Keras: AttributeError: 'ProgbarLogger' object has no attribute 'log_values'

[sarancruzer]

_This is my code_

-- coding: utf-8 --

"""
Created on Fri Mar 03 10:07:42 2017

@author: Shreyans
"""

import os
import pickle
import numpy as np
from keras.models import Sequential
import gensim
from keras.layers.recurrent import LSTM, SimpleRNN
from sklearn.model_selection import train_test_split
import theano
theano.config.optimizer = "None"
import sys

sys.setdefaultencoding("ISO-8859-1")

conversation.pickle.encode('utf-8').strip()

with open('chatbot.pickle', 'rb') as f:
(vec_x, vec_y) = pickle.load(f)
print("results")
print(vec_x)
vec_x = np.array(vec_x, dtype=np.object)
vec_y = np.array(vec_y, dtype=np.object)

x_train, x_test, y_train, y_test = train_test_split(vec_x, vec_y, test_size=0.2, random_state=1)

model = Sequential()
model.add(LSTM(output_dim=500, input_shape=x_train.shape[1:], return_sequences=True, init='glorot_normal', inner_init='glorot_normal', activation='sigmoid'))
model.add(LSTM(output_dim=300, input_shape=x_train.shape[1:], return_sequences=True, init='glorot_normal', inner_init='glorot_normal', activation='sigmoid'))
model.add(LSTM(output_dim=300, input_shape=x_train.shape[1:], return_sequences=True, init='glorot_normal', inner_init='glorot_normal', activation='sigmoid'))
model.add(LSTM(output_dim=300, input_shape=x_train.shape[1:], return_sequences=True, init='glorot_normal', inner_init='glorot_normal', activation='sigmoid'))
model.compile(loss='cosine_proximity', optimizer='adam', metrics=['accuracy'])

model.fit(x_train, y_train, nb_epoch=500, validation_data=(x_test, y_test))
model.save('LSTM5000.h5')
predictions = model.predict(x_test)
mod = gensim.models.Word2Vec.load('word2vec.bin')
[mod.most_similar([predictions[10][i]])[0] for i in range(15)]

*error getting *

model.add(LSTM(output_dim=300, input_shape=x_train.shape[1:], return_sequences=True, init='glorot_normal', inner_init='glorot_normal', activation='sigmoid'))
/usr/local/lib/python3.5/dist-packages/keras/models.py:939: UserWarning: The nb_epoch argument in fit has been renamed epochs.
warnings.warn('The nb_epoch argument in fit '
Train on 0 samples, validate on 1 samples
Epoch 1/500
Traceback (most recent call last):
File "chatbotlstmtrain.py", line 54, in
model.fit(x_train, y_train, nb_epoch=500, validation_data=(x_test, y_test))
File "/usr/local/lib/python3.5/dist-packages/keras/models.py", line 960, in fit
validation_steps=validation_steps)
File "/usr/local/lib/python3.5/dist-packages/keras/engine/training.py", line 1657, in fit
validation_steps=validation_steps)
File "/usr/local/lib/python3.5/dist-packages/keras/engine/training.py", line 1233, in _fit_loop
callbacks.on_epoch_end(epoch, epoch_logs)
File "/usr/local/lib/python3.5/dist-packages/keras/callbacks.py", line 73, in on_epoch_end
callback.on_epoch_end(epoch, logs)
File "/usr/local/lib/python3.5/dist-packages/keras/callbacks.py", line 307, in on_epoch_end
self.progbar.update(self.seen, self.log_values, force=True)
AttributeError: 'ProgbarLogger' object has no attribute 'log_values'

[majthehero]

I also have this same problem, sometimes it works, sometimes it doesn't, it doesn't happen on linux for me, only on windows. There is no obvoius cause and I can't replicate this bug at all, it will suddenly appear and then disappear on random code changes.

[bersbersbers]

Here is one way to reproduce this error using degenerate input data:

import numpy as np

import keras.layers
import keras.models
import keras.optimizers

inputs = keras.layers.Input((1, 1))
model = keras.models.Model(inputs=inputs, outputs=inputs)
model.compile(optimizer=keras.optimizers.Adam(), loss='binary_crossentropy')
model.fit(np.empty((0, 1, 1)), np.empty((0, 1, 1)))

Resulting in:

Using TensorFlow backend.
Epoch 1/1
Traceback (most recent call last):
  File "bug.py", line 10, in <module>
    model.fit(np.empty((0, 1, 1)), np.empty((0, 1, 1)))
  File "C:\Program Files\Python36\lib\site-packages\keras\engine\training.py", line 1705, in fit
    validation_steps=validation_steps)
  File "C:\Program Files\Python36\lib\site-packages\keras\engine\training.py", line 1255, in _fit_loop
    callbacks.on_epoch_end(epoch, epoch_logs)
  File "C:\Program Files\Python36\lib\site-packages\keras\callbacks.py", line 77, in on_epoch_end
    callback.on_epoch_end(epoch, logs)
  File "C:\Program Files\Python36\lib\site-packages\keras\callbacks.py", line 339, in on_epoch_end
    self.progbar.update(self.seen, self.log_values)
AttributeError: 'ProgbarLogger' object has no attribute 'log_values'

I am not saying everyone who sees this errors has degenerate input data, but if input data was being tested for degeneracy, this could help separate those cases from potential other ones.

[ZhangXiaoying0116]

将batch_size设置为较小的数字。当您的batch_size设置为比样本集大小更大的值时，会出现此错误。

[Shanlans]

I also met this problem but fixed. I found my data list is empty.

[robertlugg]

I decided to post a link to a comment I made in another issue report. That one was closed, so I don't want it to be missed. I propose a change to keras that I believe will improve usage: https://github.com/keras-team/keras/issues/3657#issuecomment-455375399

[Lsq4q4qq44]

将batch_size设置为较小的数字。当您的batch_size设置为比样本集大小更大的值时，会出现此错误。

没错确实是这个问题 十分感谢

[mrjj]

Please, see PR with fix above.
It fix this and some related bugs and contains tests for regressions and related bugs manifestation. Concerns behind change set are described as well.

[dhanpalrajpurohit]

This type of error occurs when we use a small size dataset and large batch size. So I replace batch size with 4 and it works.

Happy Coding.

[furkan59]

I also met this problem but fixed. I found my data list is empty.

Yes it is like that.