fix: correct Bi-LSTM output shape comment (n_hidden*2 not n_hidden)

[Mukller]

Fix: Incorrect output shape comment in Bi-LSTM(Attention)

The comment for the LSTM output after permute(1, 0, 2) incorrectly states
[batch_size, len_seq, n_hidden] when it should be [batch_size, len_seq, n_hidden * num_directions(=2)].

A bidirectional LSTM concatenates forward and backward hidden states, so the
last dimension is n_hidden * 2, not n_hidden.

Before (wrong comment):

output = output.permute(1, 0, 2) # output : [batch_size, len_seq, n_hidden]

After (correct comment):

output = output.permute(1, 0, 2) # output : [batch_size, len_seq, n_hidden * num_directions(=2)]

This matches the existing comments on surrounding lines (e.g., attention_net's
docstring already correctly states n_hidden * num_directions(=2)).

Closes #84

[Mukller]

Code Review: fix Bi-LSTM output shape comment

Summary

Corrects a misleading inline comment in the Bi-LSTM attention model. The comment described the LSTM output tensor as [batch_size, len_seq, n_hidden], which is the shape for a unidirectional LSTM. For a bidirectional LSTM (num_directions=2), the last dimension is n_hidden * 2.

Issues Found

#	File	Line	Issue	Severity
1	Bi-LSTM(Attention).py	~37	Comment says n_hidden but correct shape is n_hidden * num_directions(=2)	🟡 Minor (comment only)

What the Fix Does

# Before
output = output.permute(1, 0, 2) # output : [batch_size, len_seq, n_hidden]

# After
output = output.permute(1, 0, 2) # output : [batch_size, len_seq, n_hidden * num_directions(=2)]

This is a documentation-only fix. Readers following the code to understand tensor shapes would otherwise be confused when passing the output to attention_net, which operates on the full n_hidden * 2 dimension.

What Looks Good

• Single-line, zero-risk change
• Accurate dimension annotation helps readers trace tensor shapes through the attention mechanism

Verdict

Approve. Clean documentation fix that prevents a common point of confusion in bidirectional RNN code.