File size: 15,366 Bytes
d1ceb73 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 |
# coding=utf-8
# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.
#
# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
# and OPT implementations in this library. It has been modified from its
# original forms to accommodate minor architectural differences compared
# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
#
# 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.
"""Idefics model configuration"""
from ...configuration_utils import PretrainedConfig
from ...utils import logging
logger = logging.get_logger(__name__)
class IdeficsVisionConfig(PretrainedConfig):
r"""
This is the configuration class to store the configuration of a [`IdeficsModel`]. It is used to instantiate an
Idefics 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 Idefics-9B.
e.g. [HuggingFaceM4/idefics-9b](https://huggingface.co/HuggingFaceM4/idefics-9b)
Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
documentation from [`PretrainedConfig`] for more information.
Args:
hidden_size (`int`, *optional*, defaults to 768):
Dimensionality of the encoder layers and the pooler layer. (elsewhere referred to as `hidden_size`)
image_size (`int`, *optional*, defaults to 224):
The size (resolution) of each image.
intermediate_size (`int`, *optional*, defaults to 5120):
Dimensionality of the "intermediate" (i.e., feed-forward) layer in the Transformer encoder.
patch_size (`int`, *optional*, defaults to 14):
The size (resolution) of each patch.
num_hidden_layers (`int`, *optional*, defaults to 32):
Number of hidden layers in the Transformer encoder.
num_attention_heads (`int`, *optional*, defaults to 16):
Number of attention heads for each attention layer in the Transformer encoder.
image_num_channels (`int`, *optional*, defaults to `3`):
Number of image channels.
hidden_act (`str` or `function`, *optional*, defaults to `"gelu"`):
The non-linear activation function (function or string) in the encoder and pooler. If string, `"gelu"`,
`"relu"`, `"selu"` and `"gelu_new"` `"quick_gelu"` are supported.
layer_norm_eps (`float`, *optional*, defaults to 1e-5):
The epsilon used by the layer normalization layers.
attention_dropout (`float`, *optional*, defaults to 0.0):
The dropout ratio for the attention probabilities.
initializer_range (`float`, *optional*, defaults to 0.02):
The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
initializer_factor (`float`, *optional*, defaults to 1.0):
A factor for initializing all weight matrices (should be kept to 1.0, used internally for initialization
testing).
initializer_range (`float`, *optional*, defaults to 0.02):
The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
"""
model_type = "idefics"
attribute_map = {
"hidden_size": "embed_dim",
}
def __init__(
self,
embed_dim=768,
image_size=224,
intermediate_size=5120,
patch_size=14,
num_hidden_layers=32,
num_attention_heads=16,
num_channels=3,
hidden_act="gelu",
layer_norm_eps=1e-5,
attention_dropout=0.0,
initializer_range=0.02,
initializer_factor=1.0,
**kwargs,
):
self.embed_dim = embed_dim
self.image_size = image_size
self.intermediate_size = intermediate_size
self.patch_size = patch_size
self.num_hidden_layers = num_hidden_layers
self.num_attention_heads = num_attention_heads
self.num_channels = num_channels
self.layer_norm_eps = layer_norm_eps
self.attention_dropout = attention_dropout
self.initializer_range = initializer_range
self.initializer_factor = initializer_factor
self.hidden_act = hidden_act
super().__init__(**kwargs)
class IdeficsPerceiverConfig(PretrainedConfig):
r"""
This is the configuration class to store the configuration of a [`IdeficsModel`]. It is used to instantiate an
Idefics 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 Idefics-9B.
e.g. [HuggingFaceM4/idefics-9b](https://huggingface.co/HuggingFaceM4/idefics-9b)
Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
documentation from [`PretrainedConfig`] for more information.
Args:
use_resampler (`bool`, *optional*, defaults to `False`):
Whether or not to use the resampler
resampler_n_latents (`int`, *optional*, defaults to ):
Number of latent embeddings to resample ("compress") the input sequence to (usually < 128).
resampler_depth (`int`, *optional*, defaults to 6):
Depth of the Perceiver Resampler (Transformer w/ cross attention). Should be shallow (< 3).
resampler_n_heads (`int`, *optional*, defaults to 16):
Number of heads in each Transformer block (for multi-headed self-attention).
resampler_head_dim (`int`, *optional*, defaults to 96):
Dimensionality of each head projection in the Transformer block.
qk_layer_norms_perceiver (`bool`, *optional*, defaults to `False`):
Whether or not to use qk layer norms in perceiver
"""
model_type = "idefics"
def __init__(
self,
use_resampler=False,
resampler_n_latents=64,
resampler_depth=6,
resampler_n_heads=16,
resampler_head_dim=96,
qk_layer_norms_perceiver=False,
**kwargs,
):
self.use_resampler = use_resampler
self.resampler_n_latents = resampler_n_latents
self.resampler_depth = resampler_depth
self.resampler_n_heads = resampler_n_heads
self.resampler_head_dim = resampler_head_dim
self.qk_layer_norms_perceiver = qk_layer_norms_perceiver
super().__init__(**kwargs)
class IdeficsConfig(PretrainedConfig):
r"""
This is the configuration class to store the configuration of a [`IdeficsModel`]. It is used to instantiate an
Idefics 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 Idefics-9B.
e.g. [HuggingFaceM4/idefics-9b](https://huggingface.co/HuggingFaceM4/idefics-9b)
Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
documentation from [`PretrainedConfig`] for more information.
Args:
additional_vocab_size (`int`, *optional*, defaults to 0):
Additional vocabulary size of the model, typically for the special "<img>" token. Additional vocab tokens
are always trainable whereas regular vocab tokens can be frozen or not.
vocab_size (`int`, *optional*, defaults to 32000):
Vocabulary size of the Idefics model. Defines the number of different tokens that can be represented by the
`inputs_ids` passed when calling [`~IdeficsModel`]
hidden_size (`int`, *optional*, defaults to 4096):
Dimension of the hidden representations.
intermediate_size (`int`, *optional*, defaults to 11008):
Dimension of the MLP representations.
num_hidden_layers (`int`, *optional*, defaults to 32):
Number of hidden layers in the Transformer encoder.
num_attention_heads (`int`, *optional*, defaults to 32):
Number of attention heads for each attention layer in the Transformer encoder.
dropout (`float`, *optional*, defaults to 0.0):
The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
The non-linear activation function (function or string) in the decoder.
initializer_range (`float`, *optional*, defaults to 0.02):
The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
alpha_initializer (`str`, *optional*, defaults to `"zeros"`):
Initialization type for the alphas.
alphas_initializer_range (`float`, *optional*, defaults to 0.0):
The standard deviation of the truncated_normal_initializer for initializing the alphas in the Gated Cross
Attention.
alpha_type (`str`, *optional*, defaults to `"float"`):
Whether the gating alphas should be vectors or single floats.
rms_norm_eps (`float`, *optional*, defaults to 1e-6):
The epsilon used by the rms normalization layers.
use_cache (`bool`, *optional*, defaults to `True`):
Whether or not the model should return the last key/values attentions (not used by all models). Only
relevant if `config.is_decoder=True`.
pad_token_id (`int`, *optional*, defaults to 0)
Padding token id.
bos_token_id (`int`, *optional*, defaults to 1)
Beginning of stream token id.
eos_token_id (`int`, *optional*, defaults to 2)
End of stream token id.
tie_word_embeddings(`bool`, *optional*, defaults to `False`):
Whether to tie weight embeddings
cross_layer_interval (`int`, *optional*, default to 1)
Interval for cross attention (from text to image) layers.
qk_layer_norms (`bool`, *optional*, defaults to `False`): Whether to add layer norm after q and k
freeze_text_layers (`bool`, *optional*, defaults to `True`): Whether to freeze text layers
freeze_text_module_exceptions (`bool`, *optional*, defaults to `[]`):
Exceptions to freezing text layers when `freeze_text_layers` is `True`
freeze_lm_head (`bool`, *optional*, defaults to `False`): Whether to freeze lm head
freeze_vision_layers (`bool`, *optional*, defaults to `True`): Whether to freeze vision layers
freeze_vision_module_exceptions (`bool`, *optional*, defaults to `[]`):
Exceptions to freezing vision layers when `freeze_vision_layers` is `True`
use_resampler (`bool`, *optional*, defaults to `False`): Whether to use the Resampler
vision_config (`IdeficsVisionConfig`, *optional*): Custom vision config or dict
perceiver_config (`IdeficsPerceiverConfig`, *optional*): Custom perceiver config or dict
Example:
```python
>>> from transformers import IdeficsModel, IdeficsConfig
>>> # Initializing a Idefics idefics-9b style configuration
>>> configuration = IdeficsConfig()
>>> # Initializing a model from the idefics-9b style configuration
>>> model = IdeficsModel(configuration)
>>> # Accessing the model configuration
>>> configuration = model.config
```"""
model_type = "idefics"
is_composition = False
def __init__(
self,
vocab_size=32000,
additional_vocab_size=0,
hidden_size=4096,
intermediate_size=11008,
num_hidden_layers=32,
num_attention_heads=32,
dropout=0.0,
hidden_act="silu",
initializer_range=0.02,
alpha_initializer="zeros",
alphas_initializer_range=0.0,
alpha_type="float",
rms_norm_eps=1e-6,
use_cache=True,
pad_token_id=0,
bos_token_id=1,
eos_token_id=2,
tie_word_embeddings=False,
cross_layer_interval=1,
qk_layer_norms=False,
freeze_text_layers=True,
freeze_text_module_exceptions=[],
freeze_lm_head=False,
freeze_vision_layers=True,
freeze_vision_module_exceptions=[],
use_resampler=False,
vision_config=None,
perceiver_config=None,
**kwargs,
):
self.vocab_size = vocab_size
self.additional_vocab_size = additional_vocab_size
self.hidden_size = hidden_size
self.intermediate_size = intermediate_size
self.num_hidden_layers = num_hidden_layers
self.num_attention_heads = num_attention_heads
self.dropout = dropout
self.hidden_act = hidden_act
self.initializer_range = initializer_range
self.alpha_initializer = alpha_initializer
self.alphas_initializer_range = alphas_initializer_range
self.alpha_type = alpha_type
self.rms_norm_eps = rms_norm_eps
self.use_cache = use_cache
self.cross_layer_interval = cross_layer_interval
self.qk_layer_norms = qk_layer_norms
self.freeze_vision_layers = freeze_vision_layers
self.freeze_text_layers = freeze_text_layers
self.freeze_text_module_exceptions = freeze_text_module_exceptions
self.freeze_vision_module_exceptions = freeze_vision_module_exceptions
self.freeze_lm_head = freeze_lm_head
self.use_resampler = use_resampler
if perceiver_config is None:
self.perceiver_config = IdeficsPerceiverConfig()
elif isinstance(perceiver_config, dict):
self.perceiver_config = IdeficsPerceiverConfig(**perceiver_config)
elif isinstance(perceiver_config, IdeficsPerceiverConfig):
self.perceiver_config = perceiver_config
if vision_config is None:
self.vision_config = IdeficsVisionConfig()
elif isinstance(vision_config, dict):
self.vision_config = IdeficsVisionConfig(**vision_config)
elif isinstance(vision_config, IdeficsVisionConfig):
self.vision_config = vision_config
super().__init__(
pad_token_id=pad_token_id,
bos_token_id=bos_token_id,
eos_token_id=eos_token_id,
tie_word_embeddings=tie_word_embeddings,
**kwargs,
)
# IMPORTANT: Do not do any __init__ args-based checks in the constructor, since
# PretrainedConfig.from_dict first instantiates the class with the config dict and only then
# updates the config object with `kwargs` from from_pretrained, so during the instantiation
# of this object many attributes have default values and haven't yet been overridden.
# Do any required checks inside `from_pretrained` once the superclass' `from_pretrained` was run.
|