Transformer decoder in PyTorch - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is the main role of the Transformer decoder in a sequence-to-sequence model?

The Transformer decoder generates the output sequence step-by-step by attending to the encoder's output and previously generated tokens, enabling tasks like translation or text generation.

beginner

Name the three main sub-layers inside a Transformer decoder block.

1. Masked multi-head self-attention
2. Multi-head cross-attention (attends to encoder output)
3. Position-wise feed-forward network

intermediate

Why does the Transformer decoder use masked self-attention?

Masked self-attention prevents the decoder from 'seeing' future tokens during training, ensuring predictions are made only from past and current tokens, preserving the autoregressive property.

beginner

In PyTorch, which class is commonly used to implement a Transformer decoder layer?

torch.nn.TransformerDecoderLayer is used to build a single decoder layer, and torch.nn.TransformerDecoder stacks multiple such layers.

intermediate

What is the purpose of the cross-attention sub-layer in the Transformer decoder?

Cross-attention allows the decoder to focus on relevant parts of the encoder's output, helping it generate context-aware outputs based on the input sequence.

Which sub-layer in the Transformer decoder prevents the model from attending to future tokens?

A Feed-forward network

B Cross-attention

C Masked multi-head self-attention

D Layer normalization

Masked multi-head self-attention blocks future tokens to maintain autoregressive generation.

What does the cross-attention layer in the Transformer decoder attend to?

A Encoder outputs

B Future decoder tokens

C Previous decoder outputs only

D Input embeddings

Cross-attention attends to encoder outputs to incorporate input context.

In PyTorch, which class stacks multiple Transformer decoder layers?

A torch.nn.TransformerEncoder

B torch.nn.MultiheadAttention

C torch.nn.TransformerDecoderLayer

D torch.nn.TransformerDecoder

torch.nn.TransformerDecoder stacks multiple TransformerDecoderLayer instances.

Why is masking important in the Transformer decoder's self-attention?

A To prevent attending to future tokens

B To speed up training

C To reduce model size

D To normalize inputs

Masking ensures the decoder only uses past and current tokens for prediction.

Which of these is NOT a sub-layer in a Transformer decoder block?

A Masked multi-head self-attention

B Convolutional layer

C Position-wise feed-forward network

D Multi-head cross-attention

Transformer decoders do not use convolutional layers in their standard architecture.

Explain the architecture and function of a Transformer decoder layer.

Think about how the decoder processes previous outputs and encoder information step-by-step.

Describe how masking works in the Transformer decoder and why it is necessary.

Consider the order of token generation in language models.

Practice

1. What is the main role of the Transformer decoder in a sequence-to-sequence model?

easy

A. To reduce the dimensionality of input data

B. To encode input sequences into fixed-size vectors

C. To perform image classification tasks

D. To generate output tokens step-by-step using previous outputs and encoder info

Solution

1. Step 1: Understand the Transformer decoder purpose

   The Transformer decoder generates output sequences by attending to previous outputs and encoder outputs.

2. Step 2: Compare options with decoder role

   Only To generate output tokens step-by-step using previous outputs and encoder info describes generating output tokens step-by-step using previous outputs and encoder info, which matches the decoder's role.

3. Final Answer:

   To generate output tokens step-by-step using previous outputs and encoder info -> Option D

4. Quick Check:

   Decoder generates outputs stepwise = D [OK]

Hint: Decoder creates outputs using past outputs and encoder data [OK]

Common Mistakes:

• Confusing encoder and decoder roles
• Thinking decoder encodes inputs
• Assuming decoder reduces data size

2. Which of the following is the correct way to create a Transformer decoder layer in PyTorch?

easy

A. torch.nn.TransformerDecoderLayer(d_model=512, nhead=8)

B. torch.nn.TransformerEncoderLayer(d_model=512, nhead=8)

C. torch.nn.TransformerDecoder(d_model=512, nhead=8)

D. torch.nn.TransformerLayer(d_model=512, nhead=8)

Solution

1. Step 1: Identify the correct PyTorch class for a decoder layer

   PyTorch provides torch.nn.TransformerDecoderLayer to create a single decoder layer.

2. Step 2: Check options for correct class and parameters

   torch.nn.TransformerDecoderLayer(d_model=512, nhead=8) uses torch.nn.TransformerDecoderLayer with correct parameters. Others are either encoder layers or invalid classes.

3. Final Answer:

   torch.nn.TransformerDecoderLayer(d_model=512, nhead=8) -> Option A

4. Quick Check:

   Decoder layer class is TransformerDecoderLayer = A [OK]

Hint: Decoder layer class ends with 'DecoderLayer' in PyTorch [OK]

Common Mistakes:

• Using EncoderLayer instead of DecoderLayer
• Trying to instantiate TransformerDecoder directly
• Using non-existent class names

3. Given the code below, what is the shape of the output tensor?

import torch
from torch.nn import TransformerDecoder, TransformerDecoderLayer

layer = TransformerDecoderLayer(d_model=256, nhead=8)
decoder = TransformerDecoder(layer, num_layers=3)

memory = torch.rand(10, 32, 256)  # (source_seq_len, batch_size, d_model)
tgt = torch.rand(20, 32, 256)     # (target_seq_len, batch_size, d_model)

output = decoder(tgt, memory)

medium

A. (32, 20, 256)

B. (10, 32, 256)

C. (20, 32, 256)

D. (20, 256, 32)

Solution

1. Step 1: Understand input and output shapes for TransformerDecoder

   The decoder input shape is (target_seq_len, batch_size, d_model). The output shape matches the target input shape.

2. Step 2: Check given shapes and output

   Target input 'tgt' has shape (20, 32, 256), so output will have the same shape.

3. Final Answer:

   (20, 32, 256) -> Option C

4. Quick Check:

   Output shape matches target input shape = C [OK]

Hint: Output shape matches target input shape in decoder [OK]

Common Mistakes:

• Confusing source and target sequence lengths
• Mixing batch size and feature dimensions
• Assuming output shape changes after decoding

4. What is wrong with this PyTorch Transformer decoder usage?

import torch
from torch.nn import TransformerDecoder, TransformerDecoderLayer

layer = TransformerDecoderLayer(d_model=128, nhead=4)
decoder = TransformerDecoder(layer, num_layers=2)

memory = torch.rand(15, 16, 128)
tgt = torch.rand(15, 16, 64)  # Incorrect feature size

output = decoder(tgt, memory)

medium

A. The memory tensor has wrong batch size

B. The target tensor 'tgt' has wrong feature dimension

C. The number of decoder layers is too low

D. The nhead parameter should be equal to batch size

Solution

1. Step 1: Check feature dimensions of inputs

   The decoder expects target and memory tensors to have the same feature dimension (d_model=128).

2. Step 2: Identify mismatch in target tensor

   Target tensor 'tgt' has feature size 64, which does not match d_model=128, causing an error.

3. Final Answer:

   The target tensor 'tgt' has wrong feature dimension -> Option B

4. Quick Check:

   Feature size mismatch in target tensor = A [OK]

Hint: Check if target and memory feature sizes match d_model [OK]

Common Mistakes:

• Ignoring feature dimension mismatch
• Assuming batch size must match nhead
• Thinking number of layers causes this error

5. You want to build a Transformer decoder that generates text step-by-step using previous outputs and encoder memory. Which of the following is the best approach to implement the decoder's forward pass in PyTorch?

hard

A. Use a TransformerDecoder with masked target input to prevent attending to future tokens

B. Use a TransformerDecoder without masking so it sees all target tokens

C. Use only a TransformerEncoder for generation

D. Use a TransformerDecoder but feed the entire output sequence at once without masking

Solution

1. Step 1: Understand the need for masking in decoding

   During generation, the decoder must not look at future tokens to avoid cheating, so masking is required.

2. Step 2: Identify correct usage in PyTorch

   Using TransformerDecoder with a mask on target input prevents attending to future tokens, enabling step-by-step generation.

3. Final Answer:

   Use a TransformerDecoder with masked target input to prevent attending to future tokens -> Option A

4. Quick Check:

   Mask future tokens in decoder input = B [OK]

Hint: Mask future tokens in decoder input for stepwise generation [OK]

Common Mistakes:

• Not using masks and allowing future token attention
• Using encoder only for generation
• Feeding full output sequence without masking