Hugging Face integration basics in PyTorch - Model Metrics & Evaluation

Metrics & Evaluation - Hugging Face integration basics

Which metric matters for Hugging Face integration basics and WHY

When using Hugging Face models, the key metrics depend on the task. For text classification, accuracy is common because it shows how often the model predicts the right label. For tasks like question answering or translation, metrics like F1 score or BLEU are important because they measure quality beyond exact matches. Choosing the right metric helps you know if the model is truly learning and useful.

Confusion matrix example for text classification

      Actual \ Predicted | Positive | Negative
      -------------------|----------|---------
      Positive           |    40    |   10    
      Negative           |    5     |   45    
    

This matrix shows 40 true positives (TP), 45 true negatives (TN), 5 false negatives (FN), and 10 false positives (FP). From this, we calculate precision, recall, and accuracy to understand model performance.

Precision vs Recall tradeoff with Hugging Face models

Imagine a spam detector using a Hugging Face model. If it marks too many emails as spam (high recall), it might block good emails (low precision). If it only marks very sure spam (high precision), some spam gets through (low recall). Depending on your goal, you adjust the model or threshold to balance these. For example, in medical text classification, missing a disease mention (low recall) is worse than a few false alarms.

What good vs bad metric values look like

For Hugging Face text classification models:

• Good: Accuracy above 85%, Precision and Recall above 80%, F1 score balanced and high.
• Bad: Accuracy near random chance (e.g., 50% for two classes), Precision or Recall very low (below 50%), or large difference between Precision and Recall indicating imbalance.

Good metrics mean the model predicts well and is reliable for your task.

Common pitfalls in Hugging Face model metrics

• Accuracy paradox: High accuracy can be misleading if classes are imbalanced. For example, 90% accuracy if 90% data is one class but model ignores minority class.
• Data leakage: If test data leaks into training, metrics look unrealistically good.
• Overfitting: Very high training accuracy but low test accuracy means model memorizes instead of learning.
• Ignoring task-specific metrics: Using accuracy for tasks like translation where BLEU or ROUGE are better.

Self-check question

Your Hugging Face text classification model has 98% accuracy but only 12% recall on the positive class (e.g., fraud). Is it good for production? Why or why not?

Answer: No, it is not good. The low recall means the model misses most positive cases, which is critical in fraud detection. Even with high accuracy, the model fails to catch fraud, making it unreliable for production.

Key Result

For Hugging Face integration, choosing task-appropriate metrics like accuracy, precision, recall, and F1 score is key to correctly evaluating model performance.

Practice

1. What is the main purpose of the AutoTokenizer class in Hugging Face integration with PyTorch?

easy

A. To visualize model predictions

B. To convert raw text into tokens that the model can understand

C. To train a new language model from scratch

D. To save the model weights to disk

Solution

1. Step 1: Understand the role of tokenization

   Tokenization breaks down raw text into smaller pieces called tokens, which models use as input.

2. Step 2: Identify what AutoTokenizer does

   AutoTokenizer automatically loads the correct tokenizer for a given pretrained model to convert text into tokens.

3. Final Answer:

   To convert raw text into tokens that the model can understand -> Option B

4. Quick Check:

   AutoTokenizer = text to tokens [OK]

Hint: Tokenizer = text to tokens for model input [OK]

Common Mistakes:

• Confusing tokenizer with model training
• Thinking tokenizer saves model weights
• Assuming tokenizer visualizes outputs

2. Which of the following is the correct way to load a pretrained model using Hugging Face's AutoModel in PyTorch?

easy

A. model = AutoModel.from_pretrained('bert-base-uncased')

B. model = AutoModel.load('bert-base-uncased')

C. model = AutoModel('bert-base-uncased')

D. model = AutoModel.pretrained('bert-base-uncased')

Solution

1. Step 1: Recall the correct method to load pretrained models

   Hugging Face uses the from_pretrained method to load pretrained weights and configuration.

2. Step 2: Match the correct syntax

   The correct syntax is AutoModel.from_pretrained('model_name').

3. Final Answer:

   model = AutoModel.from_pretrained('bert-base-uncased') -> Option A

4. Quick Check:

   Load model with from_pretrained() [OK]

Hint: Use from_pretrained() to load models [OK]

Common Mistakes:

• Using .load() instead of .from_pretrained()
• Calling AutoModel like a function
• Using non-existent .pretrained() method

3. Given the following code snippet, what will be the shape of outputs.last_hidden_state?

from transformers import AutoTokenizer, AutoModel
import torch

tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased')
model = AutoModel.from_pretrained('bert-base-uncased')

inputs = tokenizer('Hello world!', return_tensors='pt')
outputs = model(**inputs)

medium

A. [2, 768]

B. [1, 768]

C. [1, 5, 768]

D. [768, 2]

Solution

1. Step 1: Understand the tokenizer output

   The tokenizer returns input IDs with shape [batch_size, sequence_length]. Here batch_size=1 and sequence_length=5 (tokens: [CLS], 'hello', 'world', '!', [SEP]).

2. Step 2: Understand model output shape

   The model outputs last_hidden_state with shape [batch_size, sequence_length, hidden_size]. For BERT base, hidden_size=768.

3. Final Answer:

   [1, 5, 768] -> Option C

4. Quick Check:

   Output shape = [batch, tokens, features] = [1, 5, 768] [OK]

Hint: Output shape = batch x tokens x hidden size [OK]

Common Mistakes:

• Confusing sequence length with batch size
• Ignoring hidden size dimension
• Assuming output shape is 2D instead of 3D

4. Identify the error in this Hugging Face PyTorch code snippet:

from transformers import AutoTokenizer, AutoModel

tokenizer = AutoTokenizer.from_pretrained('bert-base-uncased')
model = AutoModel.from_pretrained('bert-base-uncased')

inputs = tokenizer('Test input', return_tensors='pt')
outputs = model(inputs)
print(outputs)

medium

A. Missing .to(device) to move inputs to GPU

B. AutoModel cannot be used with BERT models

C. Tokenizer should use return_tensors='tf' instead of 'pt'

D. Model should be called with unpacked inputs: model(**inputs)

Solution

1. Step 1: Check how model is called

   The model expects keyword arguments, so inputs must be unpacked with ** to pass input_ids and attention_mask correctly.

2. Step 2: Identify the fix

   Change model(inputs) to model(**inputs) to avoid errors.

3. Final Answer:

   Model should be called with unpacked inputs: model(**inputs) -> Option D

4. Quick Check:

   Use **inputs when calling model [OK]

Hint: Call model with model(**inputs) not model(inputs) [OK]

Common Mistakes:

• Passing inputs as a single dict argument
• Confusing PyTorch and TensorFlow tensor types
• Thinking AutoModel can't load BERT

5. You want to classify text into positive or negative sentiment using a Hugging Face model in PyTorch. After tokenizing and running the model, which step correctly converts raw model outputs into class probabilities?

hard

A. Apply torch.softmax on the model's logits output

B. Use torch.argmax directly on tokenized inputs

C. Call model.eval() to get probabilities

D. Normalize input tokens before feeding to the model

Solution

1. Step 1: Understand model output for classification

   Models output raw scores called logits for each class, which are not probabilities.

2. Step 2: Convert logits to probabilities

   Applying torch.softmax on logits converts them into probabilities summing to 1.

3. Final Answer:

   Apply torch.softmax on the model's logits output -> Option A

4. Quick Check:

   Softmax = logits to probabilities [OK]

Hint: Use softmax on logits to get probabilities [OK]

Common Mistakes:

• Using argmax on inputs instead of outputs
• Confusing model.eval() with probability conversion
• Normalizing inputs instead of outputs