I was following the MNIST sample trying to replicate the codes from the lecture. However, I was getting some warnings and errors in different areas (not sure if they are related) and I am unable to complete the exercise.
First warning appears when running: mnist_dataset, mnist_info = tfds.load(name=’mnist’, with_info=True, as_supervised=True)
WARNING:tensorflow:Entity <bound method TopLevelFeature.decode_example of FeaturesDict({
‘image’: Image(shape=(28, 28, 1), dtype=tf.uint8),
‘label’: ClassLabel(shape=(), dtype=tf.int64, num_classes=10),
})> could not be transformed and will be executed as-is. Please report this to the AutoGraph team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output. Cause: Bad argument number for Name: 3, expecting 4
Then when I tried to fit the model using: model.fit(train_data, epochs=NUM_EPOCHS, validation_data=(validation_inputs, validation_targets), verbose=2)
Here’s the error: ValueError Traceback (most recent call last)
<ipython-input-96-0dc18ef0284d> in <module>
—-> 1 model.fit(train_data, epochs=NUM_EPOCHS, validation_data=(validation_inputs, validation_targets), verbose=2)
~\Anaconda3\envs\py3-TF2.0\lib\site-packages\tensorflow_core\python\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)
732 max_queue_size=max_queue_size,
733 workers=workers,
–> 734 use_multiprocessing=use_multiprocessing)
735
736 def evaluate(self,
~\Anaconda3\envs\py3-TF2.0\lib\site-packages\tensorflow_core\python\keras\engine\training_v2.py in fit(self, model, 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, **kwargs)
222 validation_data=validation_data,
223 validation_steps=validation_steps,
–> 224 distribution_strategy=strategy)
225
226 total_samples = _get_total_number_of_samples(training_data_adapter)
~\Anaconda3\envs\py3-TF2.0\lib\site-packages\tensorflow_core\python\keras\engine\training_v2.py in _process_training_inputs(model, x, y, batch_size, epochs, sample_weights, class_weights, steps_per_epoch, validation_split, validation_data, validation_steps, shuffle, distribution_strategy, max_queue_size, workers, use_multiprocessing)
561 class_weights=class_weights,
562 steps=validation_steps,
–> 563 distribution_strategy=distribution_strategy)
564 elif validation_steps:
565 raise ValueError(‘`validation_steps` should not be specified if ‘
~\Anaconda3\envs\py3-TF2.0\lib\site-packages\tensorflow_core\python\keras\engine\training_v2.py in _process_inputs(model, x, y, batch_size, epochs, sample_weights, class_weights, shuffle, steps, distribution_strategy, max_queue_size, workers, use_multiprocessing)
603 max_queue_size=max_queue_size,
604 workers=workers,
–> 605 use_multiprocessing=use_multiprocessing)
606 # As a fallback for the data type that does not work with
607 # _standardize_user_data, use the _prepare_model_with_inputs.
~\Anaconda3\envs\py3-TF2.0\lib\site-packages\tensorflow_core\python\keras\engine\data_adapter.py in __init__(self, x, y, sample_weights, batch_size, epochs, steps, shuffle, **kwargs)
239 if not batch_size:
240 raise ValueError(
–> 241 “`batch_size` or `steps` is required for `Tensor` or `NumPy`”
242 ” input data.”)
243
ValueError: `batch_size` or `steps` is required for `Tensor` or `NumPy` input data.
It looks like the provided solution (link:https://365datascience.teachable.com/courses/614390/lectures/10998874) was missing an argument referencing this line of code:
model.fit(train_data, epochs=NUM_EPOCHS, validation_data=(validation_inputs, validation_targets), verbose=2)
change to (by adding validation_steps=1) will do the trick
model.fit(train_data, epochs=NUM_EPOCHS, validation_data=(validation_inputs, validation_targets),validation_steps=1, verbose=2)
Well, Weiwei, you answered the question before we could jump in it seems! Best, The 365 Team
Hi. After doing this correction shown above, I get an error: TypeError: ‘DatasetV1Adapter’ object is not subscriptable
Hi Kari, Please update your tensorflow to the latest version! Best, The 365 Team
Performed a google search and was able to address the first issue if anyone runs into the same.
Warning mnist_dataset, mnist_info = tfds.load(name=’mnist’, with_info=True, as_supervised=True)
Installing gast 0.2.2 version fixed it.
pip install –user gast==0.2.2
Have the same problem… But really disappointed with the answer. Setting validations_steps=1 doesn’t reproduce the results of the video, and the validation accuracy is almost 100% in every step. So it doesn’t fix the problem…
Same here the validation accuracy is too high
My model was
model = Sequential()
model.add(Conv2D(32, kernel_size=(3, 3),activation='relu',input_shape=(28,28,1)))
model.add(Conv2D(64, (3, 3), activation='relu'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(128, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(10, activation='softmax'))
I just changed the
input_shape=(28,28,1)
to
input_shape=(28,28,3)
and it resolved the error