WIP: Training TASTE w/o vq using llama tokenizer (text) #4
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
GitYCC
changed the title
GitYCC
changed the title
GitYCC
reviewed
Jul 30, 2025
| make_v_proj_identity: bool = False, | ||
| is_word_level: bool = False, | ||
| skip_prefix_idx: Optional[int] = None, | ||
| vocab_size: int = None, |
Member
There was a problem hiding this comment.
new augment new_vocab_size=None
if new_vocab_size is not None:
do your work
| postfix_token_to_wrap = [tokenizer.eos_token_id] if add_eos else [] | ||
| _skip_prefix_idx = len(prefix_token_to_wrap) | ||
| logging.info(f"Tokenizer is from transformers `WhisperTokenizerFast` of transformers. Decoder prefix ids: {forced_decoder_ids}.") | ||
| if whisper_tokenizer_name_or_fpath.endswith("Llama-3.2-1B"): |
| prefix_token_to_wrap = [tokenizer.bos_token_id] if add_bos else [] | ||
| postfix_token_to_wrap = [tokenizer.eos_token_id] if add_eos else [] | ||
| _skip_prefix_idx = len(prefix_token_to_wrap) | ||
| logging.info(f"Using Llama tokenizer from {whisper_tokenizer_name_or_fpath}") |
| $RTSLM_WORK_DIR/CosyVoice/cosyvoice/bin/train.py \ | ||
| --train_engine $train_engine \ | ||
| --config $conf_fpath \ | ||
| --train_data ./data/train.data.list \ |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Training TASTE w/o vq, using llama tokenizer (Llama-3.2-1B)
taste_no_vq_llama.yaml)