Model versioning in TensorFlow - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to save the TensorFlow model with a version number.

TensorFlow

model.save('model_v[1]')

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

Asave

Bversion

Clatest

D1

Attempts:

3 left

💡 Hint

Common Mistakes

Using a string like 'version' instead of a number.

Using 'latest' which is not a version number.

The version number is typically a simple integer like '1' to indicate the model version.

2fill in blank

medium

Complete the code to load a specific version of the saved TensorFlow model.

TensorFlow

loaded_model = tf.keras.models.load_model('model_v[1]')

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

A1

Bversion

Csave

Dlatest

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'latest' which is not a folder name.

Using 'version' which is not a valid folder.

To load a specific version, use the version number like '1' that matches the saved folder.

3fill in blank

hard

Fix the error in the code to save the model with versioning correctly.

TensorFlow

model.save('model_v[1]')

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

A1.0

Bv1

C1

Dversion1

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'v1' which duplicates the 'v' in the folder name.

Using '1.0' which can cause confusion in folder naming.

The folder name should be simple and consistent, like 'model_v1'. Using '1' after 'v' is correct.

4fill in blank

hard

Fill both blanks to create a dictionary that maps model versions to their file paths.

TensorFlow

model_paths = {'v1': 'model_v[1]', 'v2': 'model_v[2]'}

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

A1

B2

C3

Dlatest

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'latest' as a folder name which is not versioned.

Mixing up the version numbers in the dictionary.

The dictionary keys 'v1' and 'v2' map to folder names 'model_v1' and 'model_v2' respectively.

5fill in blank

hard

Fill all three blanks to load a model, make a prediction, and print the result.

TensorFlow

model = tf.keras.models.load_model('model_v[1]')
prediction = model.predict([2])
print('Prediction:', [3])

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

A1

Binput_data

Cprediction

Ddata

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'data' instead of 'input_data' for prediction input.

Printing 'data' instead of the prediction result.

Load version 1 of the model, predict using 'input_data', and print the 'prediction' variable.