HDF5 format in TensorFlow - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to save a TensorFlow model in HDF5 format.

TensorFlow

model.save('my_model.[1]')

Drag options to blanks, or click blank then click option'

Ahdf5

Bhdf

Chdf6

Dh5

Attempts:

3 left

💡 Hint

Common Mistakes

Using .hdf or .hdf5 extensions which are not recognized by TensorFlow save method by default.

Using unsupported or incorrect file extensions.

The correct file extension for saving a TensorFlow model in HDF5 format is .h5.

2fill in blank

medium

Complete the code to load a TensorFlow model saved in HDF5 format.

TensorFlow

loaded_model = tf.keras.models.[1]('my_model.h5')

Drag options to blanks, or click blank then click option'

Aload

Bloadmodel

Cload_model

DloadKeras

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'load' which is not a valid function in tf.keras.models.

Misspelling the function name.

The function to load a saved Keras model is load_model.

3fill in blank

hard

Fix the error in the code to save the model in HDF5 format.

TensorFlow

model.save('model[1]')

Drag options to blanks, or click blank then click option'

A.h5

B.hdf

C.txt

D.json

Attempts:

3 left

💡 Hint

Common Mistakes

Using '.txt' or '.json' extensions which do not save the model correctly.

Omitting the file extension.

To save the model in HDF5 format, the filename must end with .h5.

4fill in blank

hard

Fill both blanks to create a dictionary comprehension that stores layer names and their weights from a model.

TensorFlow

weights_dict = {layer.[1]: layer.[2] for layer in model.layers}

Drag options to blanks, or click blank then click option'

Aname

Bweights

Cbias

Doutput

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'bias' instead of 'weights' which only gives part of the parameters.

Using 'output' which is not an attribute for this purpose.

Each layer's name is used as the key, and its weights are the values in the dictionary.

5fill in blank

hard

Fill all three blanks to create a dictionary comprehension that stores layer names and the count of their weights if the layer has weights.

TensorFlow

weights_count = {layer.[1]: len(layer.[2]) for layer in model.layers if layer.[3]

Drag options to blanks, or click blank then click option'

Aname

Bweights

Dhas_weights

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'bias' or 'output' instead of 'weights'.

Not checking if the layer has weights before counting.

The dictionary keys are layer name, the values are the length of weights, and the condition checks if the layer has_weights.