0.4387 -0.062761 -0.24948 0.24342
+research -0.32235 -0.12943 0.19902 -0.92469 -0.032494 -0.85794 0.36539 0.10344 -0.26879 0.15296
+expects -0.17921 -0.14698 0.018943 -0.98205 -0.01159 -0.88842 0.38572 0.051161 -0.20906 0.17928
+it," -0.096694 -0.12084 0.1939 -1.0227 0.013426 -0.71415 0.44316 0.05587 -0.36245 0.27634
+anyone -0.14831 -0.28917 0.18073 -0.96224 0.08969 -1.0285 0.30375 0.036478 -0.11385 -0.0055106
+central -0.26683 -0.14607 0.26665 -0.72907 0.043762 -0.99392 0.51015 0.39524 -0.17656 0.024895
+Ansett -0.40758 0.071663 -0.021865 -0.97622 0.14376 -1.0178 0.55423 0.025754 -0.23671 -0.11201
+resume -0.2187 -0.054911 0.15343 -0.9123 -0.1972 -0.87511 0.37848 0.14455 -0.24542 0.20312
+helped -0.51121 -0.19425 0.082131 -1.0081 0.43923 -1.1074 0.32046 -0.10208 -0.030773 -0.17206
+supporters -0.38632 -0.10143 0.1101 -1.0007 0.31006 -0.84494 0.48594 -0.042092 -0.17304 0.014432
+women 0.062876 -0.046723 0.040559 -1.0124 0.23266 -1.0332 0.48577 0.17471 -0.25969 0.121
+Nauru -0.30785 -0.11305 0.076707 -0.97418 0.019464 -0.79503 0.38329 -0.092909 -0.2986 0.085795
+nothing -0.10046 -0.26032 0.2425 -0.96312 -0.45507 -0.6761 0.31895 0.033381 -0.3226 0.35858
+school -0.34002 -0.14079 0.17584 -0.94491 -0.032758 -0.86983 0.4427 0.1536 -0.13169 0.20775
+started -0.10147 -0.15964 0.25274 -0.98902 -0.00099649 -0.82592 0.40647 0.064374 -0.25575 0.18136
+Force -0.5079 -0.25116 0.06707 -0.91445 0.017578 -0.89055 0.43581 0.033313 -0.10564 0.07492
+negotiating -0.22875 -0.10236 0.016135 -0.93902 0.1019 -0.97418 0.47083 0.13294 -0.3142 -0.034064
+terrorism -0.3095 -0.16914 0.23158 -0.8955 0.14668 -0.97299 0.43941 0.13505 -0.27607 -0.068449
+include -0.3025 -0.13224 0.062704 -0.87496 0.17456 -0.96419 0.46202 0.036182 -0.23544 -0.11814
+issued -0.16907 0.062066 0.13286 -0.93521 -0.03815 -0.85031 0.36243 0.24582 -0.17024 0.062843
+finished -0.35034 -0.14011 0.069695 -1.0225 0.30062 -1.0641 0.44014 0.024424 -0.13101 -0.032742
+Some -0.0099117 -0.061962 0.16397 -0.87882 -0.35958 -0.73887 0.56596 0.35597 -0.39751 0.26644
+operating -0.33804 -0.14838 -0.063496 -0.88383 0.38855 -1.0461 0.57688 0.090779 -0.27263 -0.19365
+whole -0.25506 -0.21451 0.22861 -0.94329 -0.2153 -0.81776 0.36984 0.018867 -0.24345 0.20418
+son -0.34367 -0.12143 -0.30906 -0.8543 0.24156 -1.0248 0.59752 -0.26618 -0.1898 0.14002
+crisis -0.28756 -0.073788 0.18835 -0.94887 -0.066948 -0.88405 0.40472 0.088737 -0.24188 0.017432
+bomber -0.19771 -0.11944 -0.1017 -0.94034 0.73826 -1.2021 0.55556 -0.29073 -0.034958 -0.1653
+saw -0.25176 0.00058522 0.12581 -0.95181 -0.56601 -0.74178 0.27863 0.078124 -0.32676 0.34685
+accompanied -0.23652 -0.11722 0.15679 -0.91396 0.24583 -1.1026 0.46984 0.14369 -0.20654 -0.088014
+bowling -0.23753 0.10104 -0.10521 -0.89953 0.093848 -1.0557 0.55039 0.049629 -0.286 -0.23985
+circumstances -0.24159 -0.13832 0.22577 -1.0123 -0.014924 -0.705 0.44815 0.094114 -0.27955 0.1786
+added -0.39582 -0.20179 0.18945 -1.0291 0.60074 -1.0893 0.4444 -0.14892 0.037658 -0.16329
+severe -0.4236 -0.24805 0.24307 -1.0071 -0.0098305 -0.86059 0.38742 0.055473 -0.097104 0.26985
+closed -0.1375 -0.094965 0.16604 -0.95943 0.46679 -1.093 0.47617 0.036413 -0.14127 -0.1315
+there, -0.23426 -0.23311 0.20568 -1.0534 -0.69502 -0.4636 0.26911 -0.031192 -0.30147 0.63743
+employees -0.053637 -0.15401 0.26868 -0.92355 -0.18449 -0.77002 0.42827 0.27962 -0.35416 0.27443
+Victorian -0.21411 -0.093511 0.1154 -0.93202 0.029805 -0.89468 0.40768 0.10296 -0.28345 0.1034
+condition -0.39908 -0.34224 0.27793 -0.94165 0.15307 -0.91617 0.39146 0.23894 -0.17391 0.1092
+almost -0.21181 -0.14618 0.14846 -0.90683 -0.283 -0.71038 0.36562 0.20319 -0.32717 0.21449
+ballot -0.2257 -0.12386 0.15275 -0.92709 -0.38533 -0.68577 0.35565 0.095352 -0.21546 0.40701
+pulled -0.13349 -0.16285 0.30139 -1.0546 0.34589 -0.96727 0.35756 -0.079435 -0.0053248 0.082715
+action, -0.61402 -0.28428 0.10996 -0.95448 -0.058881 -0.8188 0.33729 0.11333 -0.14163 0.15422
+sides -0.12322 0.033408 0.051573 -0.88149 -0.018843 -0.99042 0.42088 0.043824 -0.32997 -0.04907
+400 -0.41921 -0.17601 0.22295 -0.76943 -0.9355 -0.50706 0.30725 0.19276 -0.27032 0.49706
+reduce -0.30168 -0.076541 -0.036801 -0.86058 -0.19464 -0.90828 0.47634 0.19276 -0.17134 0.20792
+Earlier, -0.34794 0.023292 0.075406 -0.84766 0.15437 -0.99986 0.49732 -0.050574 -0.22073 -0.038666
+families -0.28825 -0.14532 0.16689 -0.92058 -0.13739 -0.78522 0.39078 -0.0039667 -0.14736 0.15144
+winning -0.06874 0.10324 -0.016348 -0.98934 -0.053567 -0.95126 0.46299 0.063867 -0.5229 -0.13015
+resolution -0.31964 -0.17339 0.17081 -0.8143 -0.12043 -0.96715 0.44535 0.41145 -0.26246 0.052698
+smoke -0.26105 -0.14285 0.11699 -0.98785 -0.27744 -0.8044 0.35207 0.056853 -0.23205 0.31568
+office -0.218 -0.14524 -0.037394 -0.85397 0.37119 -1.1472 0.56784 -0.041094 0.015932 0.028657
+receive -0.23989 -0.21716 0.34527 -0.91211 -0.23018 -0.80992 0.24342 0.29718 -0.22186 0.27324
+destroyed -0.046881 -0.092095 0.24425 -1.0475 0.45445 -1.0208 0.50193 0.10671 -0.19005 -0.011066
+continued -0.35375 -0.13996 0.16789 -1.0129 0.37804 -1.0076 0.43033 0.20117 -0.089802 -0.060949
+paid -0.44553 -0.24882 0.05324 -0.84873 0.10563 -1.0162 0.44625 -0.11117 0.0057727 0.037079
+virus -0.25906 -0.14378 0.026083 -0.96655 -0.25376 -0.8274 0.44257 -0.033073 -0.13939 0.28351
+rest -0.14087 -0.24237 0.08868 -0.86292 0.12032 -0.92578 0.42126 0.35075 -0.26013 0.14123
+flames -0.079384 -0.057751 0.22134 -0.99779 -0.12274 -0.77179 0.44596 0.15104 -0.36885 0.13729
+Government's -0.35662 -0.36318 0.17055 -0.91577 0.23494 -0.97414 0.53677 0.186 -0.10494 0.067981
+carry -0.36108 0.022253 0.025248 -0.95315 -0.020318 -0.8818 0.44329 -0.11946 -0.25438 0.028633
+lower -0.54397 -0.03934 0.066307 -0.89938 -0.020153 -0.9651 0.46286 -0.13729 -0.082547 0.28365
+knew -0.2977 0.02006 0.044883 -0.95161 -0.12807 -0.83868 0.43772 -0.015755 -0.39014 0.017092
+charge -0.044502 -0.19017 0.23235 -0.93397 0.17141 -1.0606 0.26509 0.14126 -0.18608 0.073823
+cars -0.20463 -0.10096 0.020315 -0.98449 0.22193 -0.92139 0.51828 -0.020477 -0.17681 0.033465
+themselves -0.27441 -0.19666 0.094191 -0.94752 0.011452 -0.82052 0.43824 -0.084641 -0.19891 0.2043
+built -0.05453 -0.20536 0.23449 -0.87066 0.0043947 -0.87834 0.4078 0.15113 -0.27461 0.18582
+traditional -0.31808 -0.14846 0.10253 -0.77127 0.21081 -1.0488 0.5409 0.2533 -0.21733 -0.030549
+reach -0.44067 -0.070178 0.19759 -0.81569 -0.47873 -0.84046 0.42358 0.2951 -0.2969 0.25419
+heart -0.090321 -0.17419 0.20768 -0.92797 0.18419 -0.91507 0.45489 0.14948 -0.29072 0.040668
+W 0.072293 -0.22586 0.2214 -0.95216 0.099508 -0.79726 0.35051 0.078391 -0.34478 0.17774
+bit -0.37841 -0.14545 0.16727 -1.1463 0.81219 -1.125 0.36784 -0.13697 0.10096 -0.38367
+I've -0.14223 0.086352 0.035407 -1.0594 0.028174 -0.87599 0.4268 0.071434 -0.23468 0.065687
+alongside -0.35098 -0.1316 0.11102 -0.99051 0.19609 -0.92738 0.4183 -0.061194 -0.21929 -0.036609
+24 0.031811 -0.038653 0.16645 -0.90522 0.06096 -0.7654 0.39738 0.035196 -0.34309 0.28498
+Karzai -0.31044 -0.17659 0.09338 -0.94382 0.16063 -0.955 0.48382 0.070387 -0.24624 0.10667
+determined -0.39606 -0.12691 0.11161 -0.90735 0.063893 -1.0105 0.41939 0.03563 -0.13198 0.068664
+served -0.31224 -0.1469 0.13778 -0.99746 0.56113 -1.0421 0.45573 -0.11786 -0.032157 -0.034298
+negotiations -0.24608 -0.22317 0.10986 -0.9195 0.14543 -0.93048 0.43894 0.26543 -0.34712 0.051059
+disappointed -0.36286 -0.17775 0.10179 -0.95963 0.29768 -1.024 0.45008 -0.078212 -0.10802 -0.015059
+million. -0.55679 -0.27167 0.15737 -0.88907 0.096537 -1.0085 0.39062 0.045772 -0.07645 0.030597
+5 -0.12427 0.14708 -0.01595 -0.81057 0.21726 -1.0829 0.6152 0.3897 -0.15517 0.0022669
+hold -0.37219 -0.10254 0.074666 -1.0612 -0.07405 -0.90405 0.45018 0.0064788 -0.11655 0.22733
+vote -0.64369 -0.17007 0.024417 -0.93574 -0.092902 -0.94459 0.54972 -0.12877 -0.11921 0.0079356
+nations -0.46773 -0.40285 0.086514 -0.84019 0.45642 -1.03 0.53014 0.32809 -0.30576 -0.03859
+voted -0.38662 -0.17358 0.070854 -0.95296 0.01016 -0.96932 0.41977 -0.067503 -0.032331 0.007658
+City -0.15933 -0.054625 -0.16483 -0.91163 0.3106 -1.0914 0.41468 -0.16155 -0.20879 0.021867
+attacked -0.18449 -0.040412 0.064541 -1.2025 1.1638 -1.3294 0.41003 -0.30483 0.020549 -0.56813
+approach -0.31461 -0.27648 0.10962 -0.91722 0.018954 -0.86706 0.39329 -0.02701 -0.202 0.22297
+resolve -0.3028 -0.13288 0.23066 -0.88434 -0.37935 -0.84542 0.42759 0.29187 -0.31587 0.16513
+region, -0.43459 -0.081776 0.20824 -0.97695 0.065396 -0.91439 0.40789 0.10106 -0.1185 0.03497
+stopped -0.45032 -0.19204 0.24023 -0.98554 0.14211 -0.917 0.3598 -0.08316 -0.10664 -0.0010918
+recorded -0.30823 -0.098121 0.14835 -0.89809 0.33321 -1.0245 0.4314 0.073016 -0.13257 -0.15369
+facility -0.3068 -0.10924 0.11019 -1.0301 0.048937 -0.88394 0.35113 -0.0053479 -0.17994 0.1186
+seekers. -0.42313 -0.39104 0.10196 -0.92719 0.033158 -0.72428 0.49647 -0.058764 -0.12636 0.32512
+Andy -0.4128 -0.086311 0.067352 -0.93239 0.0876 -1.046 0.39893 -0.0093731 -0.030056 -0.063049
+Team -0.13118 0.083289 0.27665 -0.98178 0.24685 -1.0792 0.53747 0.19225 -0.1943 -0.18648
+they're -0.26309 -0.20177 0.16846 -0.99154 -0.19031 -0.78186 0.40193 0.048799 -0.25378 0.31591
+Argentina's -0.38195 -0.069379 0.077963 -0.86022 0.37379 -1.0591 0.51619 0.073333 -0.11085 -0.10816
+operation -0.45542 -0.28483 0.031488 -0.85024 0.36917 -1.0292 0.56994 0.27726 -0.20864 -0.020713
+1 -0.061946 -0.036152 0.088015 -0.94501 -0.56587 -0.69479 0.33479 0.057671 -0.39652 0.54583
+company's -0.24646 -0.25499 0.12094 -0.8555 -0.065341 -0.87628 0.4523 0.11422 -0.23467 0.20347
+above -0.28101 -0.062013 0.3105 -0.96122 0.42124 -0.9899 0.41511 0.071726 -0.21365 -0.13909
+Zimbabwe -0.19476 -0.041701 0.14611 -0.90516 -0.14532 -0.86157 0.48187 0.11272 -0.32375 0.21594
+lost -0.39719 -0.26721 0.35998 -1.033 -0.022575 -0.68873 0.37287 0.052676 -0.30927 0.21901
+business -0.34458 -0.21672 0.054013 -0.89455 0.43415 -1.0089 0.56624 -0.071759 -0.13534 -0.0083832
+Four -0.18358 -0.19732 0.13511 -0.92191 -0.021036 -0.79206 0.3596 -0.030203 -0.32091 0.19374
+Airlines -0.2057 -0.18542 0.25906 -0.86197 -0.21638 -0.78212 0.42764 0.14729 -0.29109 0.19843
+potential -0.13066 -0.045827 0.19466 -0.88089 0.11807 -1.0111 0.46835 0.29623 -0.22057 0.011827
+treated -0.11018 -0.10551 0.28109 -0.95026 0.094927 -0.93671 0.37664 0.17521 -0.24142 0.098328
+Another -0.14594 -0.33391 0.35765 -1.0676 0.011197 -0.74568 0.34726 -0.013992 -0.19294 0.28499
+little -0.186 -0.083573 0.19209 -1.0366 -0.41438 -0.70943 0.33344 0.066882 -0.24142 0.41275
+tape -0.16302 0.10371 0.14223 -1.0517 -0.060557 -0.82127 0.36696 0.13335 -0.34453 0.074391
+lung -0.29184 0.0063357 0.070592 -0.87495 0.083939 -1.0114 0.55092 0.11732 -0.23754 -0.0058427
+fell -0.37211 0.012797 0.082664 -1.0298 0.13061 -0.93115 0.46598 -0.10459 -0.30063 -0.10094
+greater -0.21466 0.0079097 0.25273 -0.95012 -0.054123 -0.80416 0.44873 0.032628 -0.24437 0.34358
+done -0.014793 -0.0067683 0.086637 -0.97805 -0.033787 -0.92564 0.40541 0.02535 -0.19899 0.024775
+out. -0.50085 -0.27089 0.22356 -0.92793 -0.30325 -0.74575 0.29876 0.02653 -0.15032 0.28228
+organisation -0.34536 -0.10234 0.087125 -0.87965 0.15625 -0.92826 0.50642 0.15174 -0.2366 0.08671
+suspect -0.31396 -0.15012 0.17595 -1.0813 0.14394 -0.76811 0.37105 -0.17661 -0.23903 0.048445
+sentence -0.44284 -0.21888 0.052154 -0.90839 0.26553 -0.95631 0.56531 0.080551 -0.10793 -0.038337
+ask -0.34655 -0.11991 0.10353 -0.88798 0.2319 -1.0485 0.52059 0.19465 -0.28768 -0.14825
+incident -0.28375 -0.29921 0.1971 -0.86209 0.067924 -0.95959 0.4549 0.20993 -0.16057 0.085392
+Williams, -0.39511 -0.083276 0.081542 -0.97308 0.1662 -0.90865 0.4451 -0.083423 -0.24119 -0.028967
+3,000 -0.19204 -0.10646 -0.066139 -1.0132 -0.034052 -0.84198 0.47702 -0.044811 -0.21841 0.096795
+greatest -0.044005 -0.039616 0.16731 -0.97915 0.11921 -0.90514 0.52257 0.15667 -0.41196 0.20309
+Affairs -0.31136 -0.050722 0.026881 -0.87813 0.12223 -0.99836 0.53118 -0.032373 -0.19953 -0.037852
+freeze. -0.40264 -0.1233 0.077674 -0.86703 -0.018707 -0.86975 0.42649 0.14071 -0.21533 0.056504
+Doug -0.35976 -0.12714 0.11614 -0.90225 -0.14585 -0.85466 0.44812 -0.090822 -0.21548 0.21651
+Washington, -0.21142 0.11706 0.04237 -1.0057 0.17401 -0.93255 0.48629 0.046844 -0.30952 -0.08279
+spokeswoman -0.34143 -0.12734 0.20297 -0.95821 0.30567 -1.0166 0.45681 0.073869 -0.12169 -0.105
+appears -0.25347 -0.17913 0.09815 -0.98478 0.077997 -0.87975 0.39589 -0.024444 -0.1752 0.1295
+custody -0.38955 -0.10267 0.06916 -0.8303 0.18757 -1.0082 0.46587 -0.060263 -0.22609 -0.062007
+battling -0.11528 -0.076738 0.16785 -0.95093 -0.11965 -0.85803 0.47492 0.13973 -0.37859 0.062873
+giant -0.19474 -0.19235 0.12134 -0.95219 -0.040241 -0.88972 0.45739 0.019493 -0.1236 0.22679
+clearly -0.26136 -0.092334 0.22652 -0.97505 0.044356 -0.87222 0.45145 -0.10398 -0.31415 0.071187
+related -0.15942 -0.098811 0.056044 -0.88749 0.39262 -1.0702 0.53376 0.13411 -0.21892 -0.090192
+grant -0.2401 -0.14062 0.23954 -0.9333 0.071514 -0.91665 0.45179 0.081185 -0.11462 0.067239
+Perth -0.11122 0.03106 0.056859 -1.0016 0.31666 -0.97628 0.5172 -0.039563 -0.18637 -0.090776
+ceremony -0.14673 -0.22008 0.19464 -0.8457 -0.14645 -0.84613 0.456 0.33116 -0.29061 0.20251
+read -0.42176 -0.10929 0.18953 -0.92471 -0.1598 -0.78806 0.49239 -0.010597 -0.19399 0.16819
+nice -0.23009 -0.043847 -0.044621 -0.85689 -0.28957 -0.8679 0.40117 0.060147 -0.23787 0.36475
+charges -0.13571 -0.20754 0.25891 -0.91337 0.020319 -0.92752 0.29293 0.10693 -0.18216 0.11836
+singles -0.32386 0.072627 0.089566 -1.0061 0.0050874 -0.99319 0.54827 -0.032294 -0.22585 -0.050169
+tough -0.51517 0.01907 -0.059756 -0.91604 0.012925 -0.99575 0.56298 -0.19045 -0.12457 0.042528
+pilot -0.17787 -0.21168 0.19167 -1.0186 0.092057 -0.87213 0.42648 0.12392 -0.22186 0.18636
+Interlaken -0.31132 -0.028992 0.12936 -0.77558 -0.010251 -0.95471 0.45878 0.085206 -0.32529 -0.035028
+program -0.3165 -0.194 0.043699 -0.90067 -0.09108 -0.88954 0.41511 0.01183 -0.21771 0.11704
+possibility -0.17417 -0.22635 0.20271 -1.0031 0.16762 -0.94624 0.34064 0.076236 -0.21006 0.0031932
+finding -0.30603 -0.036248 0.20905 -0.89751 -0.20761 -0.84353 0.45065 0.054111 -0.33882 0.0029376
+now, -0.14095 -0.023827 0.24233 -0.97745 -0.30549 -0.70309 0.36732 -0.0015611 -0.35641 0.29118
+tomorrow -0.088968 -0.047691 0.21455 -0.88194 -0.08835 -0.9363 0.43395 0.30085 -0.30457 0.15748
+unity -0.3308 -0.27777 0.24292 -0.88044 0.088659 -0.96242 0.39934 0.30319 -0.17101 0.09274
+volunteers -0.26174 -0.057436 0.089196 -0.96116 0.074692 -0.85164 0.4566 -0.0054521 -0.20545 0.028282
+Assa -0.28009 -0.15646 0.12637 -0.92162 0.1677 -0.92001 0.53836 0.060919 -0.24258 -0.024064
+created -0.080767 0.047301 0.21443 -0.88478 0.14305 -0.98661 0.42311 0.14744 -0.27115 -0.017732
+wall -0.099802 -0.014132 0.060633 -0.85707 -0.038637 -0.96233 0.37083 0.19281 -0.28115 0.16652
+coach -0.16856 -0.29836 0.23797 -0.90876 -0.11726 -0.84583 0.42414 0.1564 -0.25516 0.15827
+recovery -0.20675 -0.18212 0.19319 -0.89779 -0.064241 -0.85309 0.38424 0.17494 -0.21933 0.17171
+Switzerland -0.079992 -0.13598 0.18184 -0.92639 0.054323 -0.93894 0.37569 0.18704 -0.28502 0.12325
+enter -0.45322 -0.01919 0.18559 -0.79178 0.14156 -1.0558 0.60923 0.11858 -0.19179 -0.02854
+doubt -0.32574 0.053422 0.0062525 -0.9487 -0.018581 -0.97107 0.45578 -0.15671 -0.24103 -0.0065333
+cause -0.50551 -0.12227 0.088099 -1.0827 0.016891 -0.88276 0.37197 -0.12661 -0.062636 0.15792
+crowd -0.40782 -0.23396 0.12535 -1.0146 0.051628 -0.88484 0.3862 -0.14947 -0.070023 0.21455
+students -0.19614 -0.093446 0.18796 -1.0112 0.32153 -0.915 0.4746 0.072986 -0.18092 0.036884
+yachts -0.10043 -0.13598 0.28857 -0.93148 0.16356 -0.94541 0.56432 0.17456 -0.27281 0.16088
+mountain -0.5485 -0.088163 0.27787 -1.0609 0.43525 -0.9536 0.4235 -0.0033296 -0.087574 -0.17432
+oil -0.49401 0.027284 0.086111 -0.77788 -0.26105 -0.89366 0.44204 0.070293 -0.21487 -0.035968
+names -0.12431 -0.10413 0.14273 -0.95384 0.0011781 -0.79293 0.37917 -0.00022532 -0.25695 0.13455
+Eve 0.026754 0.033638 0.13392 -0.94275 -0.146 -0.70248 0.46788 0.14457 -0.42444 0.19181
+boats -0.0297 -0.14933 0.3111 -1.04 0.14434 -0.76292 0.42072 0.15491 -0.32255 0.20952
+Philip -0.19464 -0.10752 0.22174 -0.94747 0.077592 -0.9701 0.41533 0.13309 -0.25159 -0.020048
+While -0.32228 -0.12627 0.1897 -0.89911 0.094805 -0.92193 0.48529 0.14728 -0.31592 0.074455
+property -0.24023 -0.26967 0.17315 -0.96327 -0.46607 -0.69738 0.31227 0.11607 -0.24229 0.4851
+River. -0.2681 -0.16975 0.18522 -0.90112 -0.08442 -0.99434 0.40489 0.13814 -0.1938 0.15909
+acting -0.53013 -0.14158 -0.054806 -0.86516 -0.18841 -0.89288 0.42395 0.0077935 -0.24085 -0.023856
+attacks, -0.26184 -0.064943 0.076111 -1.135 1.1445 -1.2878 0.43577 -0.25375 -0.021592 -0.66229
+80 0.19687 -0.0052728 0.11849 -0.99997 -0.26229 -0.74572 0.24241 0.16065 -0.50429 0.38195
+them," -0.12137 -0.21374 0.13169 -1.0482 -0.045658 -0.7064 0.35385 -0.059479 -0.23442 0.25225
+verdict -0.37523 -0.14589 0.21297 -0.90034 0.0082628 -0.88882 0.44354 -0.023509 -0.11883 0.14185
+together -0.41599 -0.11771 0.11297 -0.89298 -0.35122 -0.78593 0.44979 -0.018914 -0.2029 0.32439
+apparently -0.28669 -0.14803 0.16608 -0.90586 -0.11703 -0.78704 0.45045 0.14256 -0.27331 0.16938
+aboard -0.22315 -0.15286 0.22083 -0.9105 0.11552 -0.92815 0.47306 0.053682 -0.17567 0.17077
+area, 0.060098 -0.28676 0.48758 -1.1426 -0.19647 -0.58428 0.30434 0.084059 -0.22395 0.62962
+affected -0.31576 -0.14433 0.15533 -0.90465 0.35061 -1.0104 0.42994 0.02733 -0.076688 -0.10059
+reveal -0.29761 -0.14128 0.13282 -0.90104 -0.13577 -0.93424 0.41919 0.13418 -0.17131 0.14438
+Firefighters -0.14621 -0.1608 0.39446 -1.0887 0.32588 -0.87854 0.47296 0.21558 -0.23373 0.026986
+squad -0.11952 0.19318 0.11819 -1.0138 0.35206 -0.94827 0.57597 0.016621 -0.25446 -0.094914
+swept -0.254 -0.38474 0.3558 -0.89072 -0.2463 -0.79614 0.21236 0.2282 -0.14143 0.40372
+played -0.12248 -0.10612 0.044413 -0.99036 0.062518 -0.93714 0.40971 0.03221 -0.22713 0.076656
+agreed -0.47498 -0.17392 0.11952 -0.93832 0.12854 -1.0205 0.44601 0.026194 -0.083695 -0.075291
+hope -0.20079 -0.22468 0.37713 -1.0259 -0.34884 -0.56551 0.30759 0.068423 -0.3139 0.42521
+Hicks, -0.49856 -0.19744 0.00187 -0.99613 0.52396 -1.0283 0.50948 -0.19591 0.031717 -0.14309
+ready -0.30419 -0.20335 0.17166 -0.85772 -0.46007 -0.75412 0.38355 0.1726 -0.12917 0.40266
+department -0.29198 -0.26417 0.19732 -0.83206 0.12443 -1.0298 0.50087 0.37477 -0.10588 -0.0049605
+doubles -0.18303 0.085722 0.041611 -0.91097 -0.12315 -0.87427 0.58882 0.046346 -0.31335 0.081859
+Gillespie -0.21623 0.065488 0.036796 -0.91373 0.032026 -0.98446 0.523 0.035139 -0.25906 0.014605
+scored -0.17742 0.027689 0.062962 -0.92025 0.35127 -1.0697 0.48265 0.086444 -0.25782 -0.10568
+conflict -0.10585 -0.21113 0.17809 -0.99384 0.22581 -0.88548 0.41597 0.18941 -0.15521 0.13629
+dropped -0.32043 -0.087391 0.11451 -0.94564 0.29418 -1.0107 0.48245 0.031455 -0.16764 -0.093651
+years, -0.0011375 -0.20937 0.11103 -0.99005 0.27352 -0.91087 0.54879 0.21073 -0.28111 0.15536
+Fatah -0.42844 -0.21625 0.049941 -0.90511 0.30765 -1.0389 0.42033 -0.096833 -0.06178 -0.086817
+Friedli -0.31514 -0.2202 0.13265 -0.9082 0.12679 -0.93712 0.36831 0.074354 -0.084924 0.047795
+old -0.69719 -0.18566 0.070685 -1.0067 0.10258 -0.97605 0.44238 -0.1796 -0.052982 0.089902
+Transport -0.15508 -0.11403 0.15107 -0.92124 0.092459 -0.8699 0.4467 0.1271 -0.29248 0.15401
+agency -0.30065 -0.21441 0.035038 -0.91231 0.56437 -1.0848 0.57941 0.034436 -0.11448 -0.15499
+follows -0.017571 0.10862 0.0052477 -0.90609 0.073521 -1.0277 0.5345 0.17643 -0.28208 0.019561
+streets -0.16264 -0.015076 0.17619 -0.93528 0.30503 -0.97626 0.52447 0.13421 -0.26039 0.029433
+debt -0.29361 0.01649 0.11836 -0.9269 -0.1014 -0.8797 0.37439 -0.051927 -0.20218 0.14944
+factions -0.66215 -0.3437 0.0784 -0.91268 0.086813 -0.84447 0.43038 0.10023 -0.11111 0.12388
+dozens -0.020118 -0.041503 -0.0025768 -0.944 0.048652 -0.88489 0.45687 0.10902 -0.31275 0.12495
+Hundreds -0.18794 -0.071425 0.086088 -1.0112 0.47302 -1.0115 0.48967 0.04175 -0.17609 -0.1515
+capital, -0.12553 -0.055046 0.031502 -1.019 0.64319 -1.1797 0.51824 0.0011017 -0.14031 -0.23638
+fierce -0.32886 -0.061441 -0.0085193 -0.97711 0.18292 -0.86985 0.49664 -0.058466 -0.14711 0.11091
+Sunday -0.2073 -0.0074925 0.04898 -0.92262 0.16386 -1.0042 0.49381 -0.054376 -0.23898 -0.070552
+2001 -0.34211 -0.11376 0.19554 -0.89605 0.34288 -0.93222 0.43281 0.0060886 -0.18179 -0.055397
+attempting -0.20869 -0.03348 -0.12391 -0.94975 0.37949 -1.0764 0.48798 -0.10882 -0.1925 -0.18612
+races -0.012184 0.16111 0.060192 -0.99748 0.41683 -1.0134 0.55748 0.12327 -0.26316 -0.13532
+prior -0.39419 -0.24487 0.17659 -0.9468 0.046177 -0.91649 0.39508 -0.010464 -0.17617 0.29622
+Japanese -0.24729 -0.2236 0.2223 -0.88269 0.19642 -0.9013 0.48408 0.14182 -0.19541 -0.016586
+domestic -0.090836 -0.11496 0.22062 -0.94755 -0.021954 -0.86699 0.48926 0.26807 -0.34971 0.16311
+Internet -0.44494 -0.081498 0.22393 -0.86936 0.25624 -1.0817 0.53033 0.019088 -0.15582 -0.081715
+spread -0.34286 0.0085029 0.12235 -0.96069 -0.014924 -0.85794 0.43675 -0.0030413 -0.14867 0.16879
+create -0.16881 0.0015007 0.25886 -0.89012 -0.22962 -0.83982 0.39156 0.25736 -0.3609 0.21564
+playing -0.30897 -0.044987 -0.018034 -0.89053 -0.18541 -0.96471 0.48865 -0.052599 -0.31886 -0.044458
+growing 0.019896 -0.049953 0.18254 -0.98348 -0.16335 -0.84057 0.47503 0.18023 -0.3839 0.12714
+scheduled -0.3439 -0.11101 0.15367 -0.98378 0.19435 -1.0042 0.41828 0.024073 -0.067868 -0.007766
+factory -0.45105 -0.087521 0.030059 -0.97854 -0.027822 -1.0114 0.44714 -0.057401 -0.21847 0.092436
+knowledge -0.3498 -0.13386 0.19503 -0.9762 0.07868 -0.96534 0.3975 -0.048246 -0.17176 0.031962
+save -0.38017 -0.23925 0.33928 -0.94021 -0.35619 -0.73018 0.26854 -0.049871 -0.066799 0.41109
+holiday -0.17117 0.021268 0.12256 -0.95257 0.54454 -0.99889 0.51952 0.066868 -0.25515 -0.20265
+Timor -0.086932 -0.14251 0.27041 -0.80165 0.1293 -0.96345 0.48208 0.44571 -0.29487 0.001174
+Thursday -0.16804 0.032178 0.17198 -0.89441 0.064809 -0.94998 0.5231 0.14831 -0.32066 -0.046057
+recent -0.14141 -0.21349 0.34583 -0.80142 0.047077 -0.96236 0.44747 0.59265 -0.15062 0.12291
+revealed -0.311 -0.1451 0.11148 -0.9362 0.070507 -1.0092 0.43316 0.023024 -0.12374 0.015503
+rain -0.71506 0.070684 0.009715 -0.93109 0.32285 -0.98704 0.52257 -0.17777 -0.1588 -0.21132
+Professor -0.34146 -0.047432 0.047709 -0.85468 -0.34078 -0.80689 0.38297 0.011803 -0.29741 0.29715
+"But -0.23533 -0.072159 0.11926 -0.96619 -0.25722 -0.77987 0.46747 -0.052188 -0.22124 0.32704
+statement. -0.25781 -0.23333 0.21852 -1.0081 0.55942 -1.0465 0.4925 -0.026883 -0.06044 -0.095841
+Solomon -0.1112 -0.17747 0.30167 -0.94309 0.1185 -0.95132 0.39373 0.2722 -0.16205 0.048385
+organisations -0.31664 -0.15481 0.086478 -0.91277 0.24868 -0.91472 0.51023 0.1276 -0.26093 0.067923
+runs -0.081029 0.15482 0.10812 -0.95525 0.16914 -0.87168 0.50043 0.24514 -0.40289 -0.089365
+respond -0.12951 -0.13325 0.25617 -0.99682 -0.036371 -0.84081 0.34943 0.22771 -0.22782 0.15075
+Michael -0.27256 -0.1403 0.14789 -0.84897 0.11625 -0.89961 0.44581 -0.021928 -0.093597 0.08034
+When -0.3826 -0.067863 0.1181 -0.95835 -0.24689 -0.80123 0.4205 -0.12366 -0.28273 0.26885
+40 -0.080944 -0.030479 0.48167 -1.0393 -0.23531 -0.65996 0.23723 0.1509 -0.12914 0.17171
+Hayden -0.26217 0.13508 0.056591 -1.0181 0.060048 -0.92058 0.46848 -0.06749 -0.26086 -0.062866
+attack. -0.2187 -0.054543 0.15759 -1.1832 0.91744 -1.218 0.35545 -0.205 -0.083368 -0.48719
+Earlier -0.35822 0.034622 0.045342 -0.84545 0.2501 -1.0562 0.53407 -0.14539 -0.2124 -0.041862
+Indonesia -0.13118 0.068667 0.079238 -0.90011 -0.074519 -0.92243 0.52518 0.13945 -0.25327 0.099072
+Sarah -0.29148 -0.083626 0.10898 -0.96221 0.095324 -0.90043 0.48371 0.16895 -0.13041 0.13251
+detain -0.35919 -0.070197 0.25744 -0.8393 0.011666 -0.98708 0.45426 0.19846 -0.17789 0.032611
+Neil -0.4231 0.11533 0.060575 -0.94699 -0.31259 -0.88055 0.5079 0.012363 -0.18818 0.047914
+states 0.035746 -0.23303 0.46542 -1.061 0.50206 -0.87712 0.40336 0.011969 -0.23725 0.047316
+4,000 -0.062591 0.016686 -0.018896 -0.91001 -0.16483 -0.83083 0.41536 0.25611 -0.34508 0.053079
+things -0.051032 -0.15411 0.11416 -0.91539 -0.13999 -0.74606 0.36922 0.042876 -0.36282 0.21292
+toll -0.051031 -0.074063 -0.050378 -0.81443 -0.16458 -0.86624 0.50923 0.13427 -0.31809 0.16723
+you're -0.059447 -0.12139 0.1556 -1.0472 -0.35454 -0.73766 0.27145 0.065707 -0.33578 0.33058
+felt -0.43935 -0.029357 0.056699 -1.0074 -0.012995 -0.77477 0.40277 -0.15507 -0.20878 0.065392
+deployed 0.053162 -0.1647 0.22864 -0.97735 0.35141 -0.99622 0.42883 0.28513 -0.2113 0.045679
+Hamas, -0.33651 -0.27008 0.13846 -0.95793 0.30384 -0.96349 0.42221 0.066626 -0.14882 0.029652
+gun -0.19859 -0.12881 -0.085064 -0.77327 -0.15347 -1.0126 0.51969 -0.022194 -0.37624 -0.040505
+Senior -0.22561 -0.12378 0.17376 -0.97379 0.44761 -0.99742 0.52638 -0.025954 -0.082844 -0.089523
+plan -0.40868 -0.099606 0.13781 -0.87832 0.010708 -0.91702 0.48558 -0.034796 -0.17617 0.13689
+elected -0.13135 0.04798 0.14718 -0.88991 0.044746 -0.96795 0.40639 0.12036 -0.23581 -0.0068139
+government, -0.27946 -0.2585 0.1293 -0.85924 0.25008 -1.0282 0.60002 0.2802 -0.16703 0.092715
+north. 0.11999 0.082092 0.12268 -1.0197 0.096272 -0.83958 0.56722 0.2645 -0.31074 0.18826
+tailenders -0.37816 -0.044353 -0.0033328 -0.92389 0.26644 -1.0267 0.46643 -0.02207 -0.17795 -0.20222
+B-52 -0.33645 -0.14164 -0.034415 -1.0395 0.44756 -0.93329 0.52896 -0.1543 -0.19566 -0.048659
+advice -0.019011 -0.080755 0.13782 -1.0027 -0.14265 -0.75955 0.37166 0.25948 -0.27782 0.3537
+continues -0.31877 -0.08533 0.23952 -1.0301 0.1736 -0.92963 0.4279 0.22104 -0.16045 0.032999
+Lording -0.44379 -0.31877 -0.02095 -0.89717 0.27762 -1.0258 0.49945 -0.15816 -0.072751 -0.17291
+body -0.27807 -0.043922 -0.064788 -0.85849 0.4291 -1.0043 0.5301 -0.15222 -0.14431 -0.10526
+died. -0.28004 -0.15246 -0.050408 -0.89943 0.25872 -1.1047 0.51751 0.027965 -0.16932 0.011267
+Melbourne, -0.16467 -0.0016058 0.15283 -0.917 -0.25817 -0.83193 0.39381 0.033862 -0.38796 0.1885
+activity -0.4133 -0.23608 0.042298 -0.90558 0.024862 -0.93733 0.37843 0.10265 -0.13713 0.092931
+Krishna -0.37526 -0.19429 0.26769 -0.98535 0.13161 -0.93223 0.42722 -0.041666 -0.18585 0.028535
+crossed -0.070605 -0.052971 0.25836 -1.0456 0.26535 -1.0064 0.47807 0.18456 -0.31118 -0.08613
+described -0.21603 -0.1074 0.13553 -0.93613 0.040677 -0.93754 0.39861 0.10401 -0.22784 0.060417
+suffered -0.24294 -0.18708 0.019853 -0.95508 0.11403 -0.92028 0.46586 -0.0031084 -0.14647 0.12232
+500 -0.10623 0.18396 -0.090181 -1.01 0.1596 -0.86083 0.49228 0.10059 -0.22581 -0.014477
+militants. -0.45571 -0.19868 0.017953 -0.93778 0.71889 -1.1236 0.51809 -0.12618 -0.058003 -0.21578
+rescue -0.26812 -0.02935 0.10839 -0.92493 -0.17213 -0.96675 0.37002 0.15285 -0.25694 0.13041
+walk -0.22773 0.038052 -0.11351 -0.80286 0.32745 -1.1318 0.54419 0.0089137 -0.22752 -0.081451
+That -0.1364 -0.029955 0.15664 -0.88105 -0.49632 -0.69868 0.40827 0.12486 -0.43356 0.32152
+diplomatic -0.25385 -0.12728 0.056069 -0.94035 0.0083836 -0.90342 0.41262 -0.015823 -0.21035 0.12273
+directors -0.077749 -0.14422 0.21524 -0.95906 -0.036602 -0.91827 0.31253 0.2058 -0.25 0.12023
+concern -0.30573 -0.29248 0.31705 -0.99877 0.075426 -0.84557 0.39936 0.22679 -0.052918 0.24857
+Ricky -0.53879 -0.0035531 0.062384 -1.0496 0.017301 -0.8555 0.45923 -0.17582 -0.26331 -0.021441
+attacking -0.19124 -0.011663 0.046328 -1.0701 0.61407 -1.1884 0.46141 -0.13146 -0.16939 -0.36588
+handed -0.32916 -0.28368 0.13324 -1.0037 0.43082 -1.022 0.45323 -0.087054 -0.061207 -0.075076
+edge -0.030324 0.011877 0.20697 -0.91813 0.29694 -1.065 0.38935 0.097339 -0.26136 -0.037787
+weekend. -0.24846 -0.18026 0.12221 -0.98336 0.063706 -0.77877 0.3955 -0.036036 -0.26551 0.25798
+why -0.54003 -0.17741 0.15215 -0.98643 0.38637 -1.0131 0.40792 -0.3226 -0.1438 0.0090477
+country. -0.33879 -0.12475 0.19287 -0.88116 0.034972 -0.88351 0.47031 0.21101 -0.26327 0.064972
+promised -0.33058 -0.19621 0.14678 -0.87526 0.24515 -1.0454 0.45322 0.11116 -0.14254 -0.030859
+Radio -0.40828 -0.053057 0.12387 -0.9347 0.2561 -0.9069 0.5037 -0.12057 -0.10187 -0.042171
+According -0.40444 -0.16357 0.086887 -0.89811 0.32539 -1.0539 0.52773 -0.041895 -0.20015 -0.28409
+investigating -0.17936 -0.11615 0.037612 -0.88204 0.20279 -1.0224 0.50793 0.16582 -0.24511 -0.019877
+Sydney's 0.26561 0.12441 0.31024 -1.0951 0.18804 -0.78061 0.58081 0.22754 -0.37633 0.18482
+civil -0.27172 -0.091949 0.087172 -1.0009 0.018191 -0.81623 0.45942 0.026575 -0.21189 0.17785
+Ministry -0.67236 -0.23749 0.14073 -0.8914 0.29613 -1.0714 0.49386 -0.16486 0.15936 0.069926
+Pakistan. -0.862 -0.55687 0.208 -1.0142 0.66765 -1.1684 0.38819 -0.14924 0.07302 -0.32486
+blaze -0.45721 -0.094917 0.039373 -1.0435 0.53644 -1.008 0.48946 -0.11726 -0.058373 -0.1909
+form -0.34916 -0.017953 -0.015494 -0.90177 0.14814 -1.154 0.44586 0.098764 -0.18015 -0.20346
+showed -0.29118 -0.22855 0.27023 -1.0458 0.11953 -0.83314 0.3252 -0.025419 -0.054188 0.13709
+field -0.18158 -0.17807 0.31801 -0.99627 -0.10492 -0.7835 0.45081 0.068907 -0.17304 0.29566
+period -0.24461 -0.075453 -0.0059673 -0.87738 0.42241 -1.0123 0.52005 0.092256 -0.21545 -0.044755
+action. -0.65352 -0.39539 0.19935 -0.9616 -0.052078 -0.81066 0.31832 0.10863 -0.05101 0.15773
+threatened -0.28797 -0.095794 0.1904 -0.98768 0.20751 -0.9698 0.3816 -0.079589 -0.1922 0.073733
+game -0.19809 -0.0086769 -0.066625 -0.93597 -0.36786 -0.73464 0.42883 0.018447 -0.48203 0.24025
+open -0.30496 0.025563 -0.024002 -0.92346 0.24914 -1.0943 0.49335 -0.27707 -0.17528 -0.25764
+shows -0.24444 0.016777 0.12173 -0.95246 0.04824 -0.87266 0.541 0.10345 -0.3015 -0.0094383
+hospital. -0.0014781 -0.058618 0.13286 -0.97958 0.23553 -0.93121 0.53765 0.15109 -0.27042 0.060958
+largest -0.20056 -0.048746 0.099227 -0.95552 0.37308 -1.0641 0.49084 0.073491 -0.37554 -0.077533
+responsible -0.11128 -0.14981 0.22246 -0.95417 -0.051653 -0.87557 0.35739 0.17158 -0.29181 0.14
+completed -0.18283 -0.13835 0.19334 -0.96165 0.41236 -1.0265 0.48639 0.18093 -0.17365 -0.083775
+Authorities -0.28478 -0.16736 0.17012 -0.96326 0.24699 -0.84565 0.52337 0.086151 -0.21086 0.11229
+fall -0.18331 0.060063 0.11881 -0.83354 -0.49483 -0.83717 0.39998 0.16868 -0.24348 0.3259
+"I'm -0.26948 0.045758 0.030334 -0.99897 -0.078027 -0.95211 0.43937 -0.12327 -0.20389 0.049463
+planes -0.29143 -0.1056 0.22846 -0.94404 0.17902 -0.94397 0.40442 -0.061847 -0.11728 0.084152
+met -0.26377 0.069661 0.043273 -0.88494 0.002914 -0.90562 0.42216 0.063408 -0.097371 0.0045662
+2002 -0.45596 -0.17479 0.17879 -0.90017 0.19999 -0.89563 0.49938 -0.14118 -0.2134 -0.051498
+Crean -0.2852 -0.10927 0.11987 -0.80172 -0.63576 -0.77574 0.4467 0.17397 -0.37047 0.30685
+representing -0.24022 -0.0811 0.083087 -0.86165 -0.11595 -0.91291 0.43206 0.24796 -0.25011 -0.0014316
+review -0.18524 -0.067561 0.19797 -0.94368 -0.17154 -0.85699 0.47081 0.13944 -0.23968 0.2017
+Yallourn -0.47021 -0.021575 0.01116 -0.8892 0.058852 -0.92951 0.53884 -0.036631 -0.16029 0.057955
+quarter -0.44592 -0.039945 0.034074 -0.86233 -0.051805 -0.92575 0.52607 0.010688 -0.26031 0.076642
+speech -0.63462 -0.20998 0.072527 -0.98417 0.35921 -1.0013 0.41784 -0.19271 -0.071603 -0.13002
+secure -0.32391 -0.19069 0.12847 -1.0189 0.5232 -1.0481 0.49084 -0.11453 -0.058657 0.059392
+meeting. -0.46792 -0.037851 -0.09161 -0.91255 0.31235 -1.1162 0.48363 -0.11875 -0.094529 -0.10114
+Territory -0.26832 -0.19567 0.073029 -0.90753 -0.10559 -0.89834 0.38328 0.092238 -0.24663 0.17531
+light -0.12712 -0.16921 0.18333 -0.9921 0.54441 -1.1114 0.51773 0.15336 -0.12766 -0.21983
+Adelaide. -0.025989 0.085196 0.11826 -0.84865 -0.073398 -0.90715 0.44745 0.21695 -0.47422 -0.03469
+month, -0.12487 -0.077932 0.027586 -0.85013 0.3364 -1.0947 0.47631 0.14132 -0.23213 -0.13515
+it. -0.16097 -0.0027678 0.30021 -0.9826 -0.52734 -0.59736 0.39112 0.12545 -0.3951 0.3784
+well," -0.35334 -0.018732 -0.051846 -0.95604 -0.12878 -0.79389 0.45722 -0.23372 -0.32752 0.14005
+hoped -0.37201 -0.2701 0.36301 -1.0897 0.35876 -0.81422 0.26953 -0.10798 -0.030669 0.067099
+"That -0.16934 -0.0016635 0.11367 -0.94955 -0.065852 -0.852 0.42582 -0.0042309 -0.21283 0.10859
+voice -0.21974 -0.20248 0.1113 -0.92958 -0.18162 -0.78683 0.37163 0.098087 -0.13725 0.30602
+Strip, -0.17654 -0.14265 0.1576 -0.99836 0.57331 -1.0703 0.49526 0.014806 -0.078684 -0.062875
+rival -0.18643 -0.082184 0.053126 -0.83016 -0.29496 -0.8513 0.42921 0.22511 -0.28777 0.25562
+documents -0.037684 -0.091584 0.076233 -0.89734 0.43422 -0.98438 0.59961 0.22531 -0.33378 -0.066055
+conducted -0.1383 -0.037729 0.1417 -0.98767 0.069707 -0.88731 0.42623 0.12239 -0.22512 0.12653
+became -0.16751 -0.17281 0.14356 -0.95021 -0.21699 -0.76155 0.39371 0.10439 -0.30135 0.30855
+Three -0.11491 0.034153 0.1029 -0.96994 0.15153 -1.0152 0.51004 0.23708 -0.35258 0.01072
+drug -0.10335 -0.22204 0.27887 -0.78724 -0.11928 -0.83802 0.3648 0.28033 -0.3473 0.27315
+Channel -0.40403 -0.22723 0.099833 -0.91948 0.12351 -0.99368 0.40838 0.0019438 -0.04688 0.12416
+adequate -0.28001 -0.064033 0.2613 -0.96333 0.078272 -0.93512 0.45222 0.20864 -0.27414 0.020279
+winner -0.24263 0.052 0.17734 -1.0189 -0.040712 -0.94032 0.45878 -0.046781 -0.27556 0.16295
+Gary -0.22597 0.027422 0.087171 -0.88541 -0.061774 -0.90493 0.45046 0.10433 -0.25775 0.069412
+Costello -0.38752 -0.17864 0.023341 -0.99545 0.33056 -1.0319 0.4482 -0.083534 -0.10498 -0.01931
+Mohammad -0.39264 -0.03019 0.11398 -0.87596 0.088261 -0.86994 0.4628 -0.085696 -0.2002 0.055633
+month. -0.079094 -0.062628 0.086861 -0.85169 0.093903 -1.0001 0.42942 0.21574 -0.27447 -0.018415
+Hospital -0.069514 -0.17098 0.17142 -0.91389 0.21038 -0.94193 0.48858 0.10507 -0.16306 0.061756
+worked -0.3658 -0.18651 0.16302 -1.0202 0.20734 -0.93103 0.45036 -0.045388 0.011548 0.033307
+No -0.06644 0.047094 -0.0072464 -0.97361 0.35235 -1.0128 0.65638 0.1251 -0.27042 0.074347
+Home -0.19708 -0.16752 0.23055 -0.91908 -0.28299 -0.78355 0.37316 0.28967 -0.33267 0.1164
+finally -0.2225 0.023568 0.042913 -0.88631 -0.21756 -0.87086 0.41333 0.12565 -0.24964 0.24162
+system -0.32424 -0.11447 0.022804 -0.86424 0.41032 -1.087 0.59199 0.05113 -0.16408 -0.029677
+low -0.42441 0.058064 0.24926 -0.94843 -0.41338 -0.72884 0.31984 -0.049308 -0.14088 0.27766
+people. -0.12527 -0.067296 0.039527 -0.90466 0.085485 -0.90647 0.50767 0.094092 -0.33578 0.091639
+tell -0.32064 -0.16263 0.1469 -1.0216 -0.58306 -0.65198 0.35841 0.054218 -0.3 0.40143
+separate -0.27425 -0.12546 -0.021667 -0.87955 0.1923 -0.97197 0.5305 0.075712 -0.20981 0.095925
+Rumsfeld -0.20524 -0.15714 0.14062 -1.0314 0.0023828 -0.79159 0.43273 -0.032572 -0.2126 0.21579
+Timor's -0.14788 -0.19963 0.31471 -0.94354 0.028989 -0.85614 0.43924 0.35624 -0.2247 0.097835
+assisting -0.42728 -0.074091 -0.084522 -0.8508 0.32864 -1.0608 0.50128 -0.053758 -0.16812 -0.19792
+regional -0.38529 -0.035004 0.15997 -0.77454 0.25105 -1.0563 0.52642 0.19416 -0.086911 -0.093535
+real -0.36399 -0.25527 0.23231 -0.85246 -0.43477 -0.81053 0.37251 0.24543 -0.21747 0.35087
+travelled -0.30306 -0.11726 0.12499 -0.88953 0.034301 -0.99596 0.39319 0.059334 -0.14803 0.070926
+personnel -0.24849 -0.16882 0.1015 -1.0017 0.31558 -1.0015 0.45331 -0.048916 -0.13477 0.066244
+ability -0.27173 -0.19358 0.049492 -0.99239 0.61827 -1.1174 0.4664 -0.022641 -0.1291 -0.19477
+shopping -0.084733 0.028369 0.081703 -0.91519 0.18997 -0.90579 0.52277 0.094239 -0.38535 -0.065638
+offered -0.28105 -0.32388 0.079595 -0.95111 0.095603 -0.92738 0.43216 0.026027 -0.059018 0.13014
+well." -0.31512 -0.066211 0.095542 -0.90755 -0.52851 -0.67737 0.41563 0.076345 -0.40165 0.30207
+republic -0.29533 -0.10822 0.024253 -0.89019 -0.13072 -0.883 0.40501 0.12325 -0.22216 0.1724
+tragedy. -0.2544 -0.075438 0.20771 -0.87111 -0.17515 -0.9306 0.40912 0.14662 -0.34255 0.01546
+Sharon, -0.54722 0.14504 -0.09222 -0.86272 0.29718 -1.2241 0.55062 -0.51179 0.04589 -0.062425
+waiting -0.33496 -0.061463 -0.06793 -0.86166 -0.10127 -0.94691 0.42146 0.071077 -0.2626 -0.007498
+Health -0.12165 -0.15446 0.21313 -0.90986 -0.06678 -0.81807 0.35183 0.025407 -0.17704 0.19281
+track -0.33687 -0.28419 0.27461 -0.85502 0.094422 -0.93155 0.43375 0.019345 -0.14151 0.15344
+problems -0.23635 -0.057152 0.13733 -0.96691 -0.28434 -0.78862 0.46773 0.060729 -0.28674 0.3308
+seriously -0.29599 -0.20022 0.10944 -0.90622 0.39272 -1.011 0.52848 -0.049441 -0.11413 0.075811
+Illawarra -0.12312 -0.10931 0.231 -0.94276 -0.040194 -0.88484 0.44864 0.1819 -0.33259 0.13047
+Virgin -0.43737 -0.059672 0.022511 -0.87982 0.25952 -1.0294 0.55749 0.038893 -0.22717 -0.13892
+television -0.40775 -0.15439 0.0078305 -0.89733 0.12706 -1.0102 0.4374 -0.1198 -0.047093 0.11185
+hours, -0.16732 -0.082988 0.24497 -1.0274 -0.028222 -0.85568 0.27882 0.06291 -0.27767 0.13924
+south-west -0.032608 -0.049815 0.19882 -1.0478 0.52558 -0.92695 0.59419 0.12515 -0.38462 0.03899
+Mohammed -0.43229 -0.12045 0.13958 -0.96673 0.4586 -1.025 0.45631 -0.13927 -0.044081 -0.13244
+Washington. -0.28267 0.034559 0.095128 -1.0139 0.13886 -0.91053 0.44615 0.050118 -0.28358 -0.058168
+"His -0.40711 -0.089346 0.022946 -0.9222 -0.045735 -0.92028 0.47429 -0.021138 -0.25581 -0.011387
+landed -0.22248 -0.17765 0.11213 -0.9833 0.64399 -1.1873 0.47176 -0.053128 -0.032894 -0.17665
+individuals -0.30371 -0.097623 0.12036 -0.86899 -0.12237 -0.91775 0.39886 0.080975 -0.15119 0.13273
+resistance -0.61039 -0.31171 0.15754 -0.96852 0.24458 -0.94637 0.43058 0.0090633 -0.098713 -0.037902
+Mayor -0.27184 -0.093265 0.19896 -0.9807 -0.039304 -0.88642 0.3541 0.054288 -0.12868 0.012767
+criminal -0.2211 -0.12247 0.23261 -0.85552 0.067805 -0.98377 0.47484 0.19956 -0.25257 0.014181
+representation -0.37881 -0.21401 0.17627 -0.88162 0.080312 -0.96165 0.42385 0.2562 -0.16246 0.053188
+His -0.4522 -0.28315 0.15234 -0.87023 0.22684 -0.91788 0.45055 -0.13313 -0.10547 0.12085
+territories -0.32634 -0.23142 0.095182 -0.9543 0.14923 -0.91172 0.4212 -0.0087514 -0.13068 0.099214
+observers -0.28079 -0.11129 0.12121 -0.97759 0.33095 -0.95931 0.50448 -0.056571 -0.13655 0.021243
+Owen -0.40788 -0.0014133 -0.11052 -0.94474 0.53525 -1.2078 0.64991 -0.091348 -0.11122 -0.25833
+sending -0.51309 -0.1859 0.084034 -0.88158 0.019714 -0.9883 0.49863 -0.031452 -0.19263 -0.072645
+26 -0.0006202 0.10724 -0.022284 -0.82976 0.068064 -0.98822 0.51966 0.14629 -0.40722 -0.11155
+Sector -0.31445 -0.17617 0.15852 -0.88677 -0.069458 -0.83895 0.38908 0.031061 -0.23409 0.11124
+embassy -0.26549 -0.017364 0.21213 -0.90806 0.069404 -0.94696 0.49348 0.20868 -0.21777 -0.020424
+shuttle -0.20957 0.014762 0.13739 -1.0199 -0.025479 -0.85309 0.46138 0.04258 -0.29832 0.1862
+ban -0.53768 -0.02236 0.095289 -0.96129 -0.13797 -0.85195 0.39781 -0.14755 -0.08554 0.1538
+ANZ -0.19758 -0.16704 -0.05023 -0.91872 0.364 -0.98109 0.44476 0.053657 -0.14595 -0.0029469
+Ahmed -0.50308 -0.21174 0.097447 -1.0042 0.39655 -1.0429 0.30618 -0.12886 -0.10869 -0.094447
+request -0.1231 -0.036596 0.10909 -0.90715 0.19061 -0.95619 0.52876 0.28876 -0.30993 -0.099247
+unemployment -0.29502 -0.24977 0.1978 -0.94719 -0.065665 -0.86253 0.50078 0.13676 -0.16593 0.23636
+assistance -0.63903 -0.28618 0.075197 -0.9545 0.40815 -0.99153 0.49461 -0.078772 -0.10471 -0.10326
+Launceston -0.37547 -0.0064214 0.051219 -0.89427 -0.067292 -0.87666 0.49713 0.056742 -0.29787 0.077475
+Wayne -0.12187 0.068928 0.017605 -0.99255 0.57536 -1.0356 0.57311 0.14447 -0.252 -0.14248
+Boucher -0.52523 0.025364 -0.026015 -0.98984 0.13125 -0.97817 0.51024 -0.13463 -0.18826 -0.030091
+Indonesian -0.215 0.020276 0.034067 -0.87099 -0.005719 -0.92212 0.52643 0.056346 -0.1973 0.097429
+months, -0.10046 -0.069789 0.10003 -0.9277 0.24753 -1.0747 0.44745 0.18699 -0.26068 -0.08329
+murder -0.42637 -0.24334 0.12645 -0.78126 0.10648 -1.0982 0.47767 0.159 -0.11915 0.041442
+Whiting -0.25833 -0.023759 -0.047414 -0.88992 0.080393 -0.95249 0.46532 0.0084415 -0.28506 -0.07071
+convicted -0.087382 -0.14507 0.21741 -0.97476 0.12423 -0.92747 0.37899 0.18639 -0.23692 0.10655
+positions -0.45257 -0.27477 0.085941 -0.85921 0.18458 -0.953 0.46735 0.14453 -0.21068 0.086495
+ethnic -0.3314 -0.013393 0.139 -0.9376 0.002369 -0.81834 0.44175 -0.056331 -0.19796 0.11059
+About -0.1777 -0.16339 0.3381 -0.82766 0.019878 -0.88386 0.32885 0.3276 -0.32283 0.036568
+success -0.27618 -0.13046 0.098235 -0.9085 0.16715 -0.91378 0.51658 0.01549 -0.19144 0.0049447
+Matthew -0.40921 -0.082844 -0.0025165 -0.93614 0.19289 -0.94652 0.47355 -0.17839 -0.17524 -0.056866
+adding -0.23709 -0.1576 0.04681 -0.92208 0.30419 -1.0458 0.53208 0.030716 -0.21109 -0.13296
+afternoon, -0.54182 -0.15146 0.05249 -0.9633 0.31184 -0.9892 0.51719 -0.082694 -0.16372 -0.015537
+Several -0.13199 -0.30305 0.35307 -0.90982 -0.12317 -0.80691 0.29891 0.17346 -0.16832 0.24398
+doesn't -0.2277 0.012908 0.016402 -1.0144 -0.26756 -0.81014 0.45085 -0.052238 -0.38941 0.2301
+jets -0.1653 -0.18342 0.02247 -1.0515 0.78594 -1.0918 0.5612 -0.070135 -0.12446 -0.12714
+returning -0.22105 0.066884 0.058995 -0.93883 -0.047402 -0.98682 0.4423 0.021276 -0.28428 0.0085229
+Tasmania -0.36636 0.025534 -0.026354 -0.97525 0.6671 -1.1778 0.56041 -0.21786 -0.060114 -0.32723
+eventually -0.060651 -0.05693 0.20574 -0.99548 -0.093327 -0.84715 0.45356 0.16121 -0.27674 0.2399
+turn -0.2635 -0.051435 -0.014098 -0.92409 0.2041 -0.8606 0.59687 -0.1945 -0.073028 0.1483
+leaving -0.23264 -0.1132 0.032918 -0.9093 -0.21705 -0.873 0.42549 -0.048552 -0.30136 0.17416
+City, -0.067429 0.11399 -0.11627 -1.0337 -0.085067 -0.8824 0.47107 -0.10358 -0.38238 0.26292
+blasted -0.033591 -0.078305 0.26092 -0.99451 0.95927 -1.1888 0.50083 0.1436 -0.072922 -0.32647
+ambush. -0.47088 -0.12169 -0.068579 -1.014 0.35499 -1.134 0.47717 -0.22519 -0.063657 -0.10275
+walked -0.18161 -0.14404 0.088765 -0.92529 0.35004 -1.1199 0.35355 0.093609 -0.069034 -0.17966
+infected -0.19348 -0.038659 0.12921 -0.8858 0.38597 -1.0568 0.47456 0.073278 -0.12294 -0.08691
+connection -0.40445 -0.24466 0.095035 -0.90027 0.12359 -0.97314 0.44386 0.1951 -0.14262 0.12853
+throughout -0.31055 -0.063748 0.18935 -0.97882 0.030535 -0.94017 0.38571 0.063454 -0.20595 0.050733
+"We've -0.18621 -0.073269 0.12786 -1.0151 -0.29615 -0.82385 0.37146 0.077991 -0.22902 0.20631
+aware -0.093168 -0.26167 0.33907 -1.0078 -0.09483 -0.78338 0.28706 0.091563 -0.22648 0.2203
+initial -0.20161 -0.2792 0.020668 -0.88278 0.2253 -1.0335 0.48082 0.21186 -0.14592 0.14193
+batsmen -0.21912 -0.096826 0.10785 -1.0118 -0.1907 -0.75203 0.45697 0.12281 -0.31518 0.22863
+publicly -0.26759 -0.13199 0.13334 -0.93198 -0.064618 -0.82329 0.39054 0.032715 -0.24503 0.17005
+hijacked -0.27263 -0.11454 0.064381 -1.0713 0.32305 -1.0128 0.45111 -0.11007 -0.085915 0.031842
+hotel -0.13685 -0.15678 0.12426 -0.89428 0.094665 -0.9525 0.50199 0.049709 -0.19466 0.1286
+manager -0.36589 -0.28577 0.16006 -0.89964 -0.051954 -0.89808 0.35038 0.068212 -0.22981 0.23828
+News -0.13397 0.038487 0.11931 -0.95999 0.2688 -0.82699 0.51309 0.16196 -0.29672 -0.090084
+whereabouts -0.26045 -0.23654 0.29012 -0.94527 0.10056 -0.86472 0.39543 0.13834 -0.21111 0.14724
+SES -0.060015 -0.093493 0.14755 -0.89576 0.27132 -1.0017 0.37142 0.29515 -0.23319 -0.054476
+passed -0.23991 -0.14631 0.1053 -0.92571 0.079771 -1.0391 0.35249 0.041943 -0.10139 0.010671
+retired -0.10288 -0.21485 0.22882 -1.0209 0.22266 -0.86894 0.30943 0.031934 -0.069143 0.091702
+Cabinet -0.39858 0.091476 0.20836 -0.95545 0.50144 -1.0545 0.51298 0.020529 -0.18466 -0.28999
+Hopman -0.23384 0.056702 0.13752 -0.88843 0.11919 -1.0044 0.50717 0.089605 -0.27833 0.026196
+Colin -0.50858 -0.30501 0.20478 -0.90787 0.047238 -0.85429 0.46799 -0.025284 -0.1396 0.16909
+France -0.54875 -0.027401 -0.014226 -0.98142 0.30926 -0.9169 0.4647 -0.08488 -0.13867 -0.093333
+halt -0.37949 -0.19086 0.26888 -0.95092 0.20975 -0.81348 0.34532 -0.10359 -0.094048 0.11118
+Seles 0.021606 0.14334 0.18284 -1.0069 0.073838 -0.83509 0.57831 -0.015797 -0.31883 0.055868
+leadership -0.5245 -0.24842 -0.041373 -1.0035 0.40459 -1.0631 0.41643 -0.17426 0.05289 0.0039623
+presence -0.36084 -0.15133 -0.003308 -0.88257 -0.002605 -0.84002 0.43271 0.06422 -0.1509 0.1411
+bringing -0.16257 0.07094 -0.063496 -0.89058 -0.066668 -0.98277 0.55783 0.071073 -0.42444 -0.11064
+Ford -0.68731 -0.17513 -0.0084982 -0.83995 0.29547 -0.92423 0.58389 -0.28806 0.0021562 -0.033388
+ashes -0.035294 -0.048422 0.2007 -0.92284 -0.29905 -0.77588 0.41888 0.1914 -0.34667 0.23776
+temporary -0.42203 -0.19107 0.047354 -0.95904 0.21416 -0.96639 0.50933 -0.055362 -0.12741 0.052021
+HIV -0.24905 -0.21377 0.063751 -0.87832 -0.084386 -0.8413 0.44261 0.065612 -0.28955 0.21276
+male -0.42632 -0.037384 -0.011601 -0.80641 -0.23765 -0.79392 0.52739 -0.072474 -0.31588 0.17107
+delivered -0.27494 -0.27154 0.20049 -1.0283 0.34314 -1.0314 0.3781 0.068735 -0.10579 0.013449
+stay -0.40191 -0.46309 0.35015 -0.9483 0.43225 -0.89636 0.38019 -0.093341 -0.069494 -0.028059
+place, -0.42016 -0.094504 0.0081534 -0.91748 -0.071619 -0.95206 0.38292 -0.062778 -0.26259 0.074912
+authority -0.44484 -0.28295 0.072215 -0.97754 0.1083 -0.90135 0.47309 0.044972 -0.11922 0.18977
+whatever -0.35706 -0.091354 0.2386 -0.86837 -0.083266 -0.82038 0.40701 -0.12165 -0.25287 0.18153
+Premier -0.42991 -0.093934 0.14489 -0.9427 0.30017 -1.0132 0.45915 -0.14094 -0.052598 -0.05255
+Washington -0.1942 0.075839 0.079679 -1.0062 0.12136 -0.92512 0.45686 0.092941 -0.31464 -0.073134
+farmers -0.30328 -0.25586 0.25466 -0.96498 0.20236 -0.90681 0.33278 -0.020502 -0.032443 0.074924
+hearing -0.11652 -0.12713 0.13908 -0.89716 0.094548 -1.017 0.4162 0.2249 -0.3312 -0.083326
+disaster -0.41233 0.0090258 -0.012068 -0.89477 0.23141 -1.0535 0.51601 -0.090431 -0.11046 -0.053681
+Hare -0.31633 -0.23029 0.39142 -1.0579 -0.029476 -0.71293 0.28781 -0.037969 -0.14417 0.06339
+fair -0.2138 -0.1823 0.13421 -0.87633 -0.31357 -0.84075 0.35768 0.028203 -0.20797 0.24274
+28-year-old -0.26208 -0.20888 0.14721 -0.98162 0.19933 -0.90627 0.47913 0.070672 -0.13381 0.10318
+manslaughter -0.18954 -0.10245 0.15641 -0.90749 0.046472 -0.9511 0.47872 0.17989 -0.25487 0.052103
+Services 0.099523 -0.10298 0.32188 -1.0072 -0.14518 -0.6677 0.39707 0.38495 -0.33886 0.32082
+Emergency -0.18084 -0.069069 0.13063 -0.93422 0.3176 -1.0159 0.53012 0.11359 -0.19818 -0.063753
+relationship -0.4527 -0.31467 0.065602 -0.83727 0.16268 -0.9873 0.51063 0.17667 -0.17547 0.1002
+allegedly -0.15841 -0.11791 0.23173 -0.95427 -0.064483 -0.95716 0.24609 0.25144 -0.18054 0.16803
+happy -0.24261 -0.14874 0.16108 -0.9851 0.16835 -0.88154 0.46359 0.049752 -0.27116 0.092684
+tensions -0.27807 -0.2446 0.1275 -0.95882 0.08456 -0.87565 0.43105 0.16766 -0.21819 0.15439
+Arafat, -0.69495 0.00283 -0.11653 -0.87107 0.30845 -1.1002 0.55539 -0.46929 0.13194 -0.013451
+actor -0.5157 -0.28967 0.11812 -0.93227 -0.35857 -0.7746 0.2859 -0.082813 -0.23848 0.29371
+seemed -0.61388 -0.29496 0.10962 -1.0043 0.36095 -1.0873 0.38044 -0.22345 0.087789 -0.058102
+headed -0.50502 -0.2573 0.24772 -1.059 0.37391 -1.0235 0.31954 -0.065931 0.0087435 -0.0023838
+injuring -0.098241 -0.11688 0.064764 -0.8593 0.32638 -1.1317 0.5642 0.29006 -0.33966 -0.19568
+Neville -0.3259 -0.14089 0.040579 -0.91155 -0.02612 -0.84816 0.50193 0.010655 -0.14942 0.21091
+self-rule -0.40224 -0.15451 0.073701 -0.99873 0.14058 -0.97193 0.49833 -0.06536 -0.13114 0.10273
+we'll -0.411 -0.18358 -0.045442 -0.91371 -0.66867 -0.69162 0.34793 -0.038717 -0.17974 0.46947
+faces -0.17063 0.026559 0.13962 -0.97514 0.50471 -1.0292 0.54584 0.04527 -0.0085925 -0.078847
+aged -0.13331 -0.10156 0.20415 -0.92413 0.15444 -1.1066 0.30317 0.19216 -0.11349 -0.18242
+sign -0.38234 -0.12628 0.19185 -0.98366 0.16336 -0.84411 0.39789 -0.18097 -0.10779 -0.035669
+Jenin -0.27241 -0.039568 0.088355 -0.93481 0.50369 -1.0301 0.52204 -0.0096526 -0.18061 -0.17257
+Nablus -0.21391 -0.096137 0.018651 -0.98315 0.424 -1.0754 0.55246 -0.10657 -0.15748 -0.0084811
+concerns -0.35554 -0.28261 0.25285 -1.0106 -0.065398 -0.85027 0.4072 0.16369 -0.094971 0.29612
+service -0.044501 -0.19176 0.15741 -0.92203 0.041666 -0.79793 0.44268 0.23301 -0.1867 0.33057
+today's -0.34094 -0.15605 0.092894 -0.92911 0.1317 -0.9693 0.45755 0.017209 -0.14467 0.078745
+Mt 0.14151 -0.042268 0.20287 -0.93268 -0.25601 -0.64688 0.41099 0.23649 -0.32969 0.38915
+industry -0.534 -0.085994 0.025382 -0.84774 -0.11158 -0.92475 0.44758 0.0081253 -0.13622 0.046628
+terrorism. -0.31517 -0.12548 0.21477 -0.88978 0.24459 -0.98603 0.47557 0.087711 -0.26356 -0.10607
+often -0.52879 -0.14935 0.10029 -0.93974 0.1651 -0.9411 0.55918 -0.21536 -0.11227 0.10149
+night, 0.12298 -0.12941 0.36066 -0.93444 0.41968 -0.88449 0.4654 0.53798 -0.2872 -0.05315
+escalating -0.33829 -0.12703 0.015898 -0.83725 0.11901 -0.99878 0.52081 0.13393 -0.21464 -0.051598
+previous -0.282 -0.10094 -0.0079603 -0.88312 -0.060852 -0.93805 0.51915 0.10054 -0.25168 0.20141
+Island 0.013039 -0.059081 -0.038447 -0.94126 0.53201 -0.99614 0.5743 0.16496 -0.13518 0.12471
+levels -0.2053 -0.10492 0.20118 -0.9795 0.18489 -0.82919 0.42957 -0.077108 -0.13166 0.16859
+India's -0.51326 -0.10508 0.24936 -0.87206 0.38369 -1.0817 0.53512 -0.1711 -0.03881 -0.21276
+Antarctic -0.36612 -0.26796 0.16656 -0.99177 0.060273 -0.91185 0.41627 0.03005 -0.18043 0.18036
+"Every -0.33802 -0.1243 0.14369 -0.95371 -0.4448 -0.73827 0.37005 0.016855 -0.21363 0.35746
+extremists -0.24011 -0.086676 0.028766 -0.94564 0.35898 -1.0907 0.5522 0.07233 -0.22499 -0.05126
+locked -0.31102 -0.084622 0.083962 -0.94166 0.34729 -1.0238 0.37482 0.023288 -0.14446 -0.017709
+unable -0.17401 -0.066264 0.08375 -0.90282 -0.22468 -0.90443 0.47118 0.043548 -0.3112 0.19806
+treatment -0.27478 -0.30204 0.32846 -0.8659 -0.31595 -0.77731 0.37383 0.39064 -0.25039 0.26144
+increased -0.20034 -0.16535 0.21299 -0.95197 0.36574 -0.97734 0.40417 0.030496 -0.12069 -0.042252
+Qantas' -0.66562 -0.27424 0.20254 -0.93174 0.015123 -0.89357 0.37683 -0.083202 0.031141 -0.024249
+choosing -0.3108 -0.05691 0.059514 -0.92087 0.021155 -0.94624 0.46132 -0.060476 -0.22563 -0.024965
+Manufacturing -0.36459 -0.03665 0.12481 -0.80146 -0.04296 -0.95224 0.47252 0.065796 -0.23579 -0.080454
+Park -0.27228 -0.24721 0.18172 -1.0477 0.061452 -0.88541 0.34821 0.10405 -0.1182 0.15918
+pace -0.51173 -0.056871 -0.25776 -0.95248 0.57897 -1.1676 0.55508 -0.091836 -0.093863 -0.16639
+intelligence -0.48843 -0.18543 0.0017036 -0.99733 0.25789 -1.0857 0.47285 -0.1039 -0.082249 -0.045634
+Peres -0.064541 -0.088503 0.16119 -0.905 0.18264 -1.007 0.52766 0.23608 -0.25072 -0.0042925
+Saturday -0.20459 0.043622 0.021364 -0.91458 0.15444 -0.97015 0.59302 0.020509 -0.24214 0.02135
+allowed -0.14503 -0.055127 0.15819 -0.98083 0.27823 -1.0507 0.3977 0.12058 -0.069834 0.064531
+follow -0.020472 0.067402 0.083804 -0.92167 0.10125 -0.98843 0.51323 0.22062 -0.24816 0.065072
+food -0.16719 -0.21457 -0.032412 -0.91796 0.3043 -0.98733 0.44674 -0.0045141 -0.095009 0.0023948
+effort -0.33103 -0.19333 0.16564 -0.95268 -0.005989 -0.90712 0.40634 0.14172 -0.12277 0.14287
+contested -0.1816 -0.074223 0.16935 -1.0212 0.25551 -0.98821 0.47956 0.14203 -0.19689 0.063591
+course -0.35694 -0.093528 0.16327 -0.92286 -0.52557 -0.70208 0.25882 0.007853 -0.29627 0.31276
+focus -0.094077 -0.063998 0.084013 -0.9035 0.35784 -1.0862 0.42149 0.072761 -0.20467 -0.10303
+staying -0.32977 -0.26031 0.20705 -0.92262 -0.15255 -0.90498 0.39614 0.0032056 -0.31511 0.0019217
+questions -0.30184 -0.17847 0.037072 -0.9023 0.20693 -0.9822 0.4992 0.17968 -0.24761 0.016212
+Child 0.015704 -0.10886 0.20551 -0.90874 0.032916 -1.0135 0.40535 0.16044 -0.18068 0.1597
+Austar -0.36952 -0.13053 0.075797 -0.84449 0.016521 -0.84864 0.5294 0.071495 -0.28827 0.068427
+trade -0.15353 -0.076031 0.26592 -0.97467 -0.03202 -0.86541 0.34478 0.050255 -0.34897 0.12444
+lack -0.28125 -0.049111 0.12849 -0.858 0.024126 -0.87174 0.43113 0.075511 -0.28163 0.027479
+document -0.097176 -0.17444 0.14257 -0.85602 0.30242 -0.96306 0.56906 0.25605 -0.2711 -0.016344
+explanation -0.38645 -0.18002 0.055084 -0.86557 0.18408 -0.97789 0.48657 0.18924 -0.18177 0.083669
+Sultan -0.41645 -0.16765 0.21599 -1.0014 0.30501 -1.0955 0.46808 0.10096 -0.14567 -0.054145
+reduced -0.21167 -0.080213 0.11708 -0.94597 -0.17172 -0.89563 0.39573 0.26074 -0.2101 0.15443
+violent -0.11532 -0.13904 0.23495 -0.86285 0.3395 -1.0167 0.46774 0.32004 -0.18438 -0.10924
+understanding -0.36627 -0.18224 0.14645 -0.93409 0.068406 -0.91532 0.43653 0.039594 -0.22393 -0.033249
+farm -0.2685 -0.094387 0.36153 -1.0836 0.22668 -0.88753 0.28388 0.019861 -0.054735 -0.082909
+Lord -0.44769 -0.071266 -0.13184 -0.99477 0.64386 -1.1481 0.62058 -0.35432 0.088442 -0.19185
+nearby -0.21346 -0.21208 0.0129 -0.97925 0.22082 -0.93188 0.49707 0.01654 -0.27514 0.09273
+Toowoomba -0.11086 -0.016608 0.12698 -0.98588 0.18283 -0.93644 0.51906 0.27584 -0.1859 0.068254
+redundancy -0.37279 -0.2126 0.13548 -0.92321 0.20175 -0.91295 0.37839 0.091666 -0.15195 -0.016421
+credit -0.04897 0.00094891 0.18637 -0.93684 0.2433 -1.0365 0.50859 0.28308 -0.33012 -0.1834
+entitlements -0.22865 -0.17366 0.25398 -0.95994 0.25625 -0.97027 0.48133 0.29908 -0.17171 0.0098023
+paying -0.27379 -0.21393 0.14191 -0.81481 -0.31424 -0.86041 0.34354 0.12124 -0.28744 0.086008
+Stuart -0.32065 -0.045169 0.047934 -0.93749 -0.01232 -0.94264 0.55829 0.019187 -0.23201 0.012652
+administrators -0.38915 -0.10678 0.10779 -0.90749 0.15211 -0.9458 0.49626 -0.0076005 -0.15467 0.13405
+150 -0.10669 0.052089 -0.080921 -0.99111 0.28881 -0.88495 0.50056 0.040523 -0.24744 -0.046004
+technology -0.3845 -0.1054 0.095591 -0.89516 -0.023254 -0.88988 0.44832 -0.036531 -0.17773 0.062035
+holding -0.30153 -0.1088 0.12251 -0.9244 0.088088 -0.93542 0.44833 0.070584 -0.22506 -0.058827
+normal -0.33125 -0.08666 0.21988 -0.91425 -0.26755 -0.74247 0.3954 0.059246 -0.2519 0.29009
+Amin -0.67553 -0.21073 0.21461 -1.0165 -0.20907 -0.78149 0.34416 -0.081891 -0.18986 0.18107
+Adam -0.44946 -0.048925 0.12921 -0.91341 -0.258 -0.85795 0.404 0.10336 -0.19082 0.18
+crashed -0.28983 -0.19277 0.28008 -1.0251 0.36846 -1.0111 0.34961 0.023464 -0.10933 -0.062239
+natural -0.07196 -0.27232 0.38912 -0.98727 -0.1257 -0.81443 0.29644 0.32534 -0.29026 0.20611
+begin -0.67268 -0.064473 -0.036203 -1.0396 0.39318 -1.0634 0.52304 -0.39897 -0.10272 -0.20808
+Up -0.12987 -0.31596 0.37794 -0.92846 0.34058 -0.87118 0.55491 0.18954 -0.065122 0.20761
+celebrations -0.35697 -0.24729 0.14783 -0.91635 0.17408 -0.84252 0.47401 0.19342 -0.26351 0.14232
+reject -0.36215 -0.083687 0.019168 -0.88026 -0.34821 -0.80624 0.35637 0.060272 -0.22834 0.17106
+options -0.43006 -0.32979 0.099825 -0.8432 0.28397 -0.94099 0.45111 0.18693 -0.12492 0.077929
+single -0.36768 -0.0077289 0.12405 -0.89668 -0.13589 -0.91868 0.43771 0.023178 -0.19012 0.020062
+handling -0.19383 -0.086433 0.073078 -0.84088 -0.022429 -0.96951 0.52248 0.18913 -0.26988 -0.062818
+match. -0.10931 0.18295 -0.0038214 -0.83137 -0.24983 -0.95995 0.53487 0.20493 -0.49541 -0.1036
+summit -0.31419 0.0098797 0.035068 -0.96421 0.14451 -1.0258 0.50231 -0.018111 -0.23626 -0.029974
+talks. -0.35229 -0.15515 0.07993 -0.99309 0.11575 -0.89788 0.43005 0.018662 -0.17911 0.030351
+All -0.82992 -0.12558 0.23378 -0.8514 -0.078518 -0.93991 0.46519 -0.30953 0.013705 0.18509
+settlement -0.34912 -0.2379 0.10761 -0.96062 0.31773 -1.0578 0.52877 0.033297 -0.049265 -0.010945
+searching -0.24236 -0.17512 0.17781 -0.95078 -0.16358 -0.74868 0.38088 0.085915 -0.27398 0.19703
+dollars -0.35512 -0.096205 0.13233 -1.0636 0.25495 -0.91864 0.53955 -0.11415 -0.17314 -0.015355
+guess -0.27593 -0.11835 0.12093 -0.93565 -0.078513 -0.84341 0.43357 0.059761 -0.35287 0.10868
+Kieren -0.28648 -0.052006 0.019997 -0.93716 0.048119 -1.0201 0.5268 -0.082165 -0.20706 0.12905
+23 -0.21086 -0.057727 -0.088208 -0.96071 0.56566 -1.1161 0.58002 -0.23208 -0.13051 -0.046464
+Bonn -0.63015 -0.090793 -0.010465 -0.89356 0.37004 -1.1065 0.49491 -0.13038 -0.12322 -0.18684
+... -0.43566 -0.18061 0.06388 -0.86017 0.013587 -0.82546 0.47717 -0.017943 -0.095205 0.12177
+prepare -0.27096 -0.23845 0.17099 -0.88899 -0.17308 -0.79854 0.33358 0.18857 -0.27852 0.22166
+champion -0.37096 -0.075525 0.033513 -0.87721 0.032021 -1.0175 0.42515 0.086005 -0.22466 0.0095515
+Pollock -0.17072 0.075854 0.034762 -0.99117 0.025959 -0.84027 0.51192 0.054958 -0.34076 0.13271
+television. -0.44044 -0.21914 0.064641 -0.95829 0.15258 -0.9742 0.43971 -0.15748 -0.021531 0.11953
+begun -0.44643 -0.063631 0.062254 -0.89464 -0.26594 -0.77289 0.43425 -0.022602 -0.29919 0.059812
+coast -0.096154 -0.098287 0.25499 -0.86603 0.41534 -0.9861 0.60581 0.24712 -0.30546 -0.11746
+leave -0.48798 -0.14912 0.15468 -1.0082 0.28873 -0.94628 0.46997 -0.14209 0.015677 0.16177
+St -0.15052 -0.038151 0.074615 -0.94029 0.62285 -0.93937 0.64598 0.084345 -0.21029 -0.10022
+Sydney. 0.26468 0.080405 0.32812 -1.0438 0.026161 -0.72776 0.57149 0.22449 -0.45732 0.20119
+losing -0.30082 -0.14351 0.13768 -0.94164 -0.073193 -0.87297 0.40935 -0.035951 -0.20497 0.083264
+work. -0.33936 -0.20084 0.24815 -1.0307 0.14337 -0.84725 0.4851 0.038437 -0.18419 0.15475
+counts -0.18804 -0.080086 0.15436 -0.93251 0.17173 -0.92948 0.56095 0.42776 -0.34996 0.025217
+26-year-old -0.23247 -0.16833 0.10058 -0.95923 0.12781 -0.89816 0.45605 0.021855 -0.15746 0.094273
+suggested -0.22162 -0.0057453 0.067216 -1.0141 0.46976 -1.0677 0.51246 -0.14477 -0.10022 -0.052623
+projects -0.38763 -0.26072 0.012426 -1.01 0.17994 -0.91495 0.52402 -0.090687 -0.25331 0.078616
+understood -0.39868 -0.13931 0.061138 -0.91247 0.04974 -0.9366 0.44174 -0.038707 -0.21788 0.022325
+various -0.42015 -0.30838 0.073304 -0.85863 0.37029 -1.0839 0.46362 0.028696 -0.00035373 -0.010771
+debate -0.12791 -0.10166 0.26514 -0.92209 0.13826 -0.92997 0.45138 0.26024 -0.26033 0.064546
+Bill -0.39838 -0.033529 0.094462 -0.87625 -0.2477 -0.78666 0.481 0.13671 -0.29902 0.21293
+happens -0.21991 -0.11012 -0.0028922 -0.9438 0.12635 -0.90128 0.55503 0.0095414 -0.22928 0.17408
+Commissioner -0.49284 -0.16213 0.12342 -0.90232 0.031972 -0.97126 0.46812 -0.01738 -0.030934 0.15521
+Deputy -0.36846 -0.16084 0.078157 -0.82351 -0.050522 -0.92733 0.5193 0.07008 -0.23301 0.11732
+civilians -0.26415 -0.15822 0.068405 -0.98221 0.50329 -1.0065 0.52038 -0.038207 -0.18997 -0.052481
+threatening -0.25182 -0.045476 0.10861 -0.89748 0.2866 -1.0112 0.42588 -0.062959 -0.30542 -0.11057
+women's -0.19161 -0.071387 0.19565 -1.0621 0.22142 -1.0313 0.47957 0.15791 -0.22421 0.099062
+containment -0.24544 -0.31776 0.35774 -1.0603 0.27667 -0.94787 0.45011 0.25615 -0.13414 0.083895
+stand -0.32437 -0.29159 0.29872 -0.96611 0.10026 -0.85177 0.31083 0.053036 -0.15696 0.15546
+MacGill -0.31908 0.030595 0.093019 -0.94111 -0.093146 -0.86372 0.49992 0.057094 -0.24437 0.11818
+putting -0.38698 -0.14535 0.03712 -0.90819 0.081707 -0.9084 0.42326 -0.057029 -0.26089 -0.021234
+determine -0.4176 -0.097098 0.14469 -0.93155 0.11262 -1.0324 0.43527 0.012073 -0.099626 0.040556
+Israel. -0.47031 -0.063841 -0.37841 -0.97351 1.1357 -1.3919 0.65755 -0.52246 0.1681 -0.27245
+forecast -0.19474 -0.079622 0.15075 -1.103 0.64718 -1.1086 0.38859 0.13428 -0.16916 -0.21521
+During -0.34549 -0.037444 0.073162 -0.8123 -0.10379 -0.97766 0.4863 0.040554 -0.3512 -0.056979
+bureau -0.18654 -0.084556 0.22349 -0.94008 0.17144 -0.88225 0.50717 0.042983 -0.26211 0.16669
+findings -0.15385 0.024256 0.097106 -0.90701 -0.0062973 -0.87097 0.45966 0.092639 -0.31016 0.048742
+fear -0.13347 -0.033548 0.051886 -1.0621 0.69087 -1.067 0.49215 -0.082951 -0.19496 -0.24658
+data -0.34699 -0.17287 -0.01897 -0.93141 0.24327 -0.91246 0.46874 -0.18289 -0.18046 0.081533
+gone -0.26956 -0.29501 0.25143 -1.0867 0.19435 -0.93538 0.33428 0.0059832 -0.084104 0.060041
+record -0.3523 0.061825 0.089374 -0.8624 0.10975 -0.93357 0.47165 0.088856 -0.19873 -0.037258
+hoping -0.18615 -0.052057 0.080318 -0.91618 0.10148 -0.88347 0.48114 -0.0095034 -0.35875 -0.1065
+Israelis. -0.23908 -0.12511 -0.23259 -1.0042 0.8512 -1.1663 0.54631 -0.34526 -0.007403 -0.12883
+Hamid -0.30894 -0.25919 0.073896 -0.86696 -0.060983 -0.83277 0.46301 0.015398 -0.12849 0.15373
+present -0.28584 -0.21305 0.1917 -0.82763 -0.2293 -0.86695 0.41269 0.28361 -0.18485 0.14715
+live -0.36521 -0.27784 0.21715 -1.0707 0.52157 -1.1198 0.27499 -0.065779 0.10911 -0.14034
+ahead. -0.18178 -0.16656 0.22029 -0.91903 0.26257 -0.96658 0.42367 0.15003 -0.19482 0.13553
+warning -0.057134 -0.10892 0.21894 -0.93471 0.075207 -0.93845 0.32891 0.098712 -0.29838 -0.10296
+trapped -0.39018 -0.17152 0.16364 -0.90232 0.39176 -1.0995 0.41738 -0.031247 -0.14006 -0.14978
+markets -0.22304 0.049703 0.023357 -0.89652 0.028235 -0.93915 0.54958 0.10373 -0.31918 -0.015176
+Sergeant -0.22138 -0.073868 0.18847 -0.89862 -0.12118 -0.88672 0.44595 0.059898 -0.20127 0.20083
+Seven -0.15834 0.012167 0.075471 -0.9344 -0.0060505 -0.87636 0.57467 -0.091161 -0.18515 0.16282
+firm -0.20924 0.13692 0.229 -1.0004 0.087825 -0.74557 0.46068 0.063433 -0.19736 -0.013484
+welcomed -0.33084 -0.13075 -0.031394 -0.933 0.12591 -0.93544 0.43404 -0.0071708 -0.19188 0.02752
+responding -0.27598 -0.17165 0.25794 -0.96352 -0.021708 -0.90139 0.36233 0.13162 -0.20719 0.013575
+law -0.23204 0.098585 0.084543 -0.87951 0.045764 -1.0408 0.39644 0.083872 -0.19391 0.06397
+deputy -0.35117 -0.094578 0.069262 -0.8022 0.21323 -1.0385 0.49627 0.14174 -0.18405 -0.036922
+unidentified -0.21788 -0.19765 0.16739 -0.96515 0.26464 -1.0681 0.45954 0.19657 -0.10865 -0.040348
+clashes -0.18525 -0.090192 0.13899 -0.92914 0.19958 -0.99479 0.48018 0.0076089 -0.22087 0.057853
+ago, -0.12196 -0.25092 0.21986 -0.78537 -0.43743 -0.83485 0.35358 0.5775 -0.36967 0.27372
+replied: -0.53654 -0.30794 0.0668 -0.95327 -0.32244 -0.86974 0.32233 0.0035099 -0.12836 0.27396
+path -0.29637 -0.0050163 -0.09227 -0.91761 0.61522 -1.2084 0.51978 -0.070516 -0.022066 -0.26732
+search -0.33437 -0.14509 0.16727 -0.8863 0.019863 -0.77068 0.47079 0.019677 -0.24613 0.18844
+hundred -0.025627 -0.11448 0.18116 -0.96779 0.2532 -0.93224 0.3929 0.12679 -0.19685 -0.13179
+state. -0.18528 -0.23642 0.26504 -1.0479 0.57506 -1.0035 0.41344 -0.17807 -0.09178 -0.023322
+efforts -0.2879 -0.13298 0.14961 -0.93882 0.13162 -0.93926 0.49649 0.17908 -0.18039 0.022261
+tree -0.13421 0.049938 0.039293 -0.85099 -0.054489 -0.95255 0.48989 0.20728 -0.25372 0.081726
+telephone -0.31696 -0.22565 0.13317 -1.0219 0.2304 -1.0449 0.39416 -0.015374 -0.058222 0.064146
+problem -0.23808 -0.055263 0.21232 -0.94406 -0.52183 -0.69284 0.40925 0.16283 -0.35028 0.39899
+approached -0.3755 -0.24085 0.045111 -0.96186 0.31842 -0.96877 0.4262 -0.091481 -0.06434 0.076188
+chairman -0.24927 -0.10477 0.17401 -0.90386 0.013637 -0.93125 0.41858 0.049651 -0.20749 0.15118
+Afroz -0.3492 -0.11356 0.046444 -0.96179 0.49956 -1.1739 0.50381 -0.029897 -0.15488 -0.24651
+Monday, -0.18335 0.075194 0.19071 -0.85914 -0.074304 -0.8325 0.45894 0.23434 -0.32902 0.11567
+advance -0.35487 -0.13718 0.083065 -1.0344 0.076132 -0.86051 0.39686 0.031053 -0.1538 0.12614
diff --git a/gensim/test/test_data/pre_0_13_2_model b/gensim/test/test_data/pre_0_13_2_model
new file mode 100644
index 0000000000..25bf224c54
Binary files /dev/null and b/gensim/test/test_data/pre_0_13_2_model differ
diff --git a/gensim/test/test_data/pre_0_13_2_model.state b/gensim/test/test_data/pre_0_13_2_model.state
new file mode 100644
index 0000000000..03bc8c8a68
Binary files /dev/null and b/gensim/test/test_data/pre_0_13_2_model.state differ
diff --git a/gensim/test/test_fasttext_wrapper.py b/gensim/test/test_fasttext_wrapper.py
index e1fd2cd722..b0de495ad4 100644
--- a/gensim/test/test_fasttext_wrapper.py
+++ b/gensim/test/test_fasttext_wrapper.py
@@ -35,8 +35,10 @@ def setUp(self):
self.ft_path = os.path.join(ft_home, 'fasttext') if ft_home else None
self.corpus_file = datapath('lee_background.cor')
self.test_model_file = datapath('lee_fasttext')
+ self.test_new_model_file = datapath('lee_fasttext_new')
# Load pre-trained model to perform tests in case FastText binary isn't available in test environment
self.test_model = fasttext.FastText.load_fasttext_format(self.test_model_file)
+ self.test_new_model = fasttext.FastText.load_fasttext_format(self.test_new_model_file)
def model_sanity(self, model):
"""Even tiny models trained on any corpus should pass these sanity checks"""
@@ -54,7 +56,7 @@ def testTraining(self):
if self.ft_path is None:
logger.info("FT_HOME env variable not set, skipping test")
return # Use self.skipTest once python < 2.7 is no longer supported
- vocab_size, model_size = 1762, 10
+ vocab_size, model_size = 1763, 10
trained_model = fasttext.FastText.train(
self.ft_path, self.corpus_file, size=model_size, output_file=testfile())
@@ -139,6 +141,36 @@ def testLoadFastTextFormat(self):
self.assertEquals(self.test_model.wv.min_n, 3)
self.model_sanity(model)
+ def testLoadFastTextNewFormat(self):
+ """ Test model successfully loaded from fastText (new format) .vec and .bin files """
+ new_model = fasttext.FastText.load_fasttext_format(self.test_new_model_file)
+ vocab_size, model_size = 1763, 10
+ self.assertEqual(self.test_new_model.wv.syn0.shape, (vocab_size, model_size))
+ self.assertEqual(len(self.test_new_model.wv.vocab), vocab_size, model_size)
+ self.assertEqual(self.test_new_model.wv.syn0_all.shape, (self.test_new_model.num_ngram_vectors, model_size))
+
+ expected_vec_new = [-0.025627,
+ -0.11448,
+ 0.18116,
+ -0.96779,
+ 0.2532,
+ -0.93224,
+ 0.3929,
+ 0.12679,
+ -0.19685,
+ -0.13179] # obtained using ./fasttext print-word-vectors lee_fasttext_new.bin < queries.txt
+
+ self.assertTrue(numpy.allclose(self.test_new_model["hundred"], expected_vec_new, 0.001))
+ self.assertEquals(self.test_new_model.min_count, 5)
+ self.assertEquals(self.test_new_model.window, 5)
+ self.assertEquals(self.test_new_model.iter, 5)
+ self.assertEquals(self.test_new_model.negative, 5)
+ self.assertEquals(self.test_new_model.sample, 0.0001)
+ self.assertEquals(self.test_new_model.bucket, 1000)
+ self.assertEquals(self.test_new_model.wv.max_n, 6)
+ self.assertEquals(self.test_new_model.wv.min_n, 3)
+ self.model_sanity(new_model)
+
def testLoadModelWithNonAsciiVocab(self):
"""Test loading model with non-ascii words in vocab"""
model = fasttext.FastText.load_fasttext_format(datapath('non_ascii_fasttext'))
diff --git a/gensim/test/test_ldamallet_wrapper.py b/gensim/test/test_ldamallet_wrapper.py
index 5367236801..22d31d02e8 100644
--- a/gensim/test/test_ldamallet_wrapper.py
+++ b/gensim/test/test_ldamallet_wrapper.py
@@ -107,12 +107,18 @@ def testMallet2Model(self):
tm1 = ldamallet.LdaMallet(self.mallet_path, corpus=corpus, num_topics=2, id2word=dictionary)
tm2 = ldamallet.malletmodel2ldamodel(tm1)
for document in corpus:
- self.assertEqual(tm1[document][0], tm2[document][0])
- self.assertEqual(tm1[document][1], tm2[document][1])
- logging.debug('%d %d', tm1[document][0], tm2[document][0])
+ element1_1, element1_2 = tm1[document][0]
+ element2_1, element2_2 = tm2[document][0]
+ self.assertAlmostEqual(element1_1, element2_1)
+ self.assertAlmostEqual(element1_2, element2_2, 1)
+ element1_1, element1_2 = tm1[document][1]
+ element2_1, element2_2 = tm2[document][1]
+ self.assertAlmostEqual(element1_1, element2_1)
+ self.assertAlmostEqual(element1_2, element2_2, 1)
+ logging.debug('%d %d', element1_1, element2_1)
+ logging.debug('%d %d', element1_2, element2_2)
logging.debug('%d %d', tm1[document][1], tm2[document][1])
-
def testPersistence(self):
if not self.mallet_path:
return
diff --git a/gensim/test/test_ldamodel.py b/gensim/test/test_ldamodel.py
index c0d13a6ae5..81f29b8e42 100644
--- a/gensim/test/test_ldamodel.py
+++ b/gensim/test/test_ldamodel.py
@@ -46,7 +46,7 @@
def testfile(test_fname=''):
# temporary data will be stored to this file
- fname = 'gensim_models_' + test_fname + '.tst'
+ fname = 'gensim_models_' + test_fname + '.tst'
return os.path.join(tempfile.gettempdir(), fname)
@@ -247,9 +247,9 @@ def testGetDocumentTopics(self):
#Test case to use the get_document_topic function for the corpus
all_topics = model.get_document_topics(self.corpus, per_word_topics=True)
-
+
self.assertEqual(model.state.numdocs, len(corpus))
-
+
for topic in all_topics:
self.assertTrue(isinstance(topic, tuple))
for k, v in topic[0]: # list of doc_topics
@@ -269,9 +269,9 @@ def testGetDocumentTopics(self):
word_phi_count_na = 0
all_topics = model.get_document_topics(self.corpus, minimum_probability=0.8, minimum_phi_value=1.0, per_word_topics=True)
-
+
self.assertEqual(model.state.numdocs, len(corpus))
-
+
for topic in all_topics:
self.assertTrue(isinstance(topic, tuple))
for k, v in topic[0]: # list of doc_topics
@@ -470,6 +470,31 @@ def testLargeMmapCompressed(self):
# test loading the large model arrays with mmap
self.assertRaises(IOError, self.class_.load, fname, mmap='r')
+ def testRandomStateBackwardCompatibility(self):
+ # load a model saved using a pre-0.13.2 version of Gensim
+ pre_0_13_2_fname = datapath('pre_0_13_2_model')
+ model_pre_0_13_2 = self.class_.load(pre_0_13_2_fname)
+
+ # set `num_topics` less than `model_pre_0_13_2.num_topics` so that `model_pre_0_13_2.random_state` is used
+ model_topics = model_pre_0_13_2.print_topics(num_topics=2, num_words=3)
+
+ for i in model_topics:
+ self.assertTrue(isinstance(i[0], int))
+ self.assertTrue(isinstance(i[1], six.string_types))
+
+ # save back the loaded model using a post-0.13.2 version of Gensim
+ post_0_13_2_fname = datapath('post_0_13_2_model')
+ model_pre_0_13_2.save(post_0_13_2_fname)
+
+ # load a model saved using a post-0.13.2 version of Gensim
+ model_post_0_13_2 = self.class_.load(post_0_13_2_fname)
+ model_topics_new = model_post_0_13_2.print_topics(num_topics=2, num_words=3)
+
+ for i in model_topics_new:
+ self.assertTrue(isinstance(i[0], int))
+ self.assertTrue(isinstance(i[1], six.string_types))
+
+
#endclass TestLdaModel
diff --git a/gensim/test/test_ldavowpalwabbit_wrapper.py b/gensim/test/test_ldavowpalwabbit_wrapper.py
index bcf4f035b8..74e3591f4f 100644
--- a/gensim/test/test_ldavowpalwabbit_wrapper.py
+++ b/gensim/test/test_ldavowpalwabbit_wrapper.py
@@ -220,10 +220,12 @@ def testvwmodel2ldamodel(self):
tm1 = LdaVowpalWabbit(vw_path=self.vw_path, corpus=self.corpus, num_topics=2, id2word=self.dictionary)
tm2 = ldavowpalwabbit.vwmodel2ldamodel(tm1)
for document in self.corpus:
- self.assertEqual(tm1[document][0], tm2[document][0])
- self.assertEqual(tm1[document][1], tm2[document][1])
- logging.debug('%d %d', tm1[document][0], tm2[document][0])
- logging.debug('%d %d', tm1[document][1], tm2[document][1])
+ element1_1, element1_2 = tm1[document][0]
+ element2_1, element2_2 = tm2[document][0]
+ self.assertAlmostEqual(element1_1, element2_1)
+ self.assertAlmostEqual(element1_2, element2_2, 5)
+ logging.debug('%d %d', element1_1, element2_1)
+ logging.debug('%d %d', element1_2, element2_2)
if __name__ == '__main__':
diff --git a/gensim/test/test_phrases.py b/gensim/test/test_phrases.py
index ce0b73b65a..ba2cfc7192 100644
--- a/gensim/test/test_phrases.py
+++ b/gensim/test/test_phrases.py
@@ -127,20 +127,31 @@ def testExportPhrases(self):
"""Test Phrases bigram export_phrases functionality."""
bigram = Phrases(sentences, min_count=1, threshold=1)
- # with this setting we should get response_time and graph_minors
- bigram1_seen = False
- bigram2_seen = False
+ seen_bigrams = set()
for phrase, score in bigram.export_phrases(sentences):
- if not bigram1_seen and b'response time' == phrase:
- bigram1_seen = True
- elif not bigram2_seen and b'graph minors' == phrase:
- bigram2_seen = True
- if bigram1_seen and bigram2_seen:
- break
+ seen_bigrams.add(phrase)
+
+ assert seen_bigrams == set([
+ b'response time',
+ b'graph minors',
+ b'human interface'
+ ])
+
+ def test_multiple_bigrams_single_entry(self):
+ """ a single entry should produce multiple bigrams. """
+ bigram = Phrases(sentences, min_count=1, threshold=1)
+
+ seen_bigrams = set()
+
+ test_sentences = [['graph', 'minors', 'survey', 'human', 'interface']]
+ for phrase, score in bigram.export_phrases(test_sentences):
+ seen_bigrams.add(phrase)
- self.assertTrue(bigram1_seen)
- self.assertTrue(bigram2_seen)
+ assert seen_bigrams == set([
+ b'graph minors',
+ b'human interface'
+ ])
def testBadParameters(self):
"""Test the phrases module with bad parameters."""
diff --git a/gensim/test/test_sklearn_integration.py b/gensim/test/test_sklearn_integration.py
index a651a2b055..cb6c514612 100644
--- a/gensim/test/test_sklearn_integration.py
+++ b/gensim/test/test_sklearn_integration.py
@@ -6,27 +6,33 @@
import pickle
from scipy import sparse
-from sklearn.pipeline import Pipeline
-from sklearn.feature_extraction.text import CountVectorizer
-from sklearn.datasets import load_files
-from sklearn import linear_model
+try:
+ from sklearn.pipeline import Pipeline
+ from sklearn.feature_extraction.text import CountVectorizer
+ from sklearn.datasets import load_files
+ from sklearn import linear_model
+except ImportError:
+ raise unittest.SkipTest("Test requires scikit-learn to be installed, which is not available")
+
from gensim.sklearn_integration.sklearn_wrapper_gensim_ldamodel import SklearnWrapperLdaModel
from gensim.sklearn_integration.sklearn_wrapper_gensim_lsimodel import SklearnWrapperLsiModel
from gensim.corpora import Dictionary
from gensim import matutils
-module_path = os.path.dirname(__file__) # needed because sample data files are located in the same folder
+module_path = os.path.dirname(__file__) # needed because sample data files are located in the same folder
datapath = lambda fname: os.path.join(module_path, 'test_data', fname)
-texts = [['complier', 'system', 'computer'],
- ['eulerian', 'node', 'cycle', 'graph', 'tree', 'path'],
- ['graph', 'flow', 'network', 'graph'],
- ['loading', 'computer', 'system'],
- ['user', 'server', 'system'],
- ['tree', 'hamiltonian'],
- ['graph', 'trees'],
- ['computer', 'kernel', 'malfunction', 'computer'],
- ['server', 'system', 'computer']]
+texts = [
+ ['complier', 'system', 'computer'],
+ ['eulerian', 'node', 'cycle', 'graph', 'tree', 'path'],
+ ['graph', 'flow', 'network', 'graph'],
+ ['loading', 'computer', 'system'],
+ ['user', 'server', 'system'],
+ ['tree', 'hamiltonian'],
+ ['graph', 'trees'],
+ ['computer', 'kernel', 'malfunction', 'computer'],
+ ['server', 'system', 'computer'],
+]
dictionary = Dictionary(texts)
corpus = [dictionary.doc2bow(text) for text in texts]
@@ -44,49 +50,49 @@ def testPrintTopic(self):
self.assertTrue(isinstance(k, int))
def testTransform(self):
- texts_new = ['graph','eulerian']
+ texts_new = ['graph', 'eulerian']
bow = self.model.id2word.doc2bow(texts_new)
- X = self.model.transform(bow)
- self.assertTrue(X.shape[0], 1)
- self.assertTrue(X.shape[1], self.model.num_topics)
- texts_new = [['graph','eulerian'],['server', 'flow'], ['path', 'system']]
+ matrix = self.model.transform(bow)
+ self.assertTrue(matrix.shape[0], 1)
+ self.assertTrue(matrix.shape[1], self.model.num_topics)
+ texts_new = [['graph', 'eulerian'], ['server', 'flow'], ['path', 'system']]
bow = []
for i in texts_new:
bow.append(self.model.id2word.doc2bow(i))
- X = self.model.transform(bow)
- self.assertTrue(X.shape[0], 3)
- self.assertTrue(X.shape[1], self.model.num_topics)
+ matrix = self.model.transform(bow)
+ self.assertTrue(matrix.shape[0], 3)
+ self.assertTrue(matrix.shape[1], self.model.num_topics)
def testGetTopicDist(self):
- texts_new = ['graph','eulerian']
+ texts_new = ['graph', 'eulerian']
bow = self.model.id2word.doc2bow(texts_new)
- doc_topics, word_topics, phi_values = self.model.get_topic_dist(bow,per_word_topics=True)
+ doc_topics, word_topics, phi_values = self.model.get_topic_dist(bow, per_word_topics=True)
- for k,v in word_topics:
+ for k, v in word_topics:
self.assertTrue(isinstance(v, list))
self.assertTrue(isinstance(k, int))
- for k,v in doc_topics:
+ for k, v in doc_topics:
self.assertTrue(isinstance(v, float))
self.assertTrue(isinstance(k, int))
- for k,v in phi_values:
+ for k, v in phi_values:
self.assertTrue(isinstance(v, list))
self.assertTrue(isinstance(k, int))
def testPartialFit(self):
for i in range(10):
self.model.partial_fit(X=corpus) # fit against the model again
- doc=list(corpus)[0] # transform only the first document
+ doc = list(corpus)[0] # transform only the first document
transformed = self.model[doc]
transformed_approx = matutils.sparse2full(transformed, 2) # better approximation
- expected=[0.13, 0.87]
+ expected = [0.13, 0.87]
passed = numpy.allclose(sorted(transformed_approx), sorted(expected), atol=1e-1)
self.assertTrue(passed)
def testCSRMatrixConversion(self):
- Arr = numpy.array([[1, 2, 0], [0, 0, 3], [1, 0, 0]])
- sArr = sparse.csr_matrix(Arr)
+ arr = numpy.array([[1, 2, 0], [0, 0, 3], [1, 0, 0]])
+ sarr = sparse.csr_matrix(arr)
newmodel = SklearnWrapperLdaModel(num_topics=2, passes=100)
- newmodel.fit(sArr)
+ newmodel.fit(sarr)
topic = newmodel.print_topics()
for k, v in topic:
self.assertTrue(isinstance(v, six.string_types))
@@ -94,20 +100,21 @@ def testCSRMatrixConversion(self):
def testPipeline(self):
model = SklearnWrapperLdaModel(num_topics=2, passes=10, minimum_probability=0, random_state=numpy.random.seed(0))
- with open(datapath('mini_newsgroup'),'rb') as f:
+ with open(datapath('mini_newsgroup'), 'rb') as f:
compressed_content = f.read()
uncompressed_content = codecs.decode(compressed_content, 'zlib_codec')
cache = pickle.loads(uncompressed_content)
data = cache
- id2word=Dictionary(map(lambda x : x.split(), data.data))
+ id2word = Dictionary(map(lambda x: x.split(), data.data))
corpus = [id2word.doc2bow(i.split()) for i in data.data]
- rand = numpy.random.mtrand.RandomState(1) # set seed for getting same result
- clf=linear_model.LogisticRegression(penalty='l2', C=0.1)
+ numpy.random.mtrand.RandomState(1) # set seed for getting same result
+ clf = linear_model.LogisticRegression(penalty='l2', C=0.1)
text_lda = Pipeline((('features', model,), ('classifier', clf)))
text_lda.fit(corpus, data.target)
score = text_lda.score(corpus, data.target)
self.assertGreater(score, 0.40)
+
class TestSklearnLSIWrapper(unittest.TestCase):
def setUp(self):
self.model = SklearnWrapperLsiModel(id2word=dictionary, num_topics=2)
@@ -120,39 +127,39 @@ def testModelSanity(self):
self.assertTrue(isinstance(k, int))
def testTransform(self):
- texts_new = ['graph','eulerian']
+ texts_new = ['graph', 'eulerian']
bow = self.model.id2word.doc2bow(texts_new)
- X = self.model.transform(bow)
- self.assertTrue(X.shape[0], 1)
- self.assertTrue(X.shape[1], self.model.num_topics)
- texts_new = [['graph','eulerian'],['server', 'flow'], ['path', 'system']]
+ matrix = self.model.transform(bow)
+ self.assertTrue(matrix.shape[0], 1)
+ self.assertTrue(matrix.shape[1], self.model.num_topics)
+ texts_new = [['graph', 'eulerian'], ['server', 'flow'], ['path', 'system']]
bow = []
for i in texts_new:
bow.append(self.model.id2word.doc2bow(i))
- X = self.model.transform(bow)
- self.assertTrue(X.shape[0], 3)
- self.assertTrue(X.shape[1], self.model.num_topics)
+ matrix = self.model.transform(bow)
+ self.assertTrue(matrix.shape[0], 3)
+ self.assertTrue(matrix.shape[1], self.model.num_topics)
def testPartialFit(self):
for i in range(10):
self.model.partial_fit(X=corpus) # fit against the model again
- doc=list(corpus)[0] # transform only the first document
+ doc = list(corpus)[0] # transform only the first document
transformed = self.model[doc]
transformed_approx = matutils.sparse2full(transformed, 2) # better approximation
- expected=[1.39, 0.0]
+ expected = [1.39, 0.0]
passed = numpy.allclose(sorted(transformed_approx), sorted(expected), atol=1e-1)
self.assertTrue(passed)
def testPipeline(self):
model = SklearnWrapperLsiModel(num_topics=2)
- with open(datapath('mini_newsgroup'),'rb') as f:
+ with open(datapath('mini_newsgroup'), 'rb') as f:
compressed_content = f.read()
uncompressed_content = codecs.decode(compressed_content, 'zlib_codec')
cache = pickle.loads(uncompressed_content)
data = cache
- id2word=Dictionary(map(lambda x : x.split(), data.data))
+ id2word = Dictionary(map(lambda x: x.split(), data.data))
corpus = [id2word.doc2bow(i.split()) for i in data.data]
- clf=linear_model.LogisticRegression(penalty='l2', C=0.1)
+ clf = linear_model.LogisticRegression(penalty='l2', C=0.1)
text_lda = Pipeline((('features', model,), ('classifier', clf)))
text_lda.fit(corpus, data.target)
score = text_lda.score(corpus, data.target)
diff --git a/gensim/test/test_tmdiff.py b/gensim/test/test_tmdiff.py
new file mode 100644
index 0000000000..03a639e454
--- /dev/null
+++ b/gensim/test/test_tmdiff.py
@@ -0,0 +1,56 @@
+#!/usr/bin/env python
+# -*- coding: utf-8 -*-
+#
+# Copyright (C) 2016 Radim Rehurek