ros/LLaMA-Factory
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge pull request #5909 from hiyouga/hiyouga/dev2

Browse Source 
[data] support auto convert for single image, add image_dir argument

Former-commit-id: ced43fa0c84f7d0792694721d2c5e572c0d0e718
This commit is contained in:
hoshi-hiyouga
2024-11-02 13:43:04 +08:00
committed by GitHub
parent 29ad3783f5 e0045e8386
commit ceb701c2d4
2 changed files with 30 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files

26
src/llamafactory/data/aligner.py
View File

@@ -33,41 +33,47 @@ logger = get_logger(__name__)
def _convert_images(
images: Sequence["ImageInput"],
images: Union["ImageInput", Sequence["ImageInput"]],
dataset_attr: "DatasetAttr",
data_args: "DataArguments",
) -> Optional[List["ImageInput"]]:
r"""
Optionally concatenates image path to dataset dir when loading from local disk.
"""
if len(images) == 0:
if not isinstance(images, list):
images = [images]
elif len(images) == 0:
return None
else:
images = images[:]
images = images[:]
if dataset_attr.load_from in ["script", "file"]:
for i in range(len(images)):
if isinstance(images[i], str) and os.path.isfile(os.path.join(data_args.dataset_dir, images[i])):
images[i] = os.path.join(data_args.dataset_dir, images[i])
if isinstance(images[i], str) and os.path.isfile(os.path.join(data_args.image_dir, images[i])):
images[i] = os.path.join(data_args.image_dir, images[i])
return images
def _convert_videos(
videos: Sequence["VideoInput"],
videos: Union["VideoInput", Sequence["VideoInput"]],
dataset_attr: "DatasetAttr",
data_args: "DataArguments",
) -> Optional[List["VideoInput"]]:
r"""
Optionally concatenates video path to dataset dir when loading from local disk.
"""
if len(videos) == 0:
if not isinstance(videos, list):
videos = [videos]
elif len(videos) == 0:
return None
else:
videos = videos[:]
videos = videos[:]
if dataset_attr.load_from in ["script", "file"]:
for i in range(len(videos)):
if isinstance(videos[i], str) and os.path.isfile(os.path.join(data_args.dataset_dir, videos[i])):
videos[i] = os.path.join(data_args.dataset_dir, videos[i])
if isinstance(videos[i], str) and os.path.isfile(os.path.join(data_args.image_dir, videos[i])):
videos[i] = os.path.join(data_args.image_dir, videos[i])
return videos