Bird
0
0

Find the bug in this GRU implementation snippet: ```python gru = nn.GRU(input_size=30, hidden_size=50, batch_first=True) input_tensor = torch.randn(4, 10, 30) hidden = torch.randn(1, 4, 50) output, hidden = gru(input_tensor, hidden) print(output.shape) ```

medium📝 Debug Q7 of 15
NLP - Sequence Models for NLP
Find the bug in this GRU implementation snippet: ```python gru = nn.GRU(input_size=30, hidden_size=50, batch_first=True) input_tensor = torch.randn(4, 10, 30) hidden = torch.randn(1, 4, 50) output, hidden = gru(input_tensor, hidden) print(output.shape) ```
AHidden state shape should be (1, 4, 30) instead of (1, 4, 50)
BHidden state shape is correct; no bug present
CInput tensor shape should be (4, 30, 10) for batch_first=True
DHidden state should be initialized with zeros, not random values
Step-by-Step Solution
Solution:

  1. Step 1: Check input tensor shape with batch_first=True

    Input shape (4,10,30) matches (batch=4, seq_len=10, input_size=30).

  2. Step 2: Verify hidden state shape

    Hidden shape (1,4,50) matches (num_layers=1, batch=4, hidden_size=50).

  3. Step 3: Confirm hidden initialization method

    Random initialization is allowed; no strict requirement for zeros.

  4. Final Answer:

    Hidden state shape is correct; no bug present -> Option B

  5. Quick Check:

    Shapes match GRU expected dimensions [OK]
Quick Trick: Hidden shape = (num_layers, batch, hidden_size) always [OK]
Common Mistakes:
MISTAKES
  • Confusing input tensor dimension order
  • Assuming hidden must be zeros
  • Mixing hidden size with input size

Want More Practice?

15+ quiz questions · All difficulty levels · Free

Free Signup - Practice All Questions
More NLP Quizzes
Sentiment Analysis Advanced - Hybrid approaches - Quiz 5medium Sequence Models for NLP - Attention mechanism basics - Quiz 9hard Sequence Models for NLP - Bidirectional LSTM - Quiz 12easy Sequence Models for NLP - RNN for text classification - Quiz 10hard Text Generation - RNN-based text generation - Quiz 8hard Topic Modeling - Latent Dirichlet Allocation (LDA) - Quiz 12easy Topic Modeling - LDA with scikit-learn - Quiz 8hard Topic Modeling - Topic coherence evaluation - Quiz 9hard Word Embeddings - Pre-trained embedding usage - Quiz 12easy Word Embeddings - Why embeddings capture semantic meaning - Quiz 4medium