[Core][VLM] Support image embeddings as input (#6613)

This commit is contained in:
Roger Wang
2024-08-12 16:16:06 +08:00
committed by GitHub
parent ec2affa8ae
commit e6e42e4b17
13 changed files with 517 additions and 138 deletions
+48 -32
View File
@@ -1,4 +1,4 @@
from typing import Iterable, List, Literal, Optional, Tuple, TypedDict
from typing import Iterable, List, Literal, Optional, Tuple, TypedDict, Union
import torch
from torch import nn
@@ -31,6 +31,25 @@ _KEYS_TO_MODIFY_MAPPING = {
}
class PaliGemmaImagePixelInputs(TypedDict):
type: Literal["pixel_values"]
data: torch.Tensor
"""Shape: (batch_size, num_channels, height, width)"""
class PaliGemmaImageEmbeddingInputs(TypedDict):
type: Literal["image_embeds"]
data: torch.Tensor
"""Shape: `(batch_size, image_feature_size, hidden_size)`
`hidden_size` must match the hidden size of language model backbone.
"""
PaliGemmaImageInputs = Union[PaliGemmaImagePixelInputs,
PaliGemmaImageEmbeddingInputs]
def get_max_paligemma_image_tokens(ctx: InputContext):
hf_config = ctx.get_hf_config(PaliGemmaConfig)
vision_config = hf_config.vision_config
@@ -107,15 +126,6 @@ class PaliGemmaMultiModalProjector(nn.Module):
return hidden_states
class PaliGemmaImagePixelInputs(TypedDict):
type: Literal["pixel_values"]
data: torch.Tensor
"""Shape: (batch_size, num_channels, height, width)"""
PaliGemmaImageInputs = PaliGemmaImagePixelInputs
@MULTIMODAL_REGISTRY.register_image_input_mapper()
@MULTIMODAL_REGISTRY.register_max_image_tokens(get_max_paligemma_image_tokens)
@INPUT_REGISTRY.register_dummy_data(dummy_data_for_paligemma)
@@ -163,18 +173,30 @@ class PaliGemmaForConditionalGeneration(nn.Module, SupportsVision):
def _parse_and_validate_image_input(
self, **kwargs: object) -> Optional[PaliGemmaImageInputs]:
pixel_values = kwargs.pop("pixel_values", None)
image_embeds = kwargs.pop("image_embeds", None)
if pixel_values is None:
if pixel_values is None and image_embeds is None:
return None
if not isinstance(pixel_values, torch.Tensor):
raise ValueError("Incorrect type of pixel values. "
f"Got type: {type(pixel_values)}")
if pixel_values is not None:
if not isinstance(pixel_values, torch.Tensor):
raise ValueError("Incorrect type of pixel values. "
f"Got type: {type(pixel_values)}")
return PaliGemmaImagePixelInputs(
type="pixel_values",
data=self._validate_pixel_values(pixel_values),
)
return PaliGemmaImagePixelInputs(
type="pixel_values",
data=self._validate_pixel_values(pixel_values),
)
if image_embeds is not None:
if not isinstance(image_embeds, torch.Tensor):
raise ValueError("Incorrect type of image embeddings. "
f"Got type: {type(image_embeds)}")
return PaliGemmaImageEmbeddingInputs(
type="image_embeds",
data=image_embeds,
)
raise AssertionError("This line should be unreachable.")
def _image_pixels_to_features(
self,
@@ -187,26 +209,20 @@ class PaliGemmaForConditionalGeneration(nn.Module, SupportsVision):
return image_features
def _process_image_pixels(
self,
inputs: PaliGemmaImagePixelInputs,
) -> torch.Tensor:
assert self.vision_tower is not None
pixel_values = inputs["data"]
return self._image_pixels_to_features(
self.vision_tower,
pixel_values,
)
def _process_image_input(
self,
image_input: PaliGemmaImageInputs,
) -> torch.Tensor:
if image_input["type"] == "image_embeds":
return image_input["data"]
assert self.vision_tower is not None
image_features = self._process_image_pixels(image_input, )
pixel_values = image_input["data"]
image_features = self._image_pixels_to_features(
self.vision_tower,
pixel_values,
)
return self.multi_modal_projector(image_features)