I'm having a dimension error when I'm training a variational autoencoder and I can't figure out what I'm doing wrong. I had a dimension error in a neural network that just used Dense layers, but I solved it by adding the Flatten layer. This error can't be solved that way.
The dataset I'm using is CIFAR10.
Here's my code, along with its outputs:
(x_train, y_train), (x_test, y_test) = tf.keras.datasets.cifar10.load_data()
class Sampling(keras.layers.Layer):
def call(self, inputs):
mean, log_var = inputs
return K.random_normal(tf.shape(log_var)) * K.exp(log_var / 2) + mean
tf.random.set_seed(42)
np.random.seed(42)
codings_size = 10
inputs = keras.layers.Input(shape=(32, 32))
z = keras.layers.Flatten()(inputs)
z = keras.layers.Dense(150, activation="selu")(z)
z = keras.layers.Dense(100, activation="selu")(z)
codings_mean = keras.layers.Dense(codings_size)(z)
codings_log_var = keras.layers.Dense(codings_size)(z)
codings = Sampling()([codings_mean, codings_log_var])
variational_encoder = keras.models.Model(
inputs=[inputs], outputs=[codings_mean, codings_log_var, codings])
decoder_inputs = keras.layers.Input(shape=[codings_size])
x = keras.layers.Dense(100, activation="selu")(decoder_inputs)
x = keras.layers.Dense(150, activation="selu")(x)
x = keras.layers.Dense(28 * 28, activation="sigmoid")(x)
outputs = keras.layers.Reshape([28, 28])(x)
variational_decoder = keras.models.Model(inputs=[decoder_inputs], outputs=[outputs])
_, _, codings = variational_encoder(inputs)
reconstructions = variational_decoder(codings)
variational_ae = keras.models.Model(inputs=[inputs], outputs=[reconstructions])
variational_ae.summary()
latent_loss = -0.5 * K.sum(
1 + codings_log_var - K.exp(codings_log_var) - K.square(codings_mean),
axis=-1)
variational_ae.add_loss(K.mean(latent_loss) / 784.)
variational_ae.compile(loss="binary_crossentropy", optimizer="rmsprop", metrics=[rounded_accuracy])
history = variational_ae.fit(x_train, x_train, epochs=25, batch_size=128,
validation_data=(x_test, x_test))
Model Summary:
Model: "model_20"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
input_13 (InputLayer) [(None, 32, 32)] 0
_________________________________________________________________
model_18 (Model) [(None, 10), (None, 10), 170870
_________________________________________________________________
model_19 (Model) (None, 28, 28) 134634
=================================================================
Total params: 305,504
Trainable params: 305,504
Non-trainable params: 0
_________________________________________________________________
Traceback:
ValueError Traceback (most recent call last)
<ipython-input-163-310c1702abb5> in <module>
33 variational_ae.compile(loss="binary_crossentropy", optimizer="rmsprop", metrics=[rounded_accuracy])
34 history = variational_ae.fit(x_train, x_train, epochs=25, batch_size=128,
---> 35 validation_data=(x_test, x_test))
~/anaconda3/envs/tf2/lib/python3.7/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)
817 max_queue_size=max_queue_size,
818 workers=workers,
--> 819 use_multiprocessing=use_multiprocessing)
820
821 def evaluate(self,
~/anaconda3/envs/tf2/lib/python3.7/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, max_queue_size, workers, use_multiprocessing, **kwargs)
233 max_queue_size=max_queue_size,
234 workers=workers,
--> 235 use_multiprocessing=use_multiprocessing)
236
237 total_samples = _get_total_number_of_samples(training_data_adapter)
~/anaconda3/envs/tf2/lib/python3.7/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)
591 max_queue_size=max_queue_size,
592 workers=workers,
--> 593 use_multiprocessing=use_multiprocessing)
594 val_adapter = None
595 if validation_data:
~/anaconda3/envs/tf2/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/training_v2.py in _process_inputs(model, mode, x, y, batch_size, epochs, sample_weights, class_weights, shuffle, steps, distribution_strategy, max_queue_size, workers, use_multiprocessing)
644 standardize_function = None
645 x, y, sample_weights = standardize(
--> 646 x, y, sample_weight=sample_weights)
647 elif adapter_cls is data_adapter.ListsOfScalarsDataAdapter:
648 standardize_function = standardize
~/anaconda3/envs/tf2/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/training.py in _standardize_user_data(self, x, y, sample_weight, class_weight, batch_size, check_steps, steps_name, steps, validation_split, shuffle, extract_tensors_from_dataset)
2381 is_dataset=is_dataset,
2382 class_weight=class_weight,
-> 2383 batch_size=batch_size)
2384
2385 def _standardize_tensors(self, x, y, sample_weight, run_eagerly, dict_inputs,
~/anaconda3/envs/tf2/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/training.py in _standardize_tensors(self, x, y, sample_weight, run_eagerly, dict_inputs, is_dataset, class_weight, batch_size)
2408 feed_input_shapes,
2409 check_batch_axis=False, # Don't enforce the batch size.
-> 2410 exception_prefix='input')
2411
2412 # Get typespecs for the input data and sanitize it if necessary.
~/anaconda3/envs/tf2/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/training_utils.py in standardize_input_data(data, names, shapes, check_batch_axis, exception_prefix)
571 ': expected ' + names[i] + ' to have ' +
572 str(len(shape)) + ' dimensions, but got array '
--> 573 'with shape ' + str(data_shape))
574 if not check_batch_axis:
575 data_shape = data_shape[1:]
ValueError: Error when checking input: expected input_13 to have 3 dimensions, but got array with shape (50000, 32, 32, 3)
I tried putting shape=(32, 32, 3), but that results in the error below:
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
<ipython-input-166-45b654f7e264> in <module>
33 variational_ae.compile(loss="binary_crossentropy", optimizer="rmsprop", metrics=[rounded_accuracy])
34 history = variational_ae.fit(x_train, x_train, epochs=25, batch_size=128,
---> 35 validation_data=(x_test, x_test))
~/anaconda3/envs/tf2/lib/python3.7/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)
817 max_queue_size=max_queue_size,
818 workers=workers,
--> 819 use_multiprocessing=use_multiprocessing)
820
821 def evaluate(self,
~/anaconda3/envs/tf2/lib/python3.7/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, max_queue_size, workers, use_multiprocessing, **kwargs)
233 max_queue_size=max_queue_size,
234 workers=workers,
--> 235 use_multiprocessing=use_multiprocessing)
236
237 total_samples = _get_total_number_of_samples(training_data_adapter)
~/anaconda3/envs/tf2/lib/python3.7/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)
591 max_queue_size=max_queue_size,
592 workers=workers,
--> 593 use_multiprocessing=use_multiprocessing)
594 val_adapter = None
595 if validation_data:
~/anaconda3/envs/tf2/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/training_v2.py in _process_inputs(model, mode, x, y, batch_size, epochs, sample_weights, class_weights, shuffle, steps, distribution_strategy, max_queue_size, workers, use_multiprocessing)
644 standardize_function = None
645 x, y, sample_weights = standardize(
--> 646 x, y, sample_weight=sample_weights)
647 elif adapter_cls is data_adapter.ListsOfScalarsDataAdapter:
648 standardize_function = standardize
~/anaconda3/envs/tf2/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/training.py in _standardize_user_data(self, x, y, sample_weight, class_weight, batch_size, check_steps, steps_name, steps, validation_split, shuffle, extract_tensors_from_dataset)
2381 is_dataset=is_dataset,
2382 class_weight=class_weight,
-> 2383 batch_size=batch_size)
2384
2385 def _standardize_tensors(self, x, y, sample_weight, run_eagerly, dict_inputs,
~/anaconda3/envs/tf2/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/training.py in _standardize_tensors(self, x, y, sample_weight, run_eagerly, dict_inputs, is_dataset, class_weight, batch_size)
2487 # Additional checks to avoid users mistakenly using improper loss fns.
2488 training_utils.check_loss_and_target_compatibility(
-> 2489 y, self._feed_loss_fns, feed_output_shapes)
2490
2491 sample_weights, _, _ = training_utils.handle_partial_sample_weights(
~/anaconda3/envs/tf2/lib/python3.7/site-packages/tensorflow_core/python/keras/engine/training_utils.py in check_loss_and_target_compatibility(targets, loss_fns, output_shapes)
808 raise ValueError('A target array with shape ' + str(y.shape) +
809 ' was passed for an output of shape ' + str(shape) +
--> 810 ' while using as loss `' + loss_name + '`. '
811 'This loss expects targets to have the same shape '
812 'as the output.')
ValueError: A target array with shape (50000, 32, 32, 3) was passed for an output of shape (None, 28, 28) while using as loss `binary_crossentropy`. This loss expects targets to have the same shape as the output.
I also tried removing the Input layer (so that the first layer is the Flatten layer), but I still have to provide the input shape, or else I get an error complaining that 'Flatten' object has no attribute 'shape'. When I do provide the input shape (input_shape=(32, 32)), I get the same error.
Can someone tell me what is going wrong here and how do I fix it?
There is at least two problems in your code:
(32,32,3) vs (32,32). One possible fix is to load your images in grayscale to match your network input, or to make your network accept images with 3 channels.(32,32) vs (28,28). You need to redesign your decoder part to make it output a matrix with the same shape as your input (in your example, a (32,32) matrix).To convert your array to grayscale, you can use tf.image.rgb_to_grayscale, and use tf.squeeze to get rid of the last dimension:
x_train_grayscale = tf.squeeze(tf.image.rgb_to_grayscale(x_train),axis=-1)