CODEWITHSUNDEEP
python-3.xtensorflowkerastf.keras
Alex Kinman
Alex Kinman

Reputation: 2605

What is the difference between "from keras.models import Sequential" and "from tensorflow.python.keras.models import Sequential"?

I'm using Tensorflow 1.14 and Python 3.5. I got the following error:

UnboundLocalError: local variable 'batch_index' referenced before assignment

The full trace back is:

---------------------------------------------------------------------------
UnboundLocalError                         Traceback (most recent call last)
<timed exec> in <module>

/usr/local/lib/python3.5/dist-packages/keras/engine/training.py in fit(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_freq, max_queue_size, workers, use_multiprocessing, **kwargs)
   1237                                         steps_per_epoch=steps_per_epoch,
   1238                                         validation_steps=validation_steps,
-> 1239                                         validation_freq=validation_freq)
   1240 
   1241     def evaluate(self,

/usr/local/lib/python3.5/dist-packages/keras/engine/training_arrays.py in fit_loop(model, fit_function, fit_inputs, out_labels, batch_size, epochs, verbose, callbacks, val_function, val_inputs, shuffle, initial_epoch, steps_per_epoch, validation_steps, validation_freq)
    203                     break
    204 
--> 205             if batch_index == len(batches) - 1:  # Last batch.
    206                 if do_validation and should_run_validation(validation_freq, epoch):
    207                     val_outs = test_loop(model, val_function, val_inputs,

UnboundLocalError: local variable 'batch_index' referenced before assignment

And after trying multiple suggestions from different SO answers, I managed to fix it by switching from these import statements: 

from keras.layers import LSTM, Dense
from keras.models import Sequential

To these import statements: 

from tensorflow.python.keras.layers import LSTM, Dense
from tensorflow.python.keras.models import Sequential

This did resolve my issue, but I am puzzled: How are the two any different?

Are tf.keras and keras using different methods and classes?

Upvotes: 1

Views: 9748

Answers (1)

stephen_mugisha
stephen_mugisha

Reputation: 897

Difference between tf.keras and keras.

  • Keras: Is a high level neural network API for training neural networks. It's independent of tensorflow and can run on top of multiple backends such as tensorflow, Theano and CNTK. Documentation here

  • tf.keras: tf.keras is a specific high level implementation of the keras API in tensorflow with added support for certain tensorflow features.

Upvotes: 3

Related Questions