riaz.somc/transformers
1
0
Fork 0
mirror of https://github.com/huggingface/transformers.git synced 2025-07-31 02:02:21 +06:00
Code Issues Actions 33 Packages Projects Releases Wiki Activity

Update logits_process.py docstrings (#25971)

Browse Source
This commit is contained in:
larekrow 2023-09-12 19:36:31 +08:00 committed by GitHub
parent 3319eb5490
commit d65c4a4fed
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/transformers/generation/logits_process.py
View File

@ -272,7 +272,7 @@ class RepetitionPenaltyLogitsProcessor(LogitsProcessor):
[`LogitsProcessor`] that prevents the repetition of previous tokens through an exponential penalty. This technique
shares some similarities with coverage mechanisms and other aimed at reducing repetition. During the text
generation process, the probability distribution for the next token is determined using a formula that incorporates
token scores based on their occurrence in the generated sequence. Tokens with higher scores are less likely to be
token scores based on their occurrence in the generated sequence. Tokens with higher scores are more likely to be
selected. The formula can be seen in the original [paper](https://arxiv.org/pdf/1909.05858.pdf). According to the
paper a penalty of around 1.2 yields a good balance between truthful generation and lack of repetition.
@ -328,7 +328,7 @@ class EncoderRepetitionPenaltyLogitsProcessor(LogitsProcessor):
hallucination_penalty (`float`):
The parameter for hallucination penalty. 1.0 means no penalty.
encoder_input_ids (`torch.LongTensor`):
The encoder_input_ids that should not be repeated within the decoder ids.
The encoder_input_ids that should be repeated within the decoder ids.
"""
def __init__(self, penalty: float, encoder_input_ids: torch.LongTensor):