mirror of
https://github.com/hiyouga/LLaMA-Factory.git
synced 2025-12-16 11:50:35 +08:00
Merge branch 'main' into pixtral-patch
This commit is contained in:
Zhangchi Feng
GitHub
@@ -4,6 +4,7 @@ from io import BytesIO
|
||||
from typing import TYPE_CHECKING, Dict, List, Optional, Sequence, Tuple, TypedDict, Union
|
||||
|
||||
import numpy as np
|
||||
from transformers.image_utils import get_image_size, to_numpy_array
|
||||
from typing_extensions import override
|
||||
|
||||
from ..extras.constants import IGNORE_INDEX, IMAGE_PLACEHOLDER, VIDEO_PLACEHOLDER
|
||||
@@ -157,6 +158,7 @@ class BasePlugin:
|
||||
It holds num_patches == torch.prod(image_grid_thw)
|
||||
"""
|
||||
image_processor: "BaseImageProcessor" = getattr(processor, "image_processor")
|
||||
video_processor: "BaseImageProcessor" = getattr(processor, "video_processor", image_processor)
|
||||
input_dict = {"images": None} # default key
|
||||
if len(images) != 0:
|
||||
images = self._regularize_images(
|
||||
@@ -174,10 +176,16 @@ class BasePlugin:
|
||||
)
|
||||
input_dict["videos"] = videos
|
||||
|
||||
if input_dict.get("images", None) is not None or input_dict.get("videos", None) is not None:
|
||||
return image_processor(**input_dict, return_tensors="pt")
|
||||
else:
|
||||
return {}
|
||||
mm_inputs = {}
|
||||
if image_processor != video_processor:
|
||||
if input_dict.get("images") is not None:
|
||||
mm_inputs.update(image_processor(input_dict["images"], return_tensors="pt"))
|
||||
if input_dict.get("videos") is not None:
|
||||
mm_inputs.update(video_processor(input_dict["videos"], return_tensors="pt"))
|
||||
elif input_dict.get("images") is not None or input_dict.get("videos") is not None: # same processor (qwen2-vl)
|
||||
mm_inputs.update(image_processor(**input_dict, return_tensors="pt"))
|
||||
|
||||
return mm_inputs
|
||||
|
||||
def process_messages(
|
||||
self,
|
||||
@@ -263,6 +271,122 @@ class LlavaPlugin(BasePlugin):
|
||||
return self._get_mm_inputs(images, videos, processor)
|
||||
|
||||
|
||||
class LlavaNextPlugin(BasePlugin):
|
||||
@override
|
||||
def process_messages(
|
||||
self,
|
||||
messages: Sequence[Dict[str, str]],
|
||||
images: Sequence["ImageInput"],
|
||||
videos: Sequence["VideoInput"],
|
||||
processor: Optional["ProcessorMixin"],
|
||||
) -> List[Dict[str, str]]:
|
||||
self._validate_input(images, videos)
|
||||
num_image_tokens = 0
|
||||
messages = deepcopy(messages)
|
||||
mm_inputs = self._get_mm_inputs(images, videos, processor)
|
||||
if "image_sizes" in mm_inputs:
|
||||
image_sizes = iter(mm_inputs["image_sizes"])
|
||||
if "pixel_values" in mm_inputs:
|
||||
height, width = get_image_size(to_numpy_array(mm_inputs["pixel_values"][0][0]))
|
||||
for message in messages:
|
||||
content = message["content"]
|
||||
while self.image_token in content:
|
||||
image_size = next(image_sizes)
|
||||
orig_height, orig_width = image_size
|
||||
image_seqlen = processor._get_number_of_features(orig_height, orig_width, height, width)
|
||||
if processor.vision_feature_select_strategy == "default":
|
||||
image_seqlen -= 1
|
||||
num_image_tokens += 1
|
||||
content = content.replace(self.image_token, "{{image}}" * image_seqlen, 1)
|
||||
|
||||
message["content"] = content.replace("{{image}}", self.image_token)
|
||||
|
||||
if len(images) != num_image_tokens:
|
||||
raise ValueError("The number of images does not match the number of {} tokens".format(IMAGE_PLACEHOLDER))
|
||||
return messages
|
||||
|
||||
@override
|
||||
def get_mm_inputs(
|
||||
self,
|
||||
images: Sequence["ImageInput"],
|
||||
videos: Sequence["VideoInput"],
|
||||
imglens: Sequence[int],
|
||||
vidlens: Sequence[int],
|
||||
seqlens: Sequence[int],
|
||||
processor: Optional["ProcessorMixin"],
|
||||
) -> Dict[str, Union[List[int], "torch.Tensor"]]:
|
||||
self._validate_input(images, videos)
|
||||
res = self._get_mm_inputs(images, videos, processor)
|
||||
return res
|
||||
|
||||
|
||||
class LlavaNextVideoPlugin(BasePlugin):
|
||||
@override
|
||||
def process_messages(
|
||||
self,
|
||||
messages: Sequence[Dict[str, str]],
|
||||
images: Sequence["ImageInput"],
|
||||
videos: Sequence["VideoInput"],
|
||||
processor: Optional["ProcessorMixin"],
|
||||
) -> List[Dict[str, str]]:
|
||||
self._validate_input(images, videos)
|
||||
num_image_tokens = 0
|
||||
num_video_tokens = 0
|
||||
messages = deepcopy(messages)
|
||||
mm_inputs = self._get_mm_inputs(images, videos, processor)
|
||||
if "pixel_values" in mm_inputs:
|
||||
image_sizes = iter(mm_inputs["image_sizes"])
|
||||
height, width = get_image_size(to_numpy_array(mm_inputs["pixel_values"][0][0]))
|
||||
for message in messages:
|
||||
content = message["content"]
|
||||
|
||||
while self.image_token in content:
|
||||
image_size = next(image_sizes)
|
||||
orig_height, orig_width = image_size
|
||||
image_seqlen = processor._get_number_of_features(orig_height, orig_width, height, width)
|
||||
if processor.vision_feature_select_strategy == "default":
|
||||
image_seqlen -= 1
|
||||
num_image_tokens += 1
|
||||
content = content.replace(self.image_token, "{{image}}" * image_seqlen, 1)
|
||||
|
||||
message["content"] = content.replace("{{image}}", self.image_token)
|
||||
|
||||
if "pixel_values_videos" in mm_inputs:
|
||||
pixel_values_video = to_numpy_array(mm_inputs.get("pixel_values_videos")[0])
|
||||
height, width = get_image_size(pixel_values_video[0])
|
||||
num_frames = pixel_values_video.shape[0] # frame dim is always after batch dim
|
||||
image_seqlen = (height // processor.patch_size) * (width // processor.patch_size)
|
||||
video_seqlen = image_seqlen // 4 * num_frames # divide by 4 needed for avg pooling layer
|
||||
|
||||
for message in messages:
|
||||
content = message["content"]
|
||||
while self.video_token in content:
|
||||
num_video_tokens += 1
|
||||
content = content.replace(self.video_token, "{{video}}", 1)
|
||||
message["content"] = content.replace("{{video}}", self.video_token * video_seqlen)
|
||||
|
||||
if len(images) != num_image_tokens:
|
||||
raise ValueError("The number of images does not match the number of {} tokens".format(IMAGE_PLACEHOLDER))
|
||||
|
||||
if len(videos) != num_video_tokens:
|
||||
raise ValueError("The number of videos does not match the number of {} tokens".format(IMAGE_PLACEHOLDER))
|
||||
|
||||
return messages
|
||||
|
||||
@override
|
||||
def get_mm_inputs(
|
||||
self,
|
||||
images: Sequence["ImageInput"],
|
||||
videos: Sequence["VideoInput"],
|
||||
imglens: Sequence[int],
|
||||
vidlens: Sequence[int],
|
||||
seqlens: Sequence[int],
|
||||
processor: Optional["ProcessorMixin"],
|
||||
) -> Dict[str, Union[List[int], "torch.Tensor"]]:
|
||||
self._validate_input(images, videos)
|
||||
return self._get_mm_inputs(images, videos, processor)
|
||||
|
||||
|
||||
class PaliGemmaPlugin(BasePlugin):
|
||||
@override
|
||||
def process_messages(
|
||||
@@ -492,12 +616,78 @@ class Qwen2vlPlugin(BasePlugin):
|
||||
return self._get_mm_inputs(images, videos, processor)
|
||||
|
||||
|
||||
class VideoLlavaPlugin(BasePlugin):
|
||||
@override
|
||||
def process_messages(
|
||||
self,
|
||||
messages: Sequence[Dict[str, str]],
|
||||
images: Sequence["ImageInput"],
|
||||
videos: Sequence["VideoInput"],
|
||||
processor: Optional["ProcessorMixin"],
|
||||
) -> List[Dict[str, str]]:
|
||||
self._validate_input(images, videos)
|
||||
num_image_tokens = 0
|
||||
num_video_tokens = 0
|
||||
messages = deepcopy(messages)
|
||||
mm_inputs = self._get_mm_inputs(images, videos, processor)
|
||||
num_frames = 0
|
||||
exist_images = "pixel_values_images" in mm_inputs
|
||||
exist_videos = "pixel_values_videos" in mm_inputs
|
||||
if exist_videos or exist_images:
|
||||
if exist_images:
|
||||
height, width = get_image_size(to_numpy_array(mm_inputs.get("pixel_values_images")[0]))
|
||||
num_frames = 1
|
||||
if exist_videos:
|
||||
pixel_values_video = to_numpy_array(mm_inputs.get("pixel_values_videos")[0])
|
||||
height, width = get_image_size(pixel_values_video[0])
|
||||
num_frames = pixel_values_video.shape[0] # frame dim is always after batch dim
|
||||
image_seqlen = (height // processor.patch_size) * (width // processor.patch_size) + 1
|
||||
video_seqlen = image_seqlen * num_frames
|
||||
if processor.vision_feature_select_strategy == "default":
|
||||
image_seqlen -= 1
|
||||
for message in messages:
|
||||
content = message["content"]
|
||||
while self.image_token in content:
|
||||
num_image_tokens += 1
|
||||
content = content.replace(self.image_token, "{{image}}", 1)
|
||||
while self.video_token in content:
|
||||
num_video_tokens += 1
|
||||
content = content.replace(self.video_token, "{{video}}", 1)
|
||||
|
||||
content = content.replace("{{image}}", self.image_token * image_seqlen)
|
||||
message["content"] = content.replace("{{video}}", self.video_token * video_seqlen)
|
||||
|
||||
if len(images) != num_image_tokens:
|
||||
raise ValueError("The number of images does not match the number of {} tokens".format(self.image_token))
|
||||
|
||||
if len(videos) != num_video_tokens:
|
||||
raise ValueError("The number of videos does not match the number of {} tokens".format(self.video_token))
|
||||
|
||||
return messages
|
||||
|
||||
@override
|
||||
def get_mm_inputs(
|
||||
self,
|
||||
images: Sequence["ImageInput"],
|
||||
videos: Sequence["VideoInput"],
|
||||
imglens: Sequence[int],
|
||||
vidlens: Sequence[int],
|
||||
seqlens: Sequence[int],
|
||||
processor: Optional["ProcessorMixin"],
|
||||
) -> Dict[str, Union[List[int], "torch.Tensor"]]:
|
||||
self._validate_input(images, videos)
|
||||
return self._get_mm_inputs(images, videos, processor)
|
||||
|
||||
|
||||
PLUGINS = {
|
||||
"base": BasePlugin,
|
||||
"llava": LlavaPlugin,
|
||||
"llava_next": LlavaNextPlugin,
|
||||
"llava_next_video": LlavaNextVideoPlugin,
|
||||
"paligemma": PaliGemmaPlugin,
|
||||
"qwen2_vl": Qwen2vlPlugin,
|
||||
"pixtral": PixtralPlugin,
|
||||
"video_llava": VideoLlavaPlugin,
|
||||
}
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user