Dataset from tensors in TensorFlow - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create a TensorFlow dataset from a tensor.

TensorFlow

import tensorflow as tf

data = tf.constant([1, 2, 3, 4, 5])
dataset = tf.data.Dataset.[1](data)

for item in dataset:
    print(item.numpy())

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

Afrom_tensor_slices

Bfrom_tensor

Cfrom_list

Dfrom_array

Attempts:

3 left

💡 Hint

Common Mistakes

Using a non-existent method like from_tensor or from_list.

Confusing from_tensor_slices with other dataset creation methods.

The correct method to create a dataset from a tensor is from_tensor_slices. It slices the tensor along the first dimension to create elements.

2fill in blank

medium

Complete the code to create a dataset from two tensors representing features and labels.

TensorFlow

import tensorflow as tf

features = tf.constant([[1, 2], [3, 4], [5, 6]])
labels = tf.constant([0, 1, 0])
dataset = tf.data.Dataset.[1]((features, labels))

for feature, label in dataset:
    print(feature.numpy(), label.numpy())

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

Afrom_list

Bfrom_array

Cfrom_tensors

Dfrom_tensor_slices

Attempts:

3 left

💡 Hint

Common Mistakes

Using from_tensors which creates a dataset with one element containing all data.

Using from_list which is not a TensorFlow dataset method.

from_tensor_slices creates a dataset by slicing each tensor element-wise, pairing features and labels correctly.

3fill in blank

hard

Fix the error in the code to correctly create a dataset from a tensor.

TensorFlow

import tensorflow as tf

numbers = tf.constant([10, 20, 30])
dataset = tf.data.Dataset.[1](numbers)

for num in dataset:
    print(num.numpy())

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

Afrom_tensor

Bfrom_array

Cfrom_tensor_slices

Dfrom_list

Attempts:

3 left

💡 Hint

Common Mistakes

Using a non-existent method like from_tensor.

Confusing from_tensor_slices with other dataset creation methods.

The method from_tensor_slices correctly creates a dataset by slicing the tensor into individual elements.

4fill in blank

hard

Fill both blanks to create a dataset from features and labels and iterate over it.

TensorFlow

import tensorflow as tf

features = tf.constant([[7, 8], [9, 10]])
labels = tf.constant([1, 0])
dataset = tf.data.Dataset.[1]((features, labels))

for [2], label in dataset:
    print([2].numpy(), label.numpy())

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

Afrom_tensor_slices

Bfrom_tensors

Cfeature

Dx

Attempts:

3 left

💡 Hint

Common Mistakes

Using from_tensors which creates a dataset with one element.

Using a loop variable name that does not match the print statement.

Use from_tensor_slices to create the dataset. The loop variable for features is named feature to match the print statement.

5fill in blank

hard

Fill all three blanks to create a dataset from tensors, shuffle it, and batch it.

TensorFlow

import tensorflow as tf

features = tf.constant([[1, 2], [3, 4], [5, 6], [7, 8]])
labels = tf.constant([0, 1, 0, 1])
dataset = tf.data.Dataset.[1]((features, labels))
dataset = dataset.[2](buffer_size=4)
dataset = dataset.[3](batch_size=2)

for batch_features, batch_labels in dataset:
    print(batch_features.numpy(), batch_labels.numpy())

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

Afrom_tensor_slices

Bshuffle

Cbatch

Drepeat

Attempts:

3 left

💡 Hint

Common Mistakes

Using repeat instead of shuffle or batch.

Not shuffling before batching.

First, create the dataset with from_tensor_slices. Then shuffle it with shuffle and group elements with batch.