How to generate output?

[GenTxt]

Hello:

Your blog post was very interesting. I've converted your notebooks to python scripts and the training portion runs fine. I would like to generate output but I can't figure out how (new to keras).

Many non-keras repos use separate generation scripts which makes the process easier.

Thanks

[cpastorius101]

I'm also curious how to export the output... it takes so long to run to the script, it's hard to continuously trial and error

[GenTxt]

https://github.com/DavidG1011/LSTM-Text-Generator Check the above repo which is 4 commits ahead and includes code snippet for text generation. Haven't tested yet.

…

On Thu, May 16, 2019 at 12:29 AM cpastorius101 ***@***.***> wrote: I'm also curious how to export the output... it takes so long to run to the script, it's hard to continuously trial and error — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#1?email_source=notifications&email_token=AFMAWPJ55K3E7GTJOA6YXX3PVTPMFA5CNFSM4EZEW6KKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODVQT42Y#issuecomment-492912235>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AFMAWPNZGNB7TBEEAJBWQ4DPVTPMFANCNFSM4EZEW6KA> .

[db-ctrl]

I am also curious how to generate the text output. RNN's seem to be training okay. Just no output at the end of the run

[gtlsgamr]

Hey there, this script can generate output, just replace the model definition with the pre-trained model you are using.

# Load LSTM network and generate text
import sys
import numpy
from keras.models import Sequential
from keras.layers import Dense
from keras.layers import Dropout
from keras.layers import LSTM
from keras.callbacks import ModelCheckpoint
from keras.utils import np_utils
import os

# change working directory to the script's directory
abspath = os.path.abspath(__file__)
dname = os.path.dirname(abspath)
os.chdir(dname)

# load ascii text and covert to lowercase
filename = "sonnets.txt"
raw_text = open(filename, 'r', encoding='utf-8').read()
raw_text = raw_text.lower()

# create mapping of unique chars to integers, and a reverse mapping
chars = sorted(list(set(raw_text)))
char_to_int = dict((c, i) for i, c in enumerate(chars))
int_to_char = dict((i, c) for i, c in enumerate(chars))
# summarize the loaded data
n_chars = len(raw_text)
n_vocab = len(chars)
print ("Total Characters: ", n_chars)
print ("Total Vocab: ", n_vocab)
# prepare the dataset of input to output pairs encoded as integers
seq_length = 100
dataX = []
dataY = []
for i in range(0, n_chars - seq_length, 1):
	seq_in = raw_text[i:i + seq_length]
	seq_out = raw_text[i + seq_length]
	dataX.append([char_to_int[char] for char in seq_in])
	dataY.append(char_to_int[seq_out])
n_patterns = len(dataX)
print ("Total Patterns: ", n_patterns)
# reshape X to be [samples, time steps, features]
X = numpy.reshape(dataX, (n_patterns, seq_length, 1))
# normalize
X = X / float(n_vocab)
# one hot encode the output variable
y = np_utils.to_categorical(dataY)
# define the LSTM model
model = Sequential()
model.add(LSTM(400, input_shape=(X.shape[1], X.shape[2]), return_sequences=True))
model.add(Dropout(0.2))
model.add(LSTM(400))
model.add(Dropout(0.2))
model.add(Dense(y.shape[1], activation='softmax'))

# load the network weights
filename = "models/text_generator_400_0.2_400_0.2_100.h5"
model.load_weights(filename)
model.compile(loss='categorical_crossentropy', optimizer='adam')
# pick a random seed
start = numpy.random.randint(0, len(dataX)-1)
pattern = dataX[start]
print ("Seed:")
print ("\"", ''.join([int_to_char[value] for value in pattern]), "\"")
# generate characters
for i in range(1000):
	x = numpy.reshape(pattern, (1, len(pattern), 1))
	x = x / float(n_vocab)
	prediction = model.predict(x, verbose=0)
	index = numpy.argmax(prediction)
	result = int_to_char[index]
	seq_in = [int_to_char[value] for value in pattern]
	sys.stdout.write(result)
	pattern.append(index)
	pattern = pattern[1:len(pattern)]
print ("\nDone.")

[salmanke]

X_modified = np.reshape(X, (len(X), seq_length, 1))
X_modified = X_modified / float(len(characters))
Y_modified = np_utils.to_categorical(Y)

in this line of i am getting a error of "name 'np_utils' is not defined"
can anyone help me with this..?