TensorFlow vs PyTorch comparison - Interactive Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import TensorFlow library.

TensorFlow

import [1] as tf

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

Asklearn

Btensorflow

Ckeras

Dtorch

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'torch' which is PyTorch's library name.

Using 'keras' which is a high-level API.

TensorFlow is imported using the name 'tensorflow'.

2fill in blank

medium

Complete the code to create a tensor with PyTorch.

TensorFlow

import torch
x = [1]([1, 2, 3])

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

ATensor

BVariable

CTensorFlow

Dtensor

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'Tensor' which is incorrect capitalization.

Using 'Variable' which is deprecated.

PyTorch creates tensors using torch.tensor() with a lowercase 't'.

3fill in blank

hard

Fix the error in the TensorFlow code to create a constant tensor.

TensorFlow

import tensorflow as tf
x = tf.[1]([1, 2, 3])

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

Aconstant

BTensor

CVariable

Dtensor

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'Variable' instead of 'constant'.

Using 'Tensor' which is not a TensorFlow function.

TensorFlow creates constant tensors using tf.constant().

4fill in blank

hard

Fill both blanks to define a simple linear model in PyTorch.

TensorFlow

import torch.nn as nn
model = nn.[1](in_features=10, [2]=1)

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

ALinear

Bout_features

Cin_features

DConv2d

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'Conv2d' which is a convolutional layer.

Using 'in_features' twice instead of 'out_features'.

PyTorch linear layers are created with nn.Linear and require 'in_features' and 'out_features' parameters.

5fill in blank

hard

Fill all three blanks to compile and train a TensorFlow model.

TensorFlow

model.compile(optimizer='[1]', loss='[2]', metrics=['[3]'])
model.fit(x_train, y_train, epochs=5)

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

Aadam

Bsparse_categorical_crossentropy

Caccuracy

Dsgd

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'sgd' optimizer with incompatible loss.

Using 'categorical_crossentropy' instead of 'sparse_categorical_crossentropy' for integer labels.

Common TensorFlow compile parameters include 'adam' optimizer, 'sparse_categorical_crossentropy' loss, and 'accuracy' metric.