423A35C7/LLaMA-Factory
1
0
Fork 0
mirror of https://github.com/hiyouga/LLaMA-Factory.git synced 2025-08-23 06:12:50 +08:00
Code Issues Packages Projects Releases Wiki Activity

update

Browse Source 
Former-commit-id: 734e019cc1c120c23305d6f6bc1fb5b8e0d21b30
This commit is contained in:
simonJJJ 2024-08-28 20:22:46 +08:00
parent 8a09b1e732
commit d9fd8f26cc
1 changed files with 2 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
src/llamafactory/data/processors/supervised.py
View File

@ -68,7 +68,7 @@ def _encode_supervised_example(
) )
index += 1 index += 1
message["content"] = content.replace("<|placeholder|>", "<|image_pad|>") message["content"] = content.replace("<|placeholder|>", "<|image_pad|>")
if processor is not None and not hasattr(processor, "image_seq_length"): # llava-like models elif processor is not None and not hasattr(processor, "image_seq_length"): # llava-like models
prompt[0]["content"] = template.image_token + prompt[0]["content"] prompt[0]["content"] = template.image_token + prompt[0]["content"]
messages = prompt + response messages = prompt + response
@ -146,6 +146,7 @@ def preprocess_supervised_dataset(
response=examples["response"][i], response=examples["response"][i],
system=examples["system"][i], system=examples["system"][i],
tools=examples["tools"][i], tools=examples["tools"][i],
images=examples["images"][i],
template=template, template=template,
tokenizer=tokenizer, tokenizer=tokenizer,
processor=processor, processor=processor,