139 lines
5.7 KiB
Python
139 lines
5.7 KiB
Python
# coding=utf-8
|
|
# Copyright 2023 Microsoft Research & University of Wisconsin-Madison and the HuggingFace Inc. team. All rights reserved.
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
"""Llava model configuration"""
|
|
|
|
from ...configuration_utils import PretrainedConfig
|
|
from ...utils import logging
|
|
from ..auto import CONFIG_MAPPING, AutoConfig
|
|
|
|
|
|
logger = logging.get_logger(__name__)
|
|
|
|
|
|
class LlavaConfig(PretrainedConfig):
|
|
r"""
|
|
This is the configuration class to store the configuration of a [`LlavaForConditionalGeneration`]. It is used to instantiate an
|
|
Llava model according to the specified arguments, defining the model architecture. Instantiating a configuration
|
|
with the defaults will yield a similar configuration to that of the Llava-9B.
|
|
|
|
e.g. [llava-hf/llava-9b](https://huggingface.co/llava-hf/llava-9b)
|
|
|
|
Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
|
|
documentation from [`PretrainedConfig`] for more information.
|
|
|
|
Args:
|
|
vision_config (`Union[AutoConfig, dict]`, *optional*, defaults to `CLIPVisionConfig`):
|
|
The config object or dictionary of the vision backbone.
|
|
text_config (`Union[AutoConfig, dict]`, *optional*, defaults to `LlamaConfig`):
|
|
The config object or dictionary of the text backbone.
|
|
image_token_index (`int`, *optional*, defaults to 32000):
|
|
The image token index to encode the image prompt.
|
|
projector_hidden_act (`str`, *optional*, defaults to `"gelu"`):
|
|
The activation function used by the multimodal projector.
|
|
vision_feature_select_strategy (`str`, *optional*, defaults to `"default"`):
|
|
The feature selection strategy used to select the vision feature from the vision backbone.
|
|
Can be one of `"default"` or `"full"`.
|
|
vision_feature_layer (`Union[int, list[int]]`, *optional*, defaults to -2):
|
|
The index of the layer to select the vision feature. If multiple indices are provided,
|
|
the vision feature of the corresponding indices will be concatenated to form the
|
|
vision features.
|
|
image_seq_length (`int`, *optional*, defaults to 576):
|
|
Sequence length of one image embedding.
|
|
multimodal_projector_bias (`bool`, *optional*, defaults to `True`):
|
|
Whether to use bias in the multimodal projector.
|
|
|
|
Example:
|
|
|
|
```python
|
|
>>> from transformers import LlavaForConditionalGeneration, LlavaConfig, CLIPVisionConfig, LlamaConfig
|
|
|
|
>>> # Initializing a CLIP-vision config
|
|
>>> vision_config = CLIPVisionConfig()
|
|
|
|
>>> # Initializing a Llama config
|
|
>>> text_config = LlamaConfig()
|
|
|
|
>>> # Initializing a Llava llava-1.5-7b style configuration
|
|
>>> configuration = LlavaConfig(vision_config, text_config)
|
|
|
|
>>> # Initializing a model from the llava-1.5-7b style configuration
|
|
>>> model = LlavaForConditionalGeneration(configuration)
|
|
|
|
>>> # Accessing the model configuration
|
|
>>> configuration = model.config
|
|
```"""
|
|
|
|
model_type = "llava"
|
|
attribute_map = {
|
|
"image_token_id": "image_token_index",
|
|
}
|
|
sub_configs = {"text_config": AutoConfig, "vision_config": AutoConfig}
|
|
|
|
def __init__(
|
|
self,
|
|
vision_config=None,
|
|
text_config=None,
|
|
image_token_index=32000,
|
|
projector_hidden_act="gelu",
|
|
vision_feature_select_strategy="default",
|
|
vision_feature_layer=-2,
|
|
image_seq_length=576,
|
|
multimodal_projector_bias=True,
|
|
**kwargs,
|
|
):
|
|
self.image_token_index = image_token_index
|
|
self.projector_hidden_act = projector_hidden_act
|
|
self.image_seq_length = image_seq_length
|
|
|
|
if vision_feature_select_strategy not in ["default", "full"]:
|
|
raise ValueError(
|
|
"vision_feature_select_strategy should be one of 'default', 'full'."
|
|
f"Got: {vision_feature_select_strategy}"
|
|
)
|
|
|
|
self.vision_feature_select_strategy = vision_feature_select_strategy
|
|
self.vision_feature_layer = vision_feature_layer
|
|
|
|
if isinstance(vision_config, dict):
|
|
vision_config["model_type"] = (
|
|
vision_config["model_type"] if "model_type" in vision_config else "clip_vision_model"
|
|
)
|
|
vision_config = CONFIG_MAPPING[vision_config["model_type"]](**vision_config)
|
|
elif vision_config is None:
|
|
vision_config = CONFIG_MAPPING["clip_vision_model"](
|
|
intermediate_size=4096,
|
|
hidden_size=1024,
|
|
patch_size=14,
|
|
image_size=336,
|
|
num_hidden_layers=24,
|
|
num_attention_heads=16,
|
|
vocab_size=32000,
|
|
projection_dim=768,
|
|
)
|
|
|
|
self.vision_config = vision_config
|
|
|
|
if isinstance(text_config, dict):
|
|
text_config["model_type"] = text_config["model_type"] if "model_type" in text_config else "llama"
|
|
text_config = CONFIG_MAPPING[text_config["model_type"]](**text_config)
|
|
elif text_config is None:
|
|
text_config = CONFIG_MAPPING["llama"]()
|
|
|
|
self.text_config = text_config
|
|
self.multimodal_projector_bias = multimodal_projector_bias
|
|
|
|
super().__init__(**kwargs)
|
|
|
|
|
|
__all__ = ["LlavaConfig"]
|