Bug with papermill read notebook

SETUP.md

@@ -120,18 +120,21 @@ You also need to find where Spark is installed and set `SPARK_HOME` variable, on
 
 Then, create the file `$RECO_ENV/etc/conda/activate.d/env_vars.sh` and add:
 
-    #!/bin/sh
-    RECO_ENV=$(conda env list | grep reco_pyspark | awk '{print $NF}')
-    export PYSPARK_PYTHON=$RECO_ENV/bin/python
-    export PYSPARK_DRIVER_PYTHON=$RECO_ENV/bin/python
-    export SPARK_HOME=/dsvm/tools/spark/current
+```bash
+#!/bin/sh
+RECO_ENV=$(conda env list | grep reco_pyspark | awk '{print $NF}')
+export PYSPARK_PYTHON=$RECO_ENV/bin/python
+export PYSPARK_DRIVER_PYTHON=$RECO_ENV/bin/python
+export SPARK_HOME=/dsvm/tools/spark/current
+```
 
 This will export the variables every time we do `conda activate reco_pyspark`. To unset these variables when we deactivate the environment, create the file `$RECO_ENV/etc/conda/deactivate.d/env_vars.sh` and add:
 
-    #!/bin/sh
-    unset PYSPARK_PYTHON
-    unset PYSPARK_DRIVER_PYTHON
-
+```bash
+#!/bin/sh
+unset PYSPARK_PYTHON
+unset PYSPARK_DRIVER_PYTHON
+```
 
 </details>
 

examples/00_quick_start/sar_movielens.ipynb

@@ -455,13 +455,9 @@
  "metadata": {
   "celltoolbar": "Tags",
   "kernelspec": {
-   "display_name": "Python 3.6.12 64-bit ('sb_full': conda)",
-   "metadata": {
-    "interpreter": {
-     "hash": "f28711ae1fad89778b64817fc2d746effb845deda73edae96b2473c20b2d4f70"
-    }
-   },
-   "name": "python3"
+   "display_name": "Python (reco_base)",
+   "language": "python",
+   "name": "reco_base"
   },
   "language_info": {
    "codemirror_mode": {
@@ -473,7 +469,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.12-final"
+   "version": "3.6.11"
   }
  },
  "nbformat": 4,

examples/01_prepare_data/mind_utils.ipynb

@@ -46,14 +46,12 @@
     "import sys\n",
     "sys.path.append(\"../../\")\n",
     "import os\n",
-    "import papermill as pm\n",
     "import pandas as pd\n",
     "from collections import Counter\n",
     "from tqdm import tqdm\n",
     "import pickle\n",
     "import numpy as np\n",
-    "\n",
-    "print(\"System version: {}\".format(sys.version))\n",
+    "import scrapbook as sb\n",
     "\n",
     "from tempfile import TemporaryDirectory\n",
     "from reco_utils.dataset.mind import (download_mind,\n",
@@ -62,7 +60,9 @@
     "                                     load_glove_matrix,\n",
     "                                     word_tokenize\n",
     "                                    )\n",
-    "from reco_utils.dataset.download_utils import unzip_file"
+    "from reco_utils.dataset.download_utils import unzip_file\n",
+    "\n",
+    "print(\"System version: {}\".format(sys.version))\n"
    ]
   },
   {
@@ -418,37 +418,11 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 17,
+   "execution_count": null,
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "/Users/apple/miniconda/envs/reco_base/lib/python3.6/site-packages/ipykernel_launcher.py:1: DeprecationWarning: Function record is deprecated and will be removed in verison 1.0.0 (current version 0.19.1). Please see `scrapbook.glue` (nteract-scrapbook) as a replacement for this functionality.\n",
-      "  \"\"\"Entry point for launching an IPython kernel.\n"
-     ]
-    },
-    {
-     "data": {
-      "application/papermill.record+json": {
-       "utils_state": {
-        "embedding_exist_num": 22408,
-        "embedding_exist_num_all": 37634,
-        "subvert_num": 17,
-        "uid2index": 5000,
-        "vert_num": 17,
-        "word_num": 23404,
-        "word_num_all": 41074
-       }
-      }
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
+   "outputs": [],
    "source": [
-    "pm.record(\"utils_state\", utils_state)"
+    "sb.glue(\"utils_state\", utils_state)"
    ]
   }
  ],

examples/template.ipynb

@@ -259,9 +259,9 @@
  "metadata": {
   "celltoolbar": "Tags",
   "kernelspec": {
-   "display_name": "Python (recommender)",
+   "display_name": "Python (reco_base)",
    "language": "python",
-   "name": "recommender"
+   "name": "reco_base"
   },
   "language_info": {
    "codemirror_mode": {
@@ -273,9 +273,9 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.0"
+   "version": "3.6.11"
   }
  },
  "nbformat": 4,
  "nbformat_minor": 2
-}
+}

tests/integration/test_mind.py

@@ -1,9 +1,13 @@
 # Copyright (c) Microsoft Corporation. All rights reserved.
 # Licensed under the MIT License.
 
-import pytest
 import os
+import pytest
+import papermill as pm
+import scrapbook as sb
+
 from reco_utils.dataset.mind import download_mind, extract_mind
+from tests.notebooks_common import OUTPUT_NOTEBOOK, KERNEL_NAME
 
 
 @pytest.mark.integration
@@ -37,3 +41,25 @@ def test_extract_mind(tmp):
     assert statinfo.st_size == 59055351
     statinfo = os.stat(os.path.join(valid_path, "relation_embedding.vec"))
     assert statinfo.st_size == 1044588
+
+
+@pytest.mark.integration
+def test_mind_utils_integration(notebooks, tmp):
+    notebook_path = notebooks["mind_utils"]
+    pm.execute_notebook(
+        notebook_path,
+        OUTPUT_NOTEBOOK,
+        kernel_name=KERNEL_NAME,
+        parameters=dict(mind_type="small", word_embedding_dim=300),
+    )
+    results = sb.read_notebook(OUTPUT_NOTEBOOK).scraps.dataframe.set_index("name")[
+        "data"
+    ]
+
+    assert results["utils_state"]["vert_num"] == 17
+    assert results["utils_state"]["subvert_num"] == 17
+    assert results["utils_state"]["word_num"] == 23404
+    assert results["utils_state"]["word_num_all"] == 41074
+    assert results["utils_state"]["embedding_exist_num"] == 22408
+    assert results["utils_state"]["embedding_exist_num_all"] == 37634
+    assert results["utils_state"]["uid2index"] == 5000

tests/integration/test_notebooks_gpu.py

@@ -1,12 +1,13 @@
 # Copyright (c) Microsoft Corporation. All rights reserved.
 # Licensed under the MIT License.
 
-import papermill as pm
+import os
 import pytest
+import papermill as pm
+import scrapbook as sb
 
 from reco_utils.common.gpu_utils import get_number_gpus
 from tests.notebooks_common import OUTPUT_NOTEBOOK, KERNEL_NAME
-import os
 
 
 TOL = 0.5
@@ -48,7 +49,9 @@ def test_ncf_integration(notebooks, size, epochs, expected_values, seed):
             TOP_K=10, MOVIELENS_DATA_SIZE=size, EPOCHS=epochs, BATCH_SIZE=512, SEED=seed
         ),
     )
-    results = pm.read_notebook(OUTPUT_NOTEBOOK).dataframe.set_index("name")["value"]
+    results = sb.read_notebook(OUTPUT_NOTEBOOK).scraps.dataframe.set_index("name")[
+        "data"
+    ]
 
     for key, value in expected_values.items():
         assert results[key] == pytest.approx(value, rel=TOL, abs=ABS_TOL)
@@ -93,7 +96,9 @@ def test_ncf_deep_dive_integration(
             SEED=seed,
         ),
     )
-    results = pm.read_notebook(OUTPUT_NOTEBOOK).dataframe.set_index("name")["value"]
+    results = sb.read_notebook(OUTPUT_NOTEBOOK).scraps.dataframe.set_index("name")[
+        "data"
+    ]
 
     for key, value in expected_values.items():
         assert results[key] == pytest.approx(value, rel=TOL, abs=ABS_TOL)
@@ -123,14 +128,15 @@ def test_ncf_deep_dive_integration(
 )
 def test_fastai_integration(notebooks, size, epochs, expected_values):
     notebook_path = notebooks["fastai"]
-    pm.execute_notebook(notebook_path, OUTPUT_NOTEBOOK, kernel_name=KERNEL_NAME)
     pm.execute_notebook(
         notebook_path,
         OUTPUT_NOTEBOOK,
         kernel_name=KERNEL_NAME,
         parameters=dict(TOP_K=10, MOVIELENS_DATA_SIZE=size, EPOCHS=epochs),
     )
-    results = pm.read_notebook(OUTPUT_NOTEBOOK).dataframe.set_index("name")["value"]
+    results = sb.read_notebook(OUTPUT_NOTEBOOK).scraps.dataframe.set_index("name")[
+        "data"
+    ]
 
     for key, value in expected_values.items():
         assert results[key] == pytest.approx(value, rel=TOL, abs=ABS_TOL)
@@ -168,7 +174,9 @@ def test_xdeepfm_integration(
             RANDOM_SEED=seed,
         ),
     )
-    results = pm.read_notebook(OUTPUT_NOTEBOOK).dataframe.set_index("name")["value"]
+    results = sb.read_notebook(OUTPUT_NOTEBOOK).scraps.dataframe.set_index("name")[
+        "data"
+    ]
 
     for key, value in expected_values.items():
         assert results[key]["auc"] == pytest.approx(value["auc"], rel=TOL, abs=ABS_TOL)
@@ -215,7 +223,10 @@ def test_wide_deep_integration(notebooks, size, steps, expected_values, seed, tm
     pm.execute_notebook(
         notebook_path, OUTPUT_NOTEBOOK, kernel_name=KERNEL_NAME, parameters=params
     )
-    results = pm.read_notebook(OUTPUT_NOTEBOOK).dataframe.set_index("name")["value"]
+    results = sb.read_notebook(OUTPUT_NOTEBOOK).scraps.dataframe.set_index("name")[
+        "data"
+    ]
+
     for key, value in expected_values.items():
         assert results[key] == pytest.approx(value, rel=TOL, abs=ABS_TOL)
 
@@ -250,7 +261,10 @@ def test_slirec_quickstart_integration(
     pm.execute_notebook(
         notebook_path, OUTPUT_NOTEBOOK, kernel_name=KERNEL_NAME, parameters=params
     )
-    results = pm.read_notebook(OUTPUT_NOTEBOOK).dataframe.set_index("name")["value"]
+    results = sb.read_notebook(OUTPUT_NOTEBOOK).scraps.dataframe.set_index("name")[
+        "data"
+    ]
+
     for key, value in expected_values.items():
         assert results[key]["auc"] == pytest.approx(value["auc"], rel=TOL, abs=ABS_TOL)
         assert results[key]["logloss"] == pytest.approx(
@@ -278,14 +292,19 @@ def test_slirec_quickstart_integration(
         )
     ],
 )
-def test_nrms_quickstart_integration(notebooks, epochs, seed, MIND_type, expected_values):
+def test_nrms_quickstart_integration(
+    notebooks, epochs, seed, MIND_type, expected_values
+):
     notebook_path = notebooks["nrms_quickstart"]
 
     params = {"epochs": epochs, "seed": seed, "MIND_type": MIND_type}
     pm.execute_notebook(
         notebook_path, OUTPUT_NOTEBOOK, kernel_name=KERNEL_NAME, parameters=params
     )
-    results = pm.read_notebook(OUTPUT_NOTEBOOK).dataframe.set_index("name")["value"]
+    results = sb.read_notebook(OUTPUT_NOTEBOOK).scraps.dataframe.set_index("name")[
+        "data"
+    ]
+
     for key, value in expected_values.items():
         assert results[key]["group_auc"] == pytest.approx(
             value["group_auc"], rel=TOL, abs=ABS_TOL
@@ -321,14 +340,19 @@ def test_nrms_quickstart_integration(notebooks, epochs, seed, MIND_type, expecte
         )
     ],
 )
-def test_naml_quickstart_integration(notebooks, epochs, seed, MIND_type, expected_values):
+def test_naml_quickstart_integration(
+    notebooks, epochs, seed, MIND_type, expected_values
+):
     notebook_path = notebooks["naml_quickstart"]
 
     params = {"epochs": epochs, "seed": seed, "MIND_type": MIND_type}
     pm.execute_notebook(
         notebook_path, OUTPUT_NOTEBOOK, kernel_name=KERNEL_NAME, parameters=params
     )
-    results = pm.read_notebook(OUTPUT_NOTEBOOK).dataframe.set_index("name")["value"]
+    results = sb.read_notebook(OUTPUT_NOTEBOOK).scraps.dataframe.set_index("name")[
+        "data"
+    ]
+
     for key, value in expected_values.items():
         assert results[key]["group_auc"] == pytest.approx(
             value["group_auc"], rel=TOL, abs=ABS_TOL
@@ -364,14 +388,19 @@ def test_naml_quickstart_integration(notebooks, epochs, seed, MIND_type, expecte
         )
     ],
 )
-def test_lstur_quickstart_integration(notebooks, epochs, seed, MIND_type, expected_values):
+def test_lstur_quickstart_integration(
+    notebooks, epochs, seed, MIND_type, expected_values
+):
     notebook_path = notebooks["lstur_quickstart"]
 
     params = {"epochs": epochs, "seed": seed, "MIND_type": MIND_type}
     pm.execute_notebook(
         notebook_path, OUTPUT_NOTEBOOK, kernel_name=KERNEL_NAME, parameters=params
     )
-    results = pm.read_notebook(OUTPUT_NOTEBOOK).dataframe.set_index("name")["value"]
+    results = sb.read_notebook(OUTPUT_NOTEBOOK).scraps.dataframe.set_index("name")[
+        "data"
+    ]
+
     for key, value in expected_values.items():
         assert results[key]["group_auc"] == pytest.approx(
             value["group_auc"], rel=TOL, abs=ABS_TOL
@@ -407,14 +436,19 @@ def test_lstur_quickstart_integration(notebooks, epochs, seed, MIND_type, expect
         )
     ],
 )
-def test_npa_quickstart_integration(notebooks, epochs, seed, MIND_type, expected_values):
+def test_npa_quickstart_integration(
+    notebooks, epochs, seed, MIND_type, expected_values
+):
     notebook_path = notebooks["npa_quickstart"]
 
     params = {"epochs": epochs, "seed": seed, "MIND_type": MIND_type}
     pm.execute_notebook(
         notebook_path, OUTPUT_NOTEBOOK, kernel_name=KERNEL_NAME, parameters=params
     )
-    results = pm.read_notebook(OUTPUT_NOTEBOOK).dataframe.set_index("name")["value"]
+    results = sb.read_notebook(OUTPUT_NOTEBOOK).scraps.dataframe.set_index("name")[
+        "data"
+    ]
+
     for key, value in expected_values.items():
         assert results[key]["group_auc"] == pytest.approx(
             value["group_auc"], rel=TOL, abs=ABS_TOL
@@ -470,7 +504,9 @@ def test_lightgcn_deep_dive_integration(
             item_file=os.path.join(data_path, r"item_embeddings"),
         ),
     )
-    results = pm.read_notebook(OUTPUT_NOTEBOOK).dataframe.set_index("name")["value"]
+    results = sb.read_notebook(OUTPUT_NOTEBOOK).scraps.dataframe.set_index("name")[
+        "data"
+    ]
 
     for key, value in expected_values.items():
         assert results[key] == pytest.approx(value, rel=TOL, abs=ABS_TOL)
@@ -486,7 +522,9 @@ def test_dkn_quickstart_integration(notebooks):
         kernel_name=KERNEL_NAME,
         parameters=dict(epochs=5, batch_size=500),
     )
-    results = pm.read_notebook(OUTPUT_NOTEBOOK).dataframe.set_index("name")["value"]
+    results = sb.read_notebook(OUTPUT_NOTEBOOK).scraps.dataframe.set_index("name")[
+        "data"
+    ]
 
     assert results["res"]["auc"] == pytest.approx(0.5651, rel=TOL, abs=ABS_TOL)
     assert results["res"]["mean_mrr"] == pytest.approx(0.1639, rel=TOL, abs=ABS_TOL)