Why transformers revolutionized NLP - Test Your Understanding

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the Transformer model from the Hugging Face library.

NLP

from transformers import [1]

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

Atransformers

BTransformerModelClass

CAutoModel

DTransformerModel

Attempts:

3 left

💡 Hint

Common Mistakes

Using lowercase or plural form like 'transformers' which is the package name.

Using incorrect class names like 'TransformerModel'.

The correct import is AutoModel which is the base class for transformer models in Hugging Face.

2fill in blank

medium

Complete the code to tokenize input text using a pretrained tokenizer.

NLP

tokenizer = AutoTokenizer.from_pretrained('[1]')

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

Atransformer-base

Bbert-base-uncased

Cnlp-transformer

Dbert-large-cased

Attempts:

3 left

💡 Hint

Common Mistakes

Using non-existent model names like 'transformer-base'.

Using 'bert-large-cased' which is a different model variant.

The bert-base-uncased is a common pretrained model name used for tokenization.

3fill in blank

hard

Fix the error in the code to correctly create attention masks for input tokens.

NLP

attention_mask = (input_ids != [1]).long()

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

A0

B1

CNone

D-1

Attempts:

3 left

💡 Hint

Common Mistakes

Using 1 instead of 0, which reverses the mask meaning.

Using None or -1 which causes errors.

Attention masks use 0 to mark padding tokens, so tokens not equal to 0 are attended to.

4fill in blank

hard

Fill both blanks to create a dictionary of token ids and attention masks for model input.

NLP

inputs = {'input_ids': [1], 'attention_mask': [2]

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

Ainput_ids

Battention_mask

Ctoken_ids

Dmask

Attempts:

3 left

💡 Hint

Common Mistakes

Using incorrect variable names like 'token_ids' or 'mask'.

Swapping the values between keys.

The model expects keys 'input_ids' and 'attention_mask' with corresponding tensors.

5fill in blank

hard

Fill all three blanks to run the transformer model and get the output logits.

NLP

outputs = model([1], [2])
logits = outputs.[3]

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

Ainput_ids

Battention_mask

Clogits

Dhidden_states

Attempts:

3 left

💡 Hint

Common Mistakes

Passing wrong arguments or missing attention_mask.

Accessing 'hidden_states' instead of 'logits' for predictions.

The model is called with input_ids and attention_mask, and the logits are accessed from outputs.