Reshaping (view, reshape, squeeze, unsqueeze) in PyTorch - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

What is the shape of tensor after view operation?

Given the following PyTorch code, what will be the shape of tensor `b` after the `view` operation?

PyTorch

import torch
a = torch.randn(2, 3, 4)
b = a.view(3, 8)

A torch.Size([4, 6])

B torch.Size([6, 4])

C torch.Size([2, 12])

D torch.Size([3, 8])

💡 Hint

Remember that view reshapes the tensor without changing the total number of elements.

The original tensor has shape (2, 3, 4), total elements = 2*3*4 = 24. The view reshapes it to (3, 8), which also has 24 elements.

❓ Predict Output

intermediate

What is the output shape after squeeze operation?

What will be the shape of tensor `b` after applying `squeeze` on tensor `a`?

PyTorch

import torch
a = torch.randn(1, 5, 1, 3)
b = a.squeeze()

A torch.Size([5, 3])

B torch.Size([1, 5, 3])

C torch.Size([5, 1, 3])

D torch.Size([1, 5, 1, 3])

💡 Hint

Squeeze removes all dimensions of size 1.

The original shape is (1, 5, 1, 3). Squeeze removes dimensions with size 1, so the shape becomes (5, 3).

❓ Model Choice

advanced

Which operation correctly adds a new dimension at position 2?

You have a tensor `a` with shape (4, 5). Which PyTorch operation will add a new dimension at position 2 (0-based index), resulting in shape (4, 5, 1)?

PyTorch

import torch
a = torch.randn(4, 5)

A a.unsqueeze(1)

B a.unsqueeze(2)

C a.unsqueeze(0)

D a.unsqueeze(-2)

💡 Hint

Remember that unsqueeze adds a dimension at the specified index.

The tensor `a` has shape (4, 5). Adding a dimension at position 2 means after the second dimension, so shape becomes (4, 5, 1). `a.unsqueeze(2)` does this. `a.unsqueeze(-2)` inserts at position 1, giving (4, 1, 5).

❓ Hyperparameter

advanced

Choosing correct reshape parameters for a tensor

You have a tensor `a` with shape (2, 3, 4). You want to reshape it to a 2D tensor with 6 rows. Which of the following reshape calls achieves this without raising an error?

PyTorch

import torch
a = torch.randn(2, 3, 4)

A a.reshape(3, 8)

B a.reshape(4, 6)

C a.reshape(6, -1)

D a.reshape(6, 5)

💡 Hint

The total number of elements must remain the same after reshape. Use -1 to infer a dimension.

Tensor `a` has 24 elements. A: (3,8) has 6 rows? No. B: (6,-1) infers 4, (6,4)=24, good. C: (6,5)=30 mismatch, error. D: (4,6)=24 but 4 rows.

🔧 Debug

expert

Why does this reshape code raise an error?

Consider the following code snippet. Why does it raise an error?

PyTorch

import torch
a = torch.randn(2, 3, 4)
b = a.view(4, 3, 3)

A The total number of elements does not match between original and new shape.

B view cannot change the order of dimensions, only reshape.

C The new shape must have a dimension of size 1.

D view requires the tensor to be contiguous, and this tensor is not.

💡 Hint

Check the total number of elements in both shapes.

Original tensor has 2*3*4=24 elements. New shape (4,3,3) has 4*3*3=36 elements. Total number of elements does not match.

Practice

1. What does the torch.squeeze() function do to a tensor?

easy

A. Adds a new dimension of size 1 at a specified position

B. Removes all dimensions of size 1 from the tensor

C. Changes the tensor's data without changing its shape

D. Flattens the tensor into a 1D vector

Solution

1. Step 1: Understand the purpose of squeeze

   torch.squeeze() removes dimensions that have size 1, effectively reducing unnecessary dimensions.

2. Step 2: Differentiate from other reshape functions

   Unlike unsqueeze which adds size 1 dims, squeeze removes them without changing data.

3. Final Answer:

   Removes all dimensions of size 1 from the tensor -> Option B

4. Quick Check:

   squeeze removes size 1 dims = B [OK]

Hint: Squeeze removes dims with size 1, like cleaning empty boxes [OK]

Common Mistakes:

• Confusing squeeze with unsqueeze
• Thinking squeeze changes data values
• Assuming squeeze flattens the tensor

2. Which of the following is the correct syntax to add a new dimension at position 0 using PyTorch?

easy

A. tensor.unsqueeze(0)

B. tensor.unsqueeze(1)

C. tensor.squeeze(0)

D. tensor.reshape(0)

Solution

1. Step 1: Recall unsqueeze syntax

   unsqueeze(dim) adds a dimension of size 1 at the specified dim index.

2. Step 2: Identify correct dimension index

   To add at position 0 (the front), use unsqueeze(0). tensor.unsqueeze(1) adds at position 1, C squeezes, D is invalid.

3. Final Answer:

   tensor.unsqueeze(0) -> Option A

4. Quick Check:

   unsqueeze(0) adds dim at front = A [OK]

Hint: unsqueeze(dim) adds size 1 dim at dim index [OK]

Common Mistakes:

• Using squeeze instead of unsqueeze
• Passing wrong dimension index
• Using reshape with a single integer argument

3. Given the tensor t = torch.randn(2, 1, 3), what is the shape after t.view(2, 3)?

medium

A. (2, 3)

B. (2, 1, 3)

C. (1, 6)

D. RuntimeError due to incompatible shape

Solution

1. Step 1: Understand original tensor shape

   Tensor t has shape (2, 1, 3), total elements = 2*1*3 = 6.

2. Step 2: Check if view(2, 3) is compatible

   New shape (2, 3) also has 6 elements, so view can reshape without error.

3. Final Answer:

   (2, 3) -> Option A

4. Quick Check:

   view reshapes without changing total elements = A [OK]

Hint: view reshapes if total elements match exactly [OK]

Common Mistakes:

• Confusing view with squeeze or unsqueeze
• Assuming view changes data order
• Trying incompatible shapes causing runtime error

4. What is wrong with this code snippet?

import torch
t = torch.randn(3, 1, 4)
t = t.view(3, 3, 1, 1)

medium

A. The new shape has more elements than original tensor

B. view cannot add new dimensions, only reshape existing ones

C. view requires contiguous tensor, but t is not contiguous

D. The new shape is incompatible with the total number of elements

Solution

1. Step 1: Calculate original tensor elements

   Original shape (3,1,4) has 3*1*4 = 12 elements.

2. Step 2: Calculate new shape elements

   New shape (3,3,1,1) has 3*3*1*1 = 9 elements, which does not match.

3. Final Answer:

   The new shape is incompatible with the total number of elements -> Option D

4. Quick Check:

   view fails if total elements don't match = C [OK]

Hint: Check total elements before view; must match exactly [OK]

Common Mistakes:

• Assuming view cannot add dimensions
• Confusing contiguous memory errors
• Miscounting total elements

5. You have a tensor x of shape (4, 1, 28, 28). You want to prepare it for a fully connected layer that expects input shape (4, 784). Which sequence of PyTorch operations correctly reshapes x without changing data order?

hard

A. x = x.view(4, 28, 28)

B. x = x.unsqueeze(1).view(4, 1, 784)

C. x = x.squeeze(1).view(4, 784)

D. x = x.reshape(4, 1, 784)

Solution

1. Step 1: Understand original tensor shape

   x has shape (4, 1, 28, 28). The middle dimension is size 1, which is unnecessary for flattening.

2. Step 2: Remove the size 1 dimension

   Use squeeze(1) to remove the dimension of size 1 at index 1, resulting in shape (4, 28, 28).

3. Step 3: Flatten the last two dimensions

   Use view(4, 784) to reshape (4, 28, 28) into (4, 784) without changing data order.

4. Final Answer:

   x = x.squeeze(1).view(4, 784) -> Option C

5. Quick Check:

   Remove size 1 dim then flatten = D [OK]

Hint: Squeeze first to remove size 1 dims, then view to flatten [OK]

Common Mistakes:

• Trying to view directly without squeezing
• Using unsqueeze unnecessarily
• Confusing reshape and view effects