Skip to content
Snippets Groups Projects

Support subword and word language models

Merged Support subword and word language models
subword-and-word-lm into main
All threads resolved!
Merged Solene Tarride requested to merge subword-and-word-lm into main
All threads resolved!
Viewing commit fdb5ab7f
Prev Next
Show latest version
1 file
+ 0
− 2
Compare changes
  • Side-by-side
  • Inline
dan/ocr/decoder.py
+ 0
− 2
@@ -516,8 +516,6 @@ class CTCLanguageDecoder:
Add CTC frames between each characters to avoid duplicate removal
"""
batch_size, _, n_tokens = batch_features.shape
torch.clone(batch_features)
# visualize_debug(batch_features.exp()[0, :batch_frames[0], :].numpy(), "probs.jpg", False)
# Create tensor with high probability CTC token
high_prob = 0.99