riaz.somc/transformers
1
0
Fork 0
mirror of https://github.com/huggingface/transformers.git synced 2025-07-31 18:22:34 +06:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity

[LLaVa-NeXT] Small fixes (#30841)

Browse Source 
* First draft

* Update docstring
This commit is contained in:
NielsRogge 2024-05-16 08:19:15 +02:00 committed by GitHub
parent 4b3eb19fa7
commit 9fd606dbdb
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 8 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
src/transformers/models/llava_next/image_processing_llava_next.py
View File

@ -156,8 +156,8 @@ class LlavaNextImageProcessor(BaseImageProcessor):
number of channels in the image. Can be overridden by the `image_std` parameter in the `preprocess` method. number of channels in the image. Can be overridden by the `image_std` parameter in the `preprocess` method.
Can be overridden by the `image_std` parameter in the `preprocess` method. Can be overridden by the `image_std` parameter in the `preprocess` method.
do_pad (`bool`, *optional*, defaults to `True`): do_pad (`bool`, *optional*, defaults to `True`):
Whether to pad the image. If `True` will pad the images in the batch to the largest image in the batch Whether to pad the image. If `True`, will pad the patch dimension of the images in the batch to the largest
and create a pixel mask. Padding will be applied to the bottom and right of the image with zeros. number of patches in the batch. Padding will be applied to the bottom and right with zeros.
do_convert_rgb (`bool`, *optional*, defaults to `True`): do_convert_rgb (`bool`, *optional*, defaults to `True`):
Whether to convert the image to RGB. Whether to convert the image to RGB.
""" """
@ -203,6 +203,7 @@ class LlavaNextImageProcessor(BaseImageProcessor):
self.do_normalize = do_normalize self.do_normalize = do_normalize
self.image_mean = image_mean if image_mean is not None else OPENAI_CLIP_MEAN self.image_mean = image_mean if image_mean is not None else OPENAI_CLIP_MEAN
self.image_std = image_std if image_std is not None else OPENAI_CLIP_STD self.image_std = image_std if image_std is not None else OPENAI_CLIP_STD
self.do_pad = do_pad
self.do_convert_rgb = do_convert_rgb self.do_convert_rgb = do_convert_rgb
# Copied from transformers.models.clip.image_processing_clip.CLIPImageProcessor.resize with CLIP->LLaVa # Copied from transformers.models.clip.image_processing_clip.CLIPImageProcessor.resize with CLIP->LLaVa
@ -572,7 +573,7 @@ class LlavaNextImageProcessor(BaseImageProcessor):
do_normalize: bool = None, do_normalize: bool = None,
image_mean: Optional[Union[float, List[float]]] = None, image_mean: Optional[Union[float, List[float]]] = None,
image_std: Optional[Union[float, List[float]]] = None, image_std: Optional[Union[float, List[float]]] = None,
do_pad: Optional[bool] = True, do_pad: Optional[bool] = None,
do_convert_rgb: bool = None, do_convert_rgb: bool = None,
return_tensors: Optional[Union[str, TensorType]] = None, return_tensors: Optional[Union[str, TensorType]] = None,
data_format: Optional[ChannelDimension] = ChannelDimension.FIRST, data_format: Optional[ChannelDimension] = ChannelDimension.FIRST,
@ -609,9 +610,9 @@ class LlavaNextImageProcessor(BaseImageProcessor):
image_std (`float` or `List[float]`, *optional*, defaults to `self.image_std`): image_std (`float` or `List[float]`, *optional*, defaults to `self.image_std`):
Image standard deviation to use for normalization. Only has an effect if `do_normalize` is set to Image standard deviation to use for normalization. Only has an effect if `do_normalize` is set to
`True`. `True`.
do_pad (`bool`, *optional*, defaults to self.do_pad): do_pad (`bool`, *optional*, defaults to `self.do_pad`):
Whether to pad the image. If `True` will pad the images in the batch to the largest image in the batch Whether to pad the image. If `True`, will pad the patch dimension of the images in the batch to the largest
and create a pixel mask. Padding will be applied to the bottom and right of the image with zeros. number of patches in the batch. Padding will be applied to the bottom and right with zeros.
do_convert_rgb (`bool`, *optional*, defaults to `self.do_convert_rgb`): do_convert_rgb (`bool`, *optional*, defaults to `self.do_convert_rgb`):
Whether to convert the image to RGB. Whether to convert the image to RGB.
return_tensors (`str` or `TensorType`, *optional*): return_tensors (`str` or `TensorType`, *optional*):
@ -647,6 +648,7 @@ class LlavaNextImageProcessor(BaseImageProcessor):
do_normalize = do_normalize if do_normalize is not None else self.do_normalize do_normalize = do_normalize if do_normalize is not None else self.do_normalize
image_mean = image_mean if image_mean is not None else self.image_mean image_mean = image_mean if image_mean is not None else self.image_mean
image_std = image_std if image_std is not None else self.image_std image_std = image_std if image_std is not None else self.image_std
do_pad = do_pad if do_pad is not None else self.do_pad
do_convert_rgb = do_convert_rgb if do_convert_rgb is not None else self.do_convert_rgb do_convert_rgb = do_convert_rgb if do_convert_rgb is not None else self.do_convert_rgb
images = make_list_of_images(images) images = make_list_of_images(images)