TypeError: _standardize_user_data() got an unexpected keyword argument 'check_batch_axis'

[kaeflint]

I trained one of the example models as shown below but I had the error:

TypeError: _standardize_user_data() got an unexpected keyword argument 'check_batch_axis'

from future import print_function
import keras
from keras.datasets import mnist
from keras.models import Sequential
from keras.layers import Dense, Dropout
from keras.optimizers import RMSprop

from importance_sampling.training import ImportanceTraining

batch_size = 128
num_classes = 10
epochs = 3

the data, shuffled and split between train and test sets

(x_train, y_train), (x_test, y_test) = mnist.load_data()

x_train = x_train.reshape(60000, 784)
x_test = x_test.reshape(10000, 784)
x_train = x_train.astype('float32')
x_test = x_test.astype('float32')
x_train /= 255
x_test /= 255
print(x_train.shape[0], 'train samples')
print(x_test.shape[0], 'test samples')

convert class vectors to binary class matrices

y_train = keras.utils.to_categorical(y_train, num_classes)
y_test = keras.utils.to_categorical(y_test, num_classes)

model = Sequential()
model.add(Dense(512, activation='relu', input_shape=(784,)))
model.add(Dropout(0.2))
model.add(Dense(512, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(10, activation='softmax'))

model.summary()

model.compile(loss='categorical_crossentropy',
optimizer=RMSprop(),
metrics=['accuracy'])

history = ImportanceTraining(model, forward_batch_size=1024).fit(
x_train, y_train,
batch_size=batch_size,
epochs=epochs,
verbose=1
)
score = model.evaluate(x_test, y_test, verbose=0)
print('Test loss:', score[0])
print('Test accuracy:', score[1])

Had this error:

TypeError Traceback (most recent call last)
in ()
47 batch_size=batch_size,
48 epochs=epochs,
---> 49 verbose=1
50 )
51 score = model.evaluate(x_test, y_test, verbose=0)

/content/.local/lib/python3.6/site-packages/importance_sampling/training.py in fit(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, steps_per_epoch)
97 steps_per_epoch=steps_per_epoch,
98 verbose=verbose,
---> 99 callbacks=callbacks
100 )
101

/content/.local/lib/python3.6/site-packages/importance_sampling/training.py in fit_dataset(self, dataset, steps_per_epoch, batch_size, epochs, verbose, callbacks)
208
209 # Importance sampling is done here
--> 210 idxs, (x, y), w = sampler.sample(batch_size)
211 # Train on the sampled data
212 loss, metrics, scores = self.model.train_batch(x, y, w)

/content/.local/lib/python3.6/site-packages/importance_sampling/samplers.py in sample(self, batch_size)
60 def sample(self, batch_size):
61 # Get the importance scores of some samples
---> 62 idxs1, scores, xy = self._get_samples_with_scores(batch_size)
63
64 # Sample from the available ones

/content/.local/lib/python3.6/site-packages/importance_sampling/samplers.py in _get_samples_with_scores(self, batch_size)
126 idxs = np.random.choice(self.N, self.large_batch)
127 x, y = self.dataset.train_data[idxs]
--> 128 scores = self.model.score(x, y, batch_size=self.forward_batch_size)
129
130 return (

/content/.local/lib/python3.6/site-packages/importance_sampling/model_wrappers.py in score(self, x, y, batch_size)
93 result = np.hstack([
94 self.score_batch(xi, yi).T
---> 95 for xi, yi in self._iterate_batches(x, y, batch_size)
96 ]).T
97

/content/.local/lib/python3.6/site-packages/importance_sampling/model_wrappers.py in (.0)
93 result = np.hstack([
94 self.score_batch(xi, yi).T
---> 95 for xi, yi in self._iterate_batches(x, y, batch_size)
96 ]).T
97

/content/.local/lib/python3.6/site-packages/importance_sampling/model_wrappers.py in score_batch(self, x, y)
228 dummy_target = np.zeros((y.shape[0], 1))
229 inputs = _tolist(x) + [y, dummy_weights]
--> 230 outputs = self.model.test_on_batch(inputs, dummy_target)
231
232 return outputs[self.SCORE].ravel()

/content/.local/lib/python3.6/site-packages/transparent_keras/transparent_model.py in test_on_batch(self, x, y, sample_weight)
187 x, y,
188 sample_weight=sample_weight,
--> 189 check_batch_axis=True
190 )
191

TypeError: _standardize_user_data() got an unexpected keyword argument 'check_batch_axis'

[angeloskath]

Hey, really sorry for the late reply.

I am preparing a new release that changes quite a lot of things and fixes this bug as well. I will be pushing to the repository today but the release will probably be published after Easter because I have to update the documentation as well.

Thanks for your patience and support!

[angeloskath]

Hi,

there is a new release with a lot of new stuff that also fixes this issue. Reopen if you have any problems.

Cheers,
Angelos