初始化项目，由ModelHub XC社区提供模型
Model: Shanghai_AI_Laboratory/Mono-InternVL-2B-S1-3 Source: Original Platform
This commit is contained in:
ModelHub XC
47
.gitattributes
vendored
Normal file
47
.gitattributes
vendored
Normal file
@@ -0,0 +1,47 @@
|
||||
*.7z filter=lfs diff=lfs merge=lfs -text
|
||||
*.arrow filter=lfs diff=lfs merge=lfs -text
|
||||
*.bin filter=lfs diff=lfs merge=lfs -text
|
||||
*.bin.* filter=lfs diff=lfs merge=lfs -text
|
||||
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
||||
*.ftz filter=lfs diff=lfs merge=lfs -text
|
||||
*.gz filter=lfs diff=lfs merge=lfs -text
|
||||
*.h5 filter=lfs diff=lfs merge=lfs -text
|
||||
*.joblib filter=lfs diff=lfs merge=lfs -text
|
||||
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
||||
*.model filter=lfs diff=lfs merge=lfs -text
|
||||
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
||||
*.onnx filter=lfs diff=lfs merge=lfs -text
|
||||
*.ot filter=lfs diff=lfs merge=lfs -text
|
||||
*.parquet filter=lfs diff=lfs merge=lfs -text
|
||||
*.pb filter=lfs diff=lfs merge=lfs -text
|
||||
*.pt filter=lfs diff=lfs merge=lfs -text
|
||||
*.pth filter=lfs diff=lfs merge=lfs -text
|
||||
*.rar filter=lfs diff=lfs merge=lfs -text
|
||||
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
||||
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
||||
*.tflite filter=lfs diff=lfs merge=lfs -text
|
||||
*.tgz filter=lfs diff=lfs merge=lfs -text
|
||||
*.xz filter=lfs diff=lfs merge=lfs -text
|
||||
*.zip filter=lfs diff=lfs merge=lfs -text
|
||||
*.zstandard filter=lfs diff=lfs merge=lfs -text
|
||||
*.tfevents* filter=lfs diff=lfs merge=lfs -text
|
||||
*.db* filter=lfs diff=lfs merge=lfs -text
|
||||
*.ark* filter=lfs diff=lfs merge=lfs -text
|
||||
**/*ckpt*data* filter=lfs diff=lfs merge=lfs -text
|
||||
**/*ckpt*.meta filter=lfs diff=lfs merge=lfs -text
|
||||
**/*ckpt*.index filter=lfs diff=lfs merge=lfs -text
|
||||
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
||||
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
||||
*.gguf* filter=lfs diff=lfs merge=lfs -text
|
||||
*.ggml filter=lfs diff=lfs merge=lfs -text
|
||||
*.llamafile* filter=lfs diff=lfs merge=lfs -text
|
||||
*.pt2 filter=lfs diff=lfs merge=lfs -text
|
||||
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
||||
*.npy filter=lfs diff=lfs merge=lfs -text
|
||||
*.npz filter=lfs diff=lfs merge=lfs -text
|
||||
*.pickle filter=lfs diff=lfs merge=lfs -text
|
||||
*.pkl filter=lfs diff=lfs merge=lfs -text
|
||||
*.tar filter=lfs diff=lfs merge=lfs -text
|
||||
*.wasm filter=lfs diff=lfs merge=lfs -text
|
||||
*.zst filter=lfs diff=lfs merge=lfs -text
|
||||
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
||||
42
README.md
Normal file
42
README.md
Normal file
@@ -0,0 +1,42 @@
|
||||
---
|
||||
base_model:
|
||||
- internlm/internlm2-chat-1_8b
|
||||
language:
|
||||
- multilingual
|
||||
library_name: transformers
|
||||
license: mit
|
||||
pipeline_tag: image-text-to-text
|
||||
tags:
|
||||
- internvl
|
||||
- vision
|
||||
- ocr
|
||||
- custom_code
|
||||
- moe
|
||||
base_model_relation: merge
|
||||
---
|
||||
|
||||
# Mono-InternVL-2B-S1-3
|
||||
|
||||
This repository contains the Mono-InternVL-2B model after **S1.1 concept learning**, **S1.2 semantic learning**, and **S1.3 alignment learning**.
|
||||
|
||||
Please refer to our [**paper**](https://huggingface.co/papers/2507.12566), [**project page**](https://internvl.github.io/blog/2024-10-10-Mono-InternVL/) and [**GitHub repository**](https://github.com/OpenGVLab/mono-internvl) for introduction and usage.
|
||||
|
||||
## Citation
|
||||
|
||||
If you find this project useful in your research, please consider citing:
|
||||
|
||||
```BibTeX
|
||||
@article{mono_internvl_v1,
|
||||
title={Mono-InternVL: Pushing the Boundaries of Monolithic Multimodal Large Language Models with Endogenous Visual Pre-training},
|
||||
author={Luo, Gen and Yang, Xue and Dou, Wenhan and Wang, Zhaokai and Liu, Jiawen and Dai, Jifeng and Qiao, Yu and Zhu, Xizhou},
|
||||
journal={arXiv preprint arXiv:2410.08202},
|
||||
year={2024}
|
||||
}
|
||||
|
||||
@article{mono_internvl_v1.5,
|
||||
title={Mono-InternVL-1.5: Towards Cheaper and Faster Monolithic Multimodal Large Language Models},
|
||||
author={Luo, Gen and Dou, Wenhan and Li, Wenhao and Wang, Zhaokai and Yang, Xue and Tian, Changyao and Li, Hao and Wang, Weiyun and Wang, Wenhai and Zhu, Xizhou and Qiao, Yu and Dai, Jifeng},
|
||||
journal={arXiv preprint arXiv:2507.12566},
|
||||
year={2025}
|
||||
}
|
||||
```
|
||||
11
added_tokens.json
Normal file
11
added_tokens.json
Normal file
@@ -0,0 +1,11 @@
|
||||
{
|
||||
"</box>": 92552,
|
||||
"</img>": 92545,
|
||||
"</quad>": 92548,
|
||||
"</ref>": 92550,
|
||||
"<IMG_CONTEXT>": 92546,
|
||||
"<box>": 92551,
|
||||
"<img>": 92544,
|
||||
"<quad>": 92547,
|
||||
"<ref>": 92549
|
||||
}
|
||||
129
config.json
Normal file
129
config.json
Normal file
@@ -0,0 +1,129 @@
|
||||
{
|
||||
"_commit_hash": null,
|
||||
"_name_or_path": "./work_dirs/internvl_chat_lite/internvl_chat_v1_5_internlm2_1_8b_dynamic12_res_alignment_pt_mlp_unfreeze_attn_5e-5/checkpoint-70000",
|
||||
"architectures": [
|
||||
"InternVLChatModel"
|
||||
],
|
||||
"auto_map": {
|
||||
"AutoConfig": "configuration_internvl_chat.InternVLChatConfig",
|
||||
"AutoModel": "modeling_internvl_chat.InternVLChatModel",
|
||||
"AutoModelForCausalLM": "modeling_internvl_chat.InternVLChatModel"
|
||||
},
|
||||
"downsample_ratio": 0.5,
|
||||
"dynamic_image_size": true,
|
||||
"force_image_size": 448,
|
||||
"llm_config": {
|
||||
"_name_or_path": "./my_pretrained/internlm2-chat-1_8b_ve",
|
||||
"add_cross_attention": false,
|
||||
"architectures": [
|
||||
"InternLM2VEForCausalLM"
|
||||
],
|
||||
"attn_implementation": "flash_attention_2",
|
||||
"auto_map": {
|
||||
"AutoConfig": "configuration_internlm2_.InternLM2Config",
|
||||
"AutoModel": "modeling_internlm2_ve.InternLM2VEForCausalLM",
|
||||
"AutoModelForCausalLM": "modeling_internlm2_ve.InternLM2VEForCausalLM"
|
||||
},
|
||||
"bad_words_ids": null,
|
||||
"begin_suppress_tokens": null,
|
||||
"bias": false,
|
||||
"bos_token_id": 1,
|
||||
"chunk_size_feed_forward": 0,
|
||||
"cross_attention_hidden_size": null,
|
||||
"decoder_start_token_id": null,
|
||||
"diversity_penalty": 0.0,
|
||||
"do_sample": false,
|
||||
"early_stopping": false,
|
||||
"encoder_no_repeat_ngram_size": 0,
|
||||
"eos_token_id": 2,
|
||||
"expert_dp_comm": "none",
|
||||
"exponential_decay_length_penalty": null,
|
||||
"finetuning_task": null,
|
||||
"forced_bos_token_id": null,
|
||||
"forced_eos_token_id": null,
|
||||
"hidden_act": "silu",
|
||||
"hidden_size": 2048,
|
||||
"id2label": {
|
||||
"0": "LABEL_0",
|
||||
"1": "LABEL_1"
|
||||
},
|
||||
"initializer_range": 0.02,
|
||||
"intermediate_size": 8192,
|
||||
"is_decoder": false,
|
||||
"is_encoder_decoder": false,
|
||||
"label2id": {
|
||||
"LABEL_0": 0,
|
||||
"LABEL_1": 1
|
||||
},
|
||||
"length_penalty": 1.0,
|
||||
"max_length": 20,
|
||||
"max_position_embeddings": 32768,
|
||||
"min_length": 0,
|
||||
"model_type": "internlm2",
|
||||
"moe_top_k": 1,
|
||||
"no_repeat_ngram_size": 0,
|
||||
"num_attention_heads": 16,
|
||||
"num_beam_groups": 1,
|
||||
"num_beams": 1,
|
||||
"num_expert": -1,
|
||||
"num_hidden_layers": 24,
|
||||
"num_key_value_heads": 8,
|
||||
"num_return_sequences": 1,
|
||||
"output_attentions": false,
|
||||
"output_hidden_states": true,
|
||||
"output_scores": false,
|
||||
"pad_token_id": 2,
|
||||
"prefix": null,
|
||||
"problem_type": null,
|
||||
"pruned_heads": {},
|
||||
"remove_invalid_values": false,
|
||||
"repetition_penalty": 1.0,
|
||||
"return_dict": true,
|
||||
"return_dict_in_generate": false,
|
||||
"rms_norm_eps": 1e-05,
|
||||
"rope_scaling": null,
|
||||
"rope_theta": 1000000,
|
||||
"sep_token_id": null,
|
||||
"suppress_tokens": null,
|
||||
"task_specific_params": null,
|
||||
"temperature": 1.0,
|
||||
"tf_legacy_loss": false,
|
||||
"tie_encoder_decoder": false,
|
||||
"tie_word_embeddings": false,
|
||||
"tokenizer_class": null,
|
||||
"top_k": 50,
|
||||
"top_p": 1.0,
|
||||
"torch_dtype": "bfloat16",
|
||||
"torchscript": false,
|
||||
"transformers_version": "4.37.2",
|
||||
"typical_p": 1.0,
|
||||
"use_bfloat16": false,
|
||||
"use_cache": false,
|
||||
"vocab_size": 92553
|
||||
},
|
||||
"max_dynamic_patch": 24,
|
||||
"min_dynamic_patch": 1,
|
||||
"model_type": "internvl_chat",
|
||||
"pad2square": false,
|
||||
"ps_version": "v2",
|
||||
"select_layer": -1,
|
||||
"template": "internlm2-chat",
|
||||
"torch_dtype": "bfloat16",
|
||||
"transformers_version": null,
|
||||
"use_backbone_lora": 0,
|
||||
"use_llm_lora": 0,
|
||||
"use_thumbnail": true,
|
||||
"vision_config": {
|
||||
"architectures": [
|
||||
"InternVisionPatchModel"
|
||||
],
|
||||
"auto_map": {
|
||||
"AutoConfig": "configuration_intern_patch.InternVisionPatchConfig",
|
||||
"AutoModel": "modeling_intern_patch.InternVisionPatchModel"
|
||||
},
|
||||
"hidden_size": 1024,
|
||||
"image_size": 448,
|
||||
"model_type": "intern_vit_patch",
|
||||
"patch_size": 14
|
||||
}
|
||||
}
|
||||
1
configuration.json
Normal file
1
configuration.json
Normal file
@@ -0,0 +1 @@
|
||||
{"framework": "pytorch", "task": "image-text-to-text", "allow_remote": true}
|
||||
91
configuration_intern_patch.py
Normal file
91
configuration_intern_patch.py
Normal file
@@ -0,0 +1,91 @@
|
||||
# --------------------------------------------------------
|
||||
# InternVL
|
||||
# Copyright (c) 2024 OpenGVLab
|
||||
# Licensed under The MIT License [see LICENSE for details]
|
||||
# --------------------------------------------------------
|
||||
import os
|
||||
from typing import Union
|
||||
|
||||
from transformers.configuration_utils import PretrainedConfig
|
||||
from transformers.utils import logging
|
||||
|
||||
logger = logging.get_logger(__name__)
|
||||
|
||||
|
||||
|
||||
class InternVisionPatchConfig(PretrainedConfig):
|
||||
r"""
|
||||
This is the configuration class to store the configuration of a [`InternVisionModel`]. It is used to
|
||||
instantiate a vision encoder according to the specified arguments, defining the model architecture.
|
||||
|
||||
Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
|
||||
documentation from [`PretrainedConfig`] for more information.
|
||||
|
||||
Args:
|
||||
num_channels (`int`, *optional*, defaults to 3):
|
||||
Number of color channels in the input images (e.g., 3 for RGB).
|
||||
patch_size (`int`, *optional*, defaults to 14):
|
||||
The size (resolution) of each patch.
|
||||
image_size (`int`, *optional*, defaults to 224):
|
||||
The size (resolution) of each image.
|
||||
qkv_bias (`bool`, *optional*, defaults to `False`):
|
||||
Whether to add a bias to the queries and values in the self-attention layers.
|
||||
hidden_size (`int`, *optional*, defaults to 3200):
|
||||
Dimensionality of the encoder layers and the pooler layer.
|
||||
num_attention_heads (`int`, *optional*, defaults to 25):
|
||||
Number of attention heads for each attention layer in the Transformer encoder.
|
||||
intermediate_size (`int`, *optional*, defaults to 12800):
|
||||
Dimensionality of the "intermediate" (i.e., feed-forward) layer in the Transformer encoder.
|
||||
qk_normalization (`bool`, *optional*, defaults to `True`):
|
||||
Whether to normalize the queries and keys in the self-attention layers.
|
||||
num_hidden_layers (`int`, *optional*, defaults to 48):
|
||||
Number of hidden layers in the Transformer encoder.
|
||||
use_flash_attn (`bool`, *optional*, defaults to `True`):
|
||||
Whether to use flash attention mechanism.
|
||||
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"` ``"gelu"` are supported.
|
||||
layer_norm_eps (`float`, *optional*, defaults to 1e-6):
|
||||
The epsilon used by the layer normalization layers.
|
||||
dropout (`float`, *optional*, defaults to 0.0):
|
||||
The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
|
||||
drop_path_rate (`float`, *optional*, defaults to 0.0):
|
||||
Dropout rate for stochastic depth.
|
||||
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 0.1):
|
||||
A factor for layer scale.
|
||||
"""
|
||||
|
||||
model_type = 'intern_vit_patch'
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
patch_size=14,
|
||||
image_size=224,
|
||||
hidden_size=3200,
|
||||
**kwargs,
|
||||
):
|
||||
super().__init__(**kwargs)
|
||||
|
||||
self.hidden_size = hidden_size
|
||||
self.patch_size = patch_size
|
||||
self.image_size = image_size
|
||||
|
||||
@classmethod
|
||||
def from_pretrained(cls, pretrained_model_name_or_path: Union[str, os.PathLike], **kwargs) -> 'PretrainedConfig':
|
||||
config_dict, kwargs = cls.get_config_dict(pretrained_model_name_or_path, **kwargs)
|
||||
|
||||
if 'vision_config' in config_dict:
|
||||
config_dict = config_dict['vision_config']
|
||||
|
||||
if 'model_type' in config_dict and hasattr(cls, 'model_type') and config_dict['model_type'] != cls.model_type:
|
||||
logger.warning(
|
||||
f"You are using a model of type {config_dict['model_type']} to instantiate a model of type "
|
||||
f'{cls.model_type}. This is not supported for all configurations of models and can yield errors.'
|
||||
)
|
||||
|
||||
|
||||
return cls.from_dict(config_dict, **kwargs)
|
||||
150
configuration_internlm2.py
Normal file
150
configuration_internlm2.py
Normal file
@@ -0,0 +1,150 @@
|
||||
# Copyright (c) The InternLM team and The HuggingFace Inc. team. All rights reserved.
|
||||
#
|
||||
# This code is based on transformers/src/transformers/models/llama/configuration_llama.py
|
||||
#
|
||||
# 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.
|
||||
""" InternLM2 model configuration"""
|
||||
|
||||
from transformers.configuration_utils import PretrainedConfig
|
||||
from transformers.utils import logging
|
||||
|
||||
logger = logging.get_logger(__name__)
|
||||
|
||||
INTERNLM2_PRETRAINED_CONFIG_ARCHIVE_MAP = {}
|
||||
|
||||
|
||||
# Modified from transformers.model.llama.configuration_llama.LlamaConfig
|
||||
class InternLM2Config(PretrainedConfig):
|
||||
r"""
|
||||
This is the configuration class to store the configuration of a [`InternLM2Model`]. It is used to instantiate
|
||||
an InternLM2 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 InternLM2-7B.
|
||||
|
||||
Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
|
||||
documentation from [`PretrainedConfig`] for more information.
|
||||
|
||||
|
||||
Args:
|
||||
vocab_size (`int`, *optional*, defaults to 32000):
|
||||
Vocabulary size of the InternLM2 model. Defines the number of different tokens that can be represented by the
|
||||
`inputs_ids` passed when calling [`InternLM2Model`]
|
||||
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.
|
||||
num_key_value_heads (`int`, *optional*):
|
||||
This is the number of key_value heads that should be used to implement Grouped Query Attention. If
|
||||
`num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
|
||||
`num_key_value_heads=1 the model will use Multi Query Attention (MQA) otherwise GQA is used. When
|
||||
converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
|
||||
by meanpooling all the original heads within that group. For more details checkout [this
|
||||
paper](https://arxiv.org/pdf/2305.13245.pdf). If it is not specified, will default to
|
||||
`num_attention_heads`.
|
||||
hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
|
||||
The non-linear activation function (function or string) in the decoder.
|
||||
max_position_embeddings (`int`, *optional*, defaults to 2048):
|
||||
The maximum sequence length that this model might ever be used with. Typically set this to something large
|
||||
just in case (e.g., 512 or 1024 or 2048).
|
||||
initializer_range (`float`, *optional*, defaults to 0.02):
|
||||
The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
|
||||
rms_norm_eps (`float`, *optional*, defaults to 1e-12):
|
||||
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`.
|
||||
tie_word_embeddings(`bool`, *optional*, defaults to `False`):
|
||||
Whether to tie weight embeddings
|
||||
Example:
|
||||
|
||||
"""
|
||||
model_type = 'internlm2'
|
||||
_auto_class = 'AutoConfig'
|
||||
|
||||
def __init__( # pylint: disable=W0102
|
||||
self,
|
||||
vocab_size=103168,
|
||||
hidden_size=4096,
|
||||
intermediate_size=11008,
|
||||
num_hidden_layers=32,
|
||||
num_attention_heads=32,
|
||||
num_key_value_heads=None,
|
||||
hidden_act='silu',
|
||||
max_position_embeddings=2048,
|
||||
initializer_range=0.02,
|
||||
rms_norm_eps=1e-6,
|
||||
use_cache=True,
|
||||
pad_token_id=0,
|
||||
bos_token_id=1,
|
||||
eos_token_id=2,
|
||||
tie_word_embeddings=False,
|
||||
bias=True,
|
||||
rope_theta=10000,
|
||||
rope_scaling=None,
|
||||
attn_implementation='eager',
|
||||
**kwargs,
|
||||
):
|
||||
self.vocab_size = vocab_size
|
||||
self.max_position_embeddings = max_position_embeddings
|
||||
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.bias = bias
|
||||
|
||||
if num_key_value_heads is None:
|
||||
num_key_value_heads = num_attention_heads
|
||||
self.num_key_value_heads = num_key_value_heads
|
||||
|
||||
self.hidden_act = hidden_act
|
||||
self.initializer_range = initializer_range
|
||||
self.rms_norm_eps = rms_norm_eps
|
||||
self.use_cache = use_cache
|
||||
self.rope_theta = rope_theta
|
||||
self.rope_scaling = rope_scaling
|
||||
self._rope_scaling_validation()
|
||||
|
||||
self.attn_implementation = attn_implementation
|
||||
if self.attn_implementation is None:
|
||||
self.attn_implementation = 'eager'
|
||||
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,
|
||||
)
|
||||
|
||||
def _rope_scaling_validation(self):
|
||||
"""
|
||||
Validate the `rope_scaling` configuration.
|
||||
"""
|
||||
if self.rope_scaling is None:
|
||||
return
|
||||
|
||||
if not isinstance(self.rope_scaling, dict) or len(self.rope_scaling) != 2:
|
||||
raise ValueError(
|
||||
'`rope_scaling` must be a dictionary with with two fields, `type` and `factor`, '
|
||||
f'got {self.rope_scaling}'
|
||||
)
|
||||
rope_scaling_type = self.rope_scaling.get('type', None)
|
||||
rope_scaling_factor = self.rope_scaling.get('factor', None)
|
||||
if rope_scaling_type is None or rope_scaling_type not in ['linear', 'dynamic']:
|
||||
raise ValueError(
|
||||
f"`rope_scaling`'s type field must be one of ['linear', 'dynamic'], got {rope_scaling_type}"
|
||||
)
|
||||
if rope_scaling_factor is None or not isinstance(rope_scaling_factor, float) or rope_scaling_factor < 1.0:
|
||||
raise ValueError(f"`rope_scaling`'s factor field must be a float >= 1, got {rope_scaling_factor}")
|
||||
106
configuration_internvl_chat.py
Normal file
106
configuration_internvl_chat.py
Normal file
@@ -0,0 +1,106 @@
|
||||
# --------------------------------------------------------
|
||||
# InternVL
|
||||
# Copyright (c) 2024 OpenGVLab
|
||||
# Licensed under The MIT License [see LICENSE for details]
|
||||
# --------------------------------------------------------
|
||||
|
||||
import copy
|
||||
|
||||
from .configuration_internlm2 import InternLM2Config
|
||||
from transformers import AutoConfig, LlamaConfig, Qwen2Config
|
||||
from transformers.configuration_utils import PretrainedConfig
|
||||
from transformers.utils import logging
|
||||
|
||||
from .configuration_intern_patch import InternVisionPatchConfig
|
||||
|
||||
logger = logging.get_logger(__name__)
|
||||
|
||||
|
||||
class InternVLChatConfig(PretrainedConfig):
|
||||
model_type = 'internvl_chat'
|
||||
is_composition = True
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
vision_config=None,
|
||||
llm_config=None,
|
||||
use_backbone_lora=0,
|
||||
use_llm_lora=0,
|
||||
pad2square=False,
|
||||
select_layer=-1,
|
||||
force_image_size=None,
|
||||
downsample_ratio=0.5,
|
||||
template=None,
|
||||
dynamic_image_size=False,
|
||||
use_thumbnail=False,
|
||||
ps_version='v1',
|
||||
min_dynamic_patch=1,
|
||||
max_dynamic_patch=6,
|
||||
**kwargs):
|
||||
super().__init__(**kwargs)
|
||||
|
||||
if vision_config is None:
|
||||
vision_config = {}
|
||||
logger.info('vision_config is None. Initializing the InternVisionConfig with default values.')
|
||||
|
||||
if llm_config is None:
|
||||
llm_config = {}
|
||||
logger.info('llm_config is None. Initializing the LlamaConfig config with default values (`LlamaConfig`).')
|
||||
|
||||
if vision_config and vision_config['model_type']=='intern_vit_patch':
|
||||
self.vision_config = InternVisionPatchConfig(**vision_config)
|
||||
else:
|
||||
raise ValueError('Unsupported vision model type: {}'.format(vision_config['model_type']))
|
||||
if llm_config['architectures'][0] == 'LlamaForCausalLM':
|
||||
self.llm_config = LlamaConfig(**llm_config)
|
||||
elif llm_config['architectures'][0] == 'InternLM2ForCausalLM':
|
||||
self.llm_config = InternLM2Config(**llm_config)
|
||||
elif llm_config['architectures'][0] == 'InternLM2VEForCausalLM':
|
||||
self.llm_config = InternLM2Config(**llm_config)
|
||||
elif llm_config['architectures'][0] == 'Qwen2ForCausalLM':
|
||||
self.llm_config = Qwen2Config(**llm_config)
|
||||
else:
|
||||
raise ValueError('Unsupported architecture: {}'.format(llm_config['architectures'][0]))
|
||||
self.use_backbone_lora = use_backbone_lora
|
||||
self.use_llm_lora = use_llm_lora
|
||||
self.pad2square = pad2square
|
||||
self.select_layer = select_layer
|
||||
self.force_image_size = force_image_size
|
||||
self.downsample_ratio = downsample_ratio
|
||||
self.template = template
|
||||
self.dynamic_image_size = dynamic_image_size
|
||||
self.use_thumbnail = use_thumbnail
|
||||
self.ps_version = ps_version # pixel shuffle version
|
||||
self.min_dynamic_patch = min_dynamic_patch
|
||||
self.max_dynamic_patch = max_dynamic_patch
|
||||
|
||||
logger.info(f'vision_select_layer: {self.select_layer}')
|
||||
logger.info(f'ps_version: {self.ps_version}')
|
||||
logger.info(f'min_dynamic_patch: {self.min_dynamic_patch}')
|
||||
logger.info(f'max_dynamic_patch: {self.max_dynamic_patch}')
|
||||
|
||||
def to_dict(self):
|
||||
"""
|
||||
Serializes this instance to a Python dictionary. Override the default [`~PretrainedConfig.to_dict`].
|
||||
|
||||
Returns:
|
||||
`Dict[str, any]`: Dictionary of all the attributes that make up this configuration instance,
|
||||
"""
|
||||
output = copy.deepcopy(self.__dict__)
|
||||
output['vision_config'] = self.vision_config.to_dict()
|
||||
output['llm_config'] = self.llm_config.to_dict()
|
||||
output['model_type'] = self.__class__.model_type
|
||||
output['use_backbone_lora'] = self.use_backbone_lora
|
||||
output['use_llm_lora'] = self.use_llm_lora
|
||||
output['pad2square'] = self.pad2square
|
||||
output['select_layer'] = self.select_layer
|
||||
output['force_image_size'] = self.force_image_size
|
||||
output['downsample_ratio'] = self.downsample_ratio
|
||||
output['template'] = self.template
|
||||
output['dynamic_image_size'] = self.dynamic_image_size
|
||||
output['use_thumbnail'] = self.use_thumbnail
|
||||
output['ps_version'] = self.ps_version
|
||||
output['min_dynamic_patch'] = self.min_dynamic_patch
|
||||
output['max_dynamic_patch'] = self.max_dynamic_patch
|
||||
|
||||
return output
|
||||
406
conversation.py
Normal file
406
conversation.py
Normal file
@@ -0,0 +1,406 @@
|
||||
"""
|
||||
Conversation prompt templates.
|
||||
|
||||
We kindly request that you import fastchat instead of copying this file if you wish to use it.
|
||||
If you have changes in mind, please contribute back so the community can benefit collectively and continue to maintain these valuable templates.
|
||||
"""
|
||||
|
||||
import dataclasses
|
||||
from enum import IntEnum, auto
|
||||
from typing import Any, Dict, List, Tuple, Union
|
||||
|
||||
|
||||
class SeparatorStyle(IntEnum):
|
||||
"""Separator styles."""
|
||||
|
||||
ADD_COLON_SINGLE = auto()
|
||||
ADD_COLON_TWO = auto()
|
||||
ADD_COLON_SPACE_SINGLE = auto()
|
||||
NO_COLON_SINGLE = auto()
|
||||
NO_COLON_TWO = auto()
|
||||
ADD_NEW_LINE_SINGLE = auto()
|
||||
LLAMA2 = auto()
|
||||
CHATGLM = auto()
|
||||
CHATML = auto()
|
||||
CHATINTERN = auto()
|
||||
DOLLY = auto()
|
||||
RWKV = auto()
|
||||
PHOENIX = auto()
|
||||
ROBIN = auto()
|
||||
FALCON_CHAT = auto()
|
||||
CHATGLM3 = auto()
|
||||
INTERNVL_ZH = auto()
|
||||
MPT = auto()
|
||||
|
||||
|
||||
@dataclasses.dataclass
|
||||
class Conversation:
|
||||
"""A class that manages prompt templates and keeps all conversation history."""
|
||||
|
||||
# The name of this template
|
||||
name: str
|
||||
# The template of the system prompt
|
||||
system_template: str = '{system_message}'
|
||||
# The system message
|
||||
system_message: str = ''
|
||||
# The names of two roles
|
||||
roles: Tuple[str] = ('USER', 'ASSISTANT')
|
||||
# All messages. Each item is (role, message).
|
||||
messages: List[List[str]] = ()
|
||||
# The number of few shot examples
|
||||
offset: int = 0
|
||||
# The separator style and configurations
|
||||
sep_style: SeparatorStyle = SeparatorStyle.ADD_COLON_SINGLE
|
||||
sep: str = '\n'
|
||||
sep2: str = None
|
||||
# Stop criteria (the default one is EOS token)
|
||||
stop_str: Union[str, List[str]] = None
|
||||
# Stops generation if meeting any token in this list
|
||||
stop_token_ids: List[int] = None
|
||||
|
||||
def get_prompt(self) -> str:
|
||||
"""Get the prompt for generation."""
|
||||
system_prompt = self.system_template.format(system_message=self.system_message)
|
||||
if self.sep_style == SeparatorStyle.ADD_COLON_SINGLE:
|
||||
ret = system_prompt + self.sep
|
||||
for role, message in self.messages:
|
||||
if message:
|
||||
ret += role + ': ' + message + self.sep
|
||||
else:
|
||||
ret += role + ':'
|
||||
return ret
|
||||
elif self.sep_style == SeparatorStyle.ADD_COLON_TWO:
|
||||
seps = [self.sep, self.sep2]
|
||||
ret = system_prompt + seps[0]
|
||||
for i, (role, message) in enumerate(self.messages):
|
||||
if message:
|
||||
ret += role + ': ' + message + seps[i % 2]
|
||||
else:
|
||||
ret += role + ':'
|
||||
return ret
|
||||
elif self.sep_style == SeparatorStyle.ADD_COLON_SPACE_SINGLE:
|
||||
ret = system_prompt + self.sep
|
||||
for role, message in self.messages:
|
||||
if message:
|
||||
ret += role + ': ' + message + self.sep
|
||||
else:
|
||||
ret += role + ': ' # must be end with a space
|
||||
return ret
|
||||
elif self.sep_style == SeparatorStyle.ADD_NEW_LINE_SINGLE:
|
||||
ret = '' if system_prompt == '' else system_prompt + self.sep
|
||||
for role, message in self.messages:
|
||||
if message:
|
||||
ret += role + '\n' + message + self.sep
|
||||
else:
|
||||
ret += role + '\n'
|
||||
return ret
|
||||
elif self.sep_style == SeparatorStyle.NO_COLON_SINGLE:
|
||||
ret = system_prompt
|
||||
for role, message in self.messages:
|
||||
if message:
|
||||
ret += role + message + self.sep
|
||||
else:
|
||||
ret += role
|
||||
return ret
|
||||
elif self.sep_style == SeparatorStyle.NO_COLON_TWO:
|
||||
seps = [self.sep, self.sep2]
|
||||
ret = system_prompt
|
||||
for i, (role, message) in enumerate(self.messages):
|
||||
if message:
|
||||
ret += role + message + seps[i % 2]
|
||||
else:
|
||||
ret += role
|
||||
return ret
|
||||
elif self.sep_style == SeparatorStyle.RWKV:
|
||||
ret = system_prompt
|
||||
for i, (role, message) in enumerate(self.messages):
|
||||
if message:
|
||||
ret += (
|
||||
role
|
||||
+ ': '
|
||||
+ message.replace('\r\n', '\n').replace('\n\n', '\n')
|
||||
)
|
||||
ret += '\n\n'
|
||||
else:
|
||||
ret += role + ':'
|
||||
return ret
|
||||
elif self.sep_style == SeparatorStyle.LLAMA2:
|
||||
seps = [self.sep, self.sep2]
|
||||
if self.system_message:
|
||||
ret = system_prompt
|
||||
else:
|
||||
ret = '[INST] '
|
||||
for i, (role, message) in enumerate(self.messages):
|
||||
tag = self.roles[i % 2]
|
||||
if message:
|
||||
if i == 0:
|
||||
ret += message + ' '
|
||||
else:
|
||||
ret += tag + ' ' + message + seps[i % 2]
|
||||
else:
|
||||
ret += tag
|
||||
return ret
|
||||
elif self.sep_style == SeparatorStyle.CHATGLM:
|
||||
# source: https://huggingface.co/THUDM/chatglm-6b/blob/1d240ba371910e9282298d4592532d7f0f3e9f3e/modeling_chatglm.py#L1302-L1308
|
||||
# source2: https://huggingface.co/THUDM/chatglm2-6b/blob/e186c891cf64310ac66ef10a87e6635fa6c2a579/modeling_chatglm.py#L926
|
||||
round_add_n = 1 if self.name == 'chatglm2' else 0
|
||||
if system_prompt:
|
||||
ret = system_prompt + self.sep
|
||||
else:
|
||||
ret = ''
|
||||
|
||||
for i, (role, message) in enumerate(self.messages):
|
||||
if i % 2 == 0:
|
||||
ret += f'[Round {i//2 + round_add_n}]{self.sep}'
|
||||
|
||||
if message:
|
||||
ret += f'{role}：{message}{self.sep}'
|
||||
else:
|
||||
ret += f'{role}：'
|
||||
return ret
|
||||
elif self.sep_style == SeparatorStyle.CHATML:
|
||||
ret = '' if system_prompt == '' else system_prompt + self.sep + '\n'
|
||||
for role, message in self.messages:
|
||||
if message:
|
||||
ret += role + '\n' + message + self.sep + '\n'
|
||||
else:
|
||||
ret += role + '\n'
|
||||
return ret
|
||||
elif self.sep_style == SeparatorStyle.CHATGLM3:
|
||||
ret = ''
|
||||
if self.system_message:
|
||||
ret += system_prompt
|
||||
for role, message in self.messages:
|
||||
if message:
|
||||
ret += role + '\n' + ' ' + message
|
||||
else:
|
||||
ret += role
|
||||
return ret
|
||||
elif self.sep_style == SeparatorStyle.CHATINTERN:
|
||||
# source: https://huggingface.co/internlm/internlm-chat-7b-8k/blob/bd546fa984b4b0b86958f56bf37f94aa75ab8831/modeling_internlm.py#L771
|
||||
seps = [self.sep, self.sep2]
|
||||
ret = system_prompt
|
||||
for i, (role, message) in enumerate(self.messages):
|
||||
# if i % 2 == 0:
|
||||
# ret += "<s>"
|
||||
if message:
|
||||
ret += role + ':' + message + seps[i % 2] + '\n'
|
||||
else:
|
||||
ret += role + ':'
|
||||
return ret
|
||||
elif self.sep_style == SeparatorStyle.DOLLY:
|
||||
seps = [self.sep, self.sep2]
|
||||
ret = system_prompt
|
||||
for i, (role, message) in enumerate(self.messages):
|
||||
if message:
|
||||
ret += role + ':\n' + message + seps[i % 2]
|
||||
if i % 2 == 1:
|
||||
ret += '\n\n'
|
||||
else:
|
||||
ret += role + ':\n'
|
||||
return ret
|
||||
elif self.sep_style == SeparatorStyle.PHOENIX:
|
||||
ret = system_prompt
|
||||
for role, message in self.messages:
|
||||
if message:
|
||||
ret += role + ': ' + '<s>' + message + '</s>'
|
||||
else:
|
||||
ret += role + ': ' + '<s>'
|
||||
return ret
|
||||
elif self.sep_style == SeparatorStyle.ROBIN:
|
||||
ret = system_prompt + self.sep
|
||||
for role, message in self.messages:
|
||||
if message:
|
||||
ret += role + ':\n' + message + self.sep
|
||||
else:
|
||||
ret += role + ':\n'
|
||||
return ret
|
||||
elif self.sep_style == SeparatorStyle.FALCON_CHAT:
|
||||
ret = ''
|
||||
if self.system_message:
|
||||
ret += system_prompt + self.sep
|
||||
for role, message in self.messages:
|
||||
if message:
|
||||
ret += role + ': ' + message + self.sep
|
||||
else:
|
||||
ret += role + ':'
|
||||
|
||||
return ret
|
||||
elif self.sep_style == SeparatorStyle.INTERNVL_ZH:
|
||||
seps = [self.sep2, self.sep]
|
||||
ret = self.system_message + seps[0]
|
||||
for i, (role, message) in enumerate(self.messages):
|
||||
if message:
|
||||
ret += role + ': ' + message + seps[i % 2]
|
||||
else:
|
||||
ret += role + ':'
|
||||
return ret
|
||||
elif self.sep_style == SeparatorStyle.MPT:
|
||||
ret = system_prompt + self.sep
|
||||
for role, message in self.messages:
|
||||
if message:
|
||||
if type(message) is tuple:
|
||||
message, _, _ = message
|
||||
ret += role + message + self.sep
|
||||
else:
|
||||
ret += role
|
||||
return ret
|
||||
else:
|
||||
raise ValueError(f'Invalid style: {self.sep_style}')
|
||||
|
||||
def set_system_message(self, system_message: str):
|
||||
"""Set the system message."""
|
||||
self.system_message = system_message
|
||||
|
||||
def append_message(self, role: str, message: str):
|
||||
"""Append a new message."""
|
||||
self.messages.append([role, message])
|
||||
|
||||
def update_last_message(self, message: str):
|
||||
"""Update the last output.
|
||||
|
||||
The last message is typically set to be None when constructing the prompt,
|
||||
so we need to update it in-place after getting the response from a model.
|
||||
"""
|
||||
self.messages[-1][1] = message
|
||||
|
||||
def to_gradio_chatbot(self):
|
||||
"""Convert the conversation to gradio chatbot format."""
|
||||
ret = []
|
||||
for i, (role, msg) in enumerate(self.messages[self.offset :]):
|
||||
if i % 2 == 0:
|
||||
ret.append([msg, None])
|
||||
else:
|
||||
ret[-1][-1] = msg
|
||||
return ret
|
||||
|
||||
def to_openai_api_messages(self):
|
||||
"""Convert the conversation to OpenAI chat completion format."""
|
||||
ret = [{'role': 'system', 'content': self.system_message}]
|
||||
|
||||
for i, (_, msg) in enumerate(self.messages[self.offset :]):
|
||||
if i % 2 == 0:
|
||||
ret.append({'role': 'user', 'content': msg})
|
||||
else:
|
||||
if msg is not None:
|
||||
ret.append({'role': 'assistant', 'content': msg})
|
||||
return ret
|
||||
|
||||
def copy(self):
|
||||
return Conversation(
|
||||
name=self.name,
|
||||
system_template=self.system_template,
|
||||
system_message=self.system_message,
|
||||
roles=self.roles,
|
||||
messages=[[x, y] for x, y in self.messages],
|
||||
offset=self.offset,
|
||||
sep_style=self.sep_style,
|
||||
sep=self.sep,
|
||||
sep2=self.sep2,
|
||||
stop_str=self.stop_str,
|
||||
stop_token_ids=self.stop_token_ids,
|
||||
)
|
||||
|
||||
def dict(self):
|
||||
return {
|
||||
'template_name': self.name,
|
||||
'system_message': self.system_message,
|
||||
'roles': self.roles,
|
||||
'messages': self.messages,
|
||||
'offset': self.offset,
|
||||
}
|
||||
|
||||
|
||||
# A global registry for all conversation templates
|
||||
conv_templates: Dict[str, Conversation] = {}
|
||||
|
||||
|
||||
def register_conv_template(template: Conversation, override: bool = False):
|
||||
"""Register a new conversation template."""
|
||||
if not override:
|
||||
assert (
|
||||
template.name not in conv_templates
|
||||
), f'{template.name} has been registered.'
|
||||
|
||||
conv_templates[template.name] = template
|
||||
|
||||
|
||||
def get_conv_template(name: str) -> Conversation:
|
||||
"""Get a conversation template."""
|
||||
return conv_templates[name].copy()
|
||||
|
||||
|
||||
# InternVL-Chat-V1-1 template
|
||||
register_conv_template(
|
||||
Conversation(
|
||||
name='internvl_zh',
|
||||
system_template='',
|
||||
roles=('<human>', '<bot>'),
|
||||
sep_style=SeparatorStyle.INTERNVL_ZH,
|
||||
sep='</s>',
|
||||
sep2=' ',
|
||||
)
|
||||
)
|
||||
|
||||
|
||||
# Both Hermes-2 and internlm2-chat are chatml-format conversation templates. The difference
|
||||
# is that during training, the preprocessing function for the Hermes-2 template doesn't add
|
||||
# <s> at the beginning of the tokenized sequence, while the internlm2-chat template does.
|
||||
# Therefore, they are completely equivalent during inference.
|
||||
register_conv_template(
|
||||
Conversation(
|
||||
name='Hermes-2',
|
||||
system_template='<|im_start|>system\n{system_message}',
|
||||
# note: The new system prompt was not used here to avoid changes in benchmark performance.
|
||||
# system_message='我是书生·万象，英文名是InternVL，是由上海人工智能实验室及多家合作单位联合开发的多模态大语言模型。',
|
||||
system_message='你是由上海人工智能实验室联合商汤科技开发的书生多模态大模型，英文名叫InternVL, 是一个有用无害的人工智能助手。',
|
||||
roles=('<|im_start|>user\n', '<|im_start|>assistant\n'),
|
||||
sep_style=SeparatorStyle.MPT,
|
||||
sep='<|im_end|>',
|
||||
stop_token_ids=[
|
||||
2,
|
||||
6,
|
||||
7,
|
||||
8,
|
||||
],
|
||||
stop_str='<|endoftext|>',
|
||||
)
|
||||
)
|
||||
|
||||
|
||||
register_conv_template(
|
||||
Conversation(
|
||||
name='internlm2-chat',
|
||||
system_template='<|im_start|>system\n{system_message}',
|
||||
# note: The new system prompt was not used here to avoid changes in benchmark performance.
|
||||
# system_message='我是书生·万象，英文名是InternVL，是由上海人工智能实验室及多家合作单位联合开发的多模态大语言模型。',
|
||||
system_message='你是由上海人工智能实验室联合商汤科技开发的书生多模态大模型，英文名叫InternVL, 是一个有用无害的人工智能助手。',
|
||||
roles=('<|im_start|>user\n', '<|im_start|>assistant\n'),
|
||||
sep_style=SeparatorStyle.MPT,
|
||||
sep='<|im_end|>',
|
||||
stop_token_ids=[
|
||||
2,
|
||||
92543,
|
||||
92542
|
||||
]
|
||||
)
|
||||
)
|
||||
|
||||
|
||||
register_conv_template(
|
||||
Conversation(
|
||||
name='phi3-chat',
|
||||
system_template='<|system|>\n{system_message}',
|
||||
# note: The new system prompt was not used here to avoid changes in benchmark performance.
|
||||
# system_message='我是书生·万象，英文名是InternVL，是由上海人工智能实验室及多家合作单位联合开发的多模态大语言模型。',
|
||||
system_message='你是由上海人工智能实验室联合商汤科技开发的书生多模态大模型，英文名叫InternVL, 是一个有用无害的人工智能助手。',
|
||||
roles=('<|user|>\n', '<|assistant|>\n'),
|
||||
sep_style=SeparatorStyle.MPT,
|
||||
sep='<|end|>',
|
||||
stop_token_ids=[
|
||||
2,
|
||||
32000,
|
||||
32007
|
||||
]
|
||||
)
|
||||
)
|
||||
BIN
examples/image1.jpg
Normal file
BIN
examples/image1.jpg
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 76 KiB |
4
generation_config.json
Normal file
4
generation_config.json
Normal file
@@ -0,0 +1,4 @@
|
||||
{
|
||||
"_from_model_config": true,
|
||||
"transformers_version": "4.37.2"
|
||||
}
|
||||
BIN
images/fig1.jpg
Normal file
BIN
images/fig1.jpg
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 802 KiB |
BIN
images/fig2.jpg
Normal file
BIN
images/fig2.jpg
Normal file
Binary file not shown.
|
After Width: | Height: | Size: 570 KiB |
3
model-00001-of-00002.safetensors
Normal file
3
model-00001-of-00002.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:562f1a24f81023ff1cca3dbe22d3cfd937237b12550c129f3f6f70bac9f3ca44
|
||||
size 4971164000
|
||||
3
model-00002-of-00002.safetensors
Normal file
3
model-00002-of-00002.safetensors
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:c5c8cf2460188642fd87ca46468178789d5711ac51a2f451c7b2a84bf8688ba5
|
||||
size 1251579280
|
||||
260
model.safetensors.index.json
Normal file
260
model.safetensors.index.json
Normal file
@@ -0,0 +1,260 @@
|
||||
{
|
||||
"metadata": {
|
||||
"total_size": 6222710784
|
||||
},
|
||||
"weight_map": {
|
||||
"language_model.model.layers.0.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.0.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.0.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.0.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.0.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.0.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.0.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.0.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.0.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.0.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.1.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.1.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.1.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.1.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.1.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.1.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.1.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.1.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.1.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.1.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.10.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.10.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.10.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.10.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.10.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.10.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.10.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.10.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.10.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.10.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.11.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.11.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.11.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.11.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.11.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.11.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.11.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.11.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.11.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.11.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.12.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.12.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.12.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.12.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.12.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.12.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.12.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.12.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.12.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.12.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.13.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.13.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.13.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.13.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.13.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.13.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.13.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.13.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.13.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.13.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.14.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.14.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.14.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.14.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.14.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.14.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.14.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.14.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.14.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.14.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.15.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.15.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.15.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.15.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.15.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.15.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.15.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.15.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.15.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.15.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.16.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.16.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.16.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.16.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.16.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.16.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.16.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.16.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.16.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.16.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.17.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.17.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.17.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.17.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.17.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.17.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.17.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.17.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.17.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.17.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.18.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.18.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.18.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.18.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.18.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.18.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.18.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.18.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.18.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.18.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.19.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.19.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.19.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.19.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.19.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.19.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.19.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.19.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.19.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.19.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.2.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.2.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.2.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.2.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.2.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.2.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.2.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.2.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.2.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.2.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.20.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.20.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.20.attention_norm.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.20.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.20.feed_forward.w2.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.20.feed_forward.w3.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.20.feed_forward_ve.w1.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.20.feed_forward_ve.w2.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.20.feed_forward_ve.w3.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.20.ffn_norm.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.21.attention.wo.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.21.attention.wqkv.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.21.attention_norm.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.21.feed_forward.w1.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.21.feed_forward.w2.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.21.feed_forward.w3.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.21.feed_forward_ve.w1.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.21.feed_forward_ve.w2.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.21.feed_forward_ve.w3.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.21.ffn_norm.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.22.attention.wo.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.22.attention.wqkv.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.22.attention_norm.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.22.feed_forward.w1.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.22.feed_forward.w2.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.22.feed_forward.w3.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.22.feed_forward_ve.w1.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.22.feed_forward_ve.w2.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.22.feed_forward_ve.w3.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.22.ffn_norm.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.23.attention.wo.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.23.attention.wqkv.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.23.attention_norm.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.23.feed_forward.w1.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.23.feed_forward.w2.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.23.feed_forward.w3.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.23.feed_forward_ve.w1.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.23.feed_forward_ve.w2.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.23.feed_forward_ve.w3.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.23.ffn_norm.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.layers.3.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.3.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.3.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.3.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.3.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.3.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.3.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.3.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.3.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.3.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.4.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.4.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.4.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.4.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.4.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.4.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.4.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.4.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.4.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.4.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.5.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.5.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.5.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.5.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.5.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.5.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.5.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.5.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.5.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.5.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.6.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.6.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.6.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.6.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.6.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.6.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.6.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.6.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.6.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.6.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.7.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.7.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.7.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.7.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.7.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.7.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.7.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.7.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.7.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.7.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.8.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.8.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.8.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.8.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.8.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.8.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.8.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.8.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.8.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.8.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.9.attention.wo.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.9.attention.wqkv.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.9.attention_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.9.feed_forward.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.9.feed_forward.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.9.feed_forward.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.9.feed_forward_ve.w1.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.9.feed_forward_ve.w2.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.9.feed_forward_ve.w3.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.layers.9.ffn_norm.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.model.norm.weight": "model-00002-of-00002.safetensors",
|
||||
"language_model.model.tok_embeddings.weight": "model-00001-of-00002.safetensors",
|
||||
"language_model.output.weight": "model-00002-of-00002.safetensors",
|
||||
"mlp1.0.bias": "model-00002-of-00002.safetensors",
|
||||
"mlp1.0.weight": "model-00002-of-00002.safetensors",
|
||||
"mlp1.1.bias": "model-00002-of-00002.safetensors",
|
||||
"mlp1.1.weight": "model-00002-of-00002.safetensors",
|
||||
"mlp1.3.bias": "model-00002-of-00002.safetensors",
|
||||
"mlp1.3.weight": "model-00002-of-00002.safetensors",
|
||||
"vision_model.embeddings.class_embedding": "model-00001-of-00002.safetensors",
|
||||
"vision_model.embeddings.patch_embedding.bias": "model-00001-of-00002.safetensors",
|
||||
"vision_model.embeddings.patch_embedding.weight": "model-00001-of-00002.safetensors",
|
||||
"vision_model.embeddings.position_embedding": "model-00001-of-00002.safetensors"
|
||||
}
|
||||
}
|
||||
122
modeling_intern_patch.py
Normal file
122
modeling_intern_patch.py
Normal file
@@ -0,0 +1,122 @@
|
||||
# --------------------------------------------------------
|
||||
# InternVL
|
||||
# Copyright (c) 2024 OpenGVLab
|
||||
# Licensed under The MIT License [see LICENSE for details]
|
||||
# --------------------------------------------------------
|
||||
from typing import Optional, Tuple, Union
|
||||
|
||||
import torch
|
||||
import torch.nn.functional as F
|
||||
import torch.utils.checkpoint
|
||||
from torch import nn
|
||||
from transformers.modeling_outputs import (
|
||||
BaseModelOutputWithPooling)
|
||||
from transformers.modeling_utils import PreTrainedModel
|
||||
from transformers.utils import logging
|
||||
|
||||
from .configuration_intern_patch import InternVisionPatchConfig
|
||||
|
||||
logger = logging.get_logger(__name__)
|
||||
|
||||
|
||||
class InternVisionEmbeddings(nn.Module):
|
||||
def __init__(self, config: InternVisionPatchConfig):
|
||||
super().__init__()
|
||||
self.config = config
|
||||
self.embed_dim = config.hidden_size
|
||||
self.image_size = config.image_size
|
||||
self.patch_size = config.patch_size
|
||||
|
||||
self.class_embedding = nn.Parameter(
|
||||
torch.randn(1, 1, self.embed_dim),
|
||||
)
|
||||
|
||||
self.patch_embedding = nn.Conv2d(
|
||||
in_channels=3, out_channels=self.embed_dim, kernel_size=self.patch_size, stride=self.patch_size
|
||||
)
|
||||
|
||||
self.num_patches = (self.image_size // self.patch_size) ** 2
|
||||
self.num_positions = self.num_patches + 1
|
||||
|
||||
self.position_embedding = nn.Parameter(torch.randn(1, self.num_positions, self.embed_dim))
|
||||
|
||||
def _get_pos_embed(self, pos_embed, H, W):
|
||||
target_dtype = pos_embed.dtype
|
||||
pos_embed = pos_embed.float().reshape(
|
||||
1, self.image_size // self.patch_size, self.image_size // self.patch_size, -1).permute(0, 3, 1, 2)
|
||||
pos_embed = F.interpolate(pos_embed, size=(H, W), mode='bicubic', align_corners=False). \
|
||||
reshape(1, -1, H * W).permute(0, 2, 1).to(target_dtype)
|
||||
return pos_embed
|
||||
|
||||
def forward(self, pixel_values: torch.FloatTensor) -> torch.Tensor:
|
||||
target_dtype = self.patch_embedding.weight.dtype
|
||||
patch_embeds = self.patch_embedding(pixel_values) # shape = [*, channel, width, height]
|
||||
batch_size, _, height, width = patch_embeds.shape
|
||||
patch_embeds = patch_embeds.flatten(2).transpose(1, 2)
|
||||
class_embeds = self.class_embedding.expand(batch_size, 1, -1).to(target_dtype)
|
||||
embeddings = torch.cat([class_embeds, patch_embeds], dim=1)
|
||||
position_embedding = torch.cat([
|
||||
self.position_embedding[:, :1, :],
|
||||
self._get_pos_embed(self.position_embedding[:, 1:, :], height, width)
|
||||
], dim=1)
|
||||
embeddings = embeddings + position_embedding.to(target_dtype)
|
||||
return embeddings
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
class InternVisionPatchModel(PreTrainedModel):
|
||||
main_input_name = 'pixel_values'
|
||||
config_class = InternVisionPatchConfig
|
||||
_no_split_modules = ['InternVisionEncoderLayer']
|
||||
|
||||
def __init__(self, config: InternVisionPatchConfig):
|
||||
super().__init__(config)
|
||||
self.config = config
|
||||
self.embeddings = InternVisionEmbeddings(config)
|
||||
def resize_pos_embeddings(self, old_size, new_size, patch_size):
|
||||
pos_emb = self.embeddings.position_embedding
|
||||
_, num_positions, embed_dim = pos_emb.shape
|
||||
cls_emb = pos_emb[:, :1, :]
|
||||
pos_emb = pos_emb[:, 1:, :].reshape(1, old_size // patch_size, old_size // patch_size, -1).permute(0, 3, 1, 2)
|
||||
pos_emb = F.interpolate(pos_emb.float(), size=new_size // patch_size, mode='bicubic', align_corners=False)
|
||||
pos_emb = pos_emb.to(cls_emb.dtype).reshape(1, embed_dim, -1).permute(0, 2, 1)
|
||||
pos_emb = torch.cat([cls_emb, pos_emb], dim=1)
|
||||
self.embeddings.position_embedding = nn.Parameter(pos_emb)
|
||||
self.embeddings.image_size = new_size
|
||||
logger.info('Resized position embeddings from {} to {}'.format(old_size, new_size))
|
||||
|
||||
def get_input_embeddings(self):
|
||||
return self.embeddings
|
||||
|
||||
|
||||
def forward(
|
||||
self,
|
||||
pixel_values: Optional[torch.FloatTensor] = None,
|
||||
output_hidden_states: Optional[bool] = None,
|
||||
return_dict: Optional[bool] = None,
|
||||
pixel_embeds: Optional[torch.FloatTensor] = None,
|
||||
) -> Union[Tuple, BaseModelOutputWithPooling]:
|
||||
|
||||
return_dict = return_dict if return_dict is not None else self.config.use_return_dict
|
||||
|
||||
if pixel_values is None:
|
||||
raise ValueError('You have to specify pixel_values')
|
||||
|
||||
|
||||
if len(pixel_values.shape) == 4:
|
||||
hidden_states = self.embeddings(pixel_values)[:,1:]
|
||||
else:
|
||||
raise ValueError(f'wrong pixel_values size: {pixel_values.shape}')
|
||||
|
||||
|
||||
if not return_dict:
|
||||
return (hidden_states, None,None)
|
||||
|
||||
return BaseModelOutputWithPooling(
|
||||
last_hidden_state=hidden_states,
|
||||
pooler_output=None,
|
||||
hidden_states=None,
|
||||
attentions=None,
|
||||
)
|
||||
1458
modeling_internlm2_ve.py
Normal file
1458
modeling_internlm2_ve.py
Normal file
File diff suppressed because it is too large
Load Diff
468
modeling_internvl_chat.py
Normal file
468
modeling_internvl_chat.py
Normal file
@@ -0,0 +1,468 @@
|
||||
# --------------------------------------------------------
|
||||
# InternVL
|
||||
# Copyright (c) 2024 OpenGVLab
|
||||
# Licensed under The MIT License [see LICENSE for details]
|
||||
# --------------------------------------------------------
|
||||
import warnings
|
||||
from typing import Any, List, Optional, Tuple, Union
|
||||
|
||||
import torch.distributed as dist
|
||||
import torch.utils.checkpoint
|
||||
import transformers
|
||||
from .conversation import get_conv_template
|
||||
from .modeling_internlm2_ve import InternLM2VEForCausalLM
|
||||
from peft import LoraConfig, get_peft_model
|
||||
from torch import nn
|
||||
from torch.nn import CrossEntropyLoss
|
||||
from transformers import (AutoModel, GenerationConfig, LlamaForCausalLM,
|
||||
LlamaTokenizer, Qwen2ForCausalLM)
|
||||
from transformers.modeling_outputs import CausalLMOutputWithPast
|
||||
from transformers.modeling_utils import PreTrainedModel
|
||||
from transformers.utils import ModelOutput, logging
|
||||
import math
|
||||
from .configuration_internvl_chat import InternVLChatConfig
|
||||
from .modeling_intern_patch import InternVisionPatchModel
|
||||
from dataclasses import dataclass
|
||||
|
||||
logger = logging.get_logger(__name__)
|
||||
|
||||
|
||||
def version_cmp(v1, v2, op='eq'):
|
||||
import operator
|
||||
|
||||
from packaging import version
|
||||
op_func = getattr(operator, op)
|
||||
return op_func(version.parse(v1), version.parse(v2))
|
||||
|
||||
@dataclass
|
||||
class CausalLMOutputWithVisualMask(ModelOutput):
|
||||
"""
|
||||
Base class for causal language model (or autoregressive) outputs.
|
||||
|
||||
Args:
|
||||
loss (`torch.FloatTensor` of shape `(1,)`, *optional*, returned when `labels` is provided):
|
||||
Language modeling loss (for next-token prediction).
|
||||
logits (`torch.FloatTensor` of shape `(batch_size, sequence_length, config.vocab_size)`):
|
||||
Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
|
||||
past_key_values (`tuple(tuple(torch.FloatTensor))`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`):
|
||||
Tuple of `tuple(torch.FloatTensor)` of length `config.n_layers`, with each tuple having 2 tensors of shape
|
||||
`(batch_size, num_heads, sequence_length, embed_size_per_head)`)
|
||||
|
||||
Contains pre-computed hidden-states (key and values in the self-attention blocks) that can be used (see
|
||||
`past_key_values` input) to speed up sequential decoding.
|
||||
hidden_states (`tuple(torch.FloatTensor)`, *optional*, returned when `output_hidden_states=True` is passed or when `config.output_hidden_states=True`):
|
||||
Tuple of `torch.FloatTensor` (one for the output of the embeddings, if the model has an embedding layer, +
|
||||
one for the output of each layer) of shape `(batch_size, sequence_length, hidden_size)`.
|
||||
|
||||
Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.
|
||||
attentions (`tuple(torch.FloatTensor)`, *optional*, returned when `output_attentions=True` is passed or when `config.output_attentions=True`):
|
||||
Tuple of `torch.FloatTensor` (one for each layer) of shape `(batch_size, num_heads, sequence_length,
|
||||
sequence_length)`.
|
||||
|
||||
Attentions weights after the attention softmax, used to compute the weighted average in the self-attention
|
||||
heads.
|
||||
"""
|
||||
|
||||
loss: Optional[torch.FloatTensor] = None
|
||||
logits: torch.FloatTensor = None
|
||||
past_key_values: Optional[Tuple[Tuple[torch.FloatTensor]]] = None
|
||||
hidden_states: Optional[Tuple[torch.FloatTensor]] = None
|
||||
attentions: Optional[Tuple[torch.FloatTensor]] = None
|
||||
visual_token_mask : Optional[torch.FloatTensor] = None
|
||||
|
||||
|
||||
class InternVLChatModel(PreTrainedModel):
|
||||
config_class = InternVLChatConfig
|
||||
main_input_name = 'pixel_values'
|
||||
_no_split_modules = ['InternVisionPatchModel', 'LlamaDecoderLayer', 'InternLM2DecoderLayer',
|
||||
'Phi3DecoderLayer', 'Qwen2DecoderLayer']
|
||||
_supports_flash_attn_2 = True
|
||||
|
||||
def __init__(self, config: InternVLChatConfig, vision_model=None, language_model=None):
|
||||
super().__init__(config)
|
||||
|
||||
assert version_cmp(transformers.__version__, '4.37.0', 'ge')
|
||||
image_size = config.force_image_size or config.vision_config.image_size
|
||||
patch_size = config.vision_config.patch_size
|
||||
self.patch_size = patch_size
|
||||
self.select_layer = config.select_layer
|
||||
self.template = config.template
|
||||
self.num_image_token = int((image_size // patch_size) ** 2 * (config.downsample_ratio ** 2))
|
||||
self.downsample_ratio = config.downsample_ratio
|
||||
self.ps_version = config.ps_version
|
||||
self.llm_arch_name = config.llm_config.architectures[0]
|
||||
self.use_visual_token_mask=False
|
||||
|
||||
logger.info(f'num_image_token: {self.num_image_token}')
|
||||
logger.info(f'ps_version: {self.ps_version}')
|
||||
if vision_model is not None:
|
||||
self.vision_model = vision_model
|
||||
else:
|
||||
self.vision_model = InternVisionPatchModel(config.vision_config)
|
||||
|
||||
|
||||
if language_model is not None:
|
||||
self.language_model = language_model
|
||||
else:
|
||||
if config.llm_config.architectures[0] == 'LlamaForCausalLM':
|
||||
self.language_model = LlamaForCausalLM(config.llm_config)
|
||||
elif config.llm_config.architectures[0] == 'Qwen2ForCausalLM':
|
||||
self.language_model = Qwen2ForCausalLM(config.llm_config)
|
||||
elif config.llm_config.architectures[0]=='InternLM2VEForCausalLM':
|
||||
self.language_model=InternLM2VEForCausalLM(config.llm_config)
|
||||
self.use_visual_token_mask=True
|
||||
else:
|
||||
raise NotImplementedError(f'{config.llm_config.architectures[0]} is not implemented.')
|
||||
|
||||
vit_hidden_size = config.vision_config.hidden_size
|
||||
llm_hidden_size = config.llm_config.hidden_size
|
||||
|
||||
self.mlp1 = nn.Sequential(
|
||||
nn.LayerNorm(vit_hidden_size * int(1 / self.downsample_ratio) ** 2),
|
||||
nn.Linear(vit_hidden_size * int(1 / self.downsample_ratio) ** 2, llm_hidden_size),
|
||||
nn.GELU(),
|
||||
nn.Linear(llm_hidden_size, llm_hidden_size)
|
||||
)
|
||||
|
||||
self.img_context_token_id = None
|
||||
self.conv_template = get_conv_template(self.template)
|
||||
if hasattr(config, 'system_message'):
|
||||
self.system_message = config.system_message
|
||||
else:
|
||||
self.system_message = self.conv_template.system_message
|
||||
self.num_samples = 0
|
||||
|
||||
if config.use_backbone_lora:
|
||||
self.wrap_backbone_lora(r=config.use_backbone_lora, lora_alpha=2 * config.use_backbone_lora)
|
||||
|
||||
if config.use_llm_lora:
|
||||
self.wrap_llm_lora(r=config.use_llm_lora, lora_alpha=2 * config.use_llm_lora)
|
||||
|
||||
def wrap_backbone_lora(self, r=128, lora_alpha=256, lora_dropout=0.05):
|
||||
lora_config = LoraConfig(
|
||||
r=r,
|
||||
target_modules=['attn.qkv', 'attn.proj', 'mlp.fc1', 'mlp.fc2'],
|
||||
lora_alpha=lora_alpha,
|
||||
lora_dropout=lora_dropout,
|
||||
)
|
||||
self.vision_model = get_peft_model(self.vision_model, lora_config)
|
||||
self.vision_model.print_trainable_parameters()
|
||||
|
||||
def wrap_llm_lora(self, r=128, lora_alpha=256, lora_dropout=0.05):
|
||||
# Determine the target modules based on the architecture of the language model
|
||||
if self.llm_arch_name == 'InternLM2ForCausalLM' or self.llm_arch_name == 'InternLM2VEForCausalLM':
|
||||
target_modules = ['attention.wqkv', 'attention.wo', 'feed_forward.w1', 'feed_forward.w2', 'feed_forward.w3']
|
||||
elif self.llm_arch_name in ['Qwen2ForCausalLM', 'LlamaForCausalLM']:
|
||||
target_modules = ['self_attn.q_proj', 'self_attn.k_proj', 'self_attn.v_proj', 'self_attn.o_proj',
|
||||
'mlp.gate_proj', 'mlp.down_proj', 'mlp.up_proj']
|
||||
else:
|
||||
raise NotImplemented
|
||||
lora_config = LoraConfig(
|
||||
r=r,
|
||||
target_modules=target_modules,
|
||||
lora_alpha=lora_alpha,
|
||||
lora_dropout=lora_dropout,
|
||||
task_type='CAUSAL_LM'
|
||||
)
|
||||
self.language_model = get_peft_model(self.language_model, lora_config)
|
||||
self.language_model.enable_input_require_grads()
|
||||
self.language_model.print_trainable_parameters()
|
||||
|
||||
def forward(
|
||||
self,
|
||||
pixel_values: torch.FloatTensor,
|
||||
input_ids: torch.LongTensor = None,
|
||||
attention_mask: Optional[torch.Tensor] = None,
|
||||
position_ids: Optional[torch.LongTensor] = None,
|
||||
image_flags: Optional[torch.LongTensor] = None,
|
||||
past_key_values: Optional[List[torch.FloatTensor]] = None,
|
||||
labels: Optional[torch.LongTensor] = None,
|
||||
use_cache: Optional[bool] = None,
|
||||
output_attentions: Optional[bool] = None,
|
||||
output_hidden_states: Optional[bool] = None,
|
||||
return_dict: Optional[bool] = None,
|
||||
) -> Union[Tuple, CausalLMOutputWithVisualMask]:
|
||||
return_dict = return_dict if return_dict is not None else self.config.use_return_dict
|
||||
|
||||
image_flags = image_flags.squeeze(-1)
|
||||
input_embeds = self.language_model.get_input_embeddings()(input_ids).clone()
|
||||
|
||||
vit_embeds = self.extract_feature(pixel_values)
|
||||
vit_embeds = vit_embeds[image_flags == 1]
|
||||
vit_batch_size = pixel_values.shape[0]
|
||||
|
||||
B, N, C = input_embeds.shape
|
||||
input_embeds = input_embeds.reshape(B * N, C)
|
||||
|
||||
if torch.distributed.is_initialized() and torch.distributed.get_rank() == 0:
|
||||
print(f'dynamic ViT batch size: {vit_batch_size}, images per sample: {vit_batch_size / B}, dynamic token length: {N}')
|
||||
|
||||
input_ids = input_ids.reshape(B * N)
|
||||
selected = (input_ids == self.img_context_token_id)
|
||||
try:
|
||||
input_embeds[selected] = input_embeds[selected] * 0.0 + vit_embeds.reshape(-1, C)
|
||||
ignore_flag = False
|
||||
except Exception as e:
|
||||
vit_embeds = vit_embeds.reshape(-1, C)
|
||||
print(f'warning: {e}, input_embeds[selected].shape={input_embeds[selected].shape}, '
|
||||
f'vit_embeds.shape={vit_embeds.shape}')
|
||||
n_token = selected.sum()
|
||||
input_embeds[selected] = input_embeds[selected] * 0.0 + vit_embeds[:n_token]
|
||||
ignore_flag = True
|
||||
|
||||
input_embeds = input_embeds.reshape(B, N, C)
|
||||
|
||||
if self.llm_arch_name=='InternLM2VEForCausalLM':
|
||||
visual_token_mask = selected.reshape(B,N,1).to(input_embeds.dtype)
|
||||
outputs = self.language_model(
|
||||
inputs_embeds=input_embeds,
|
||||
attention_mask=attention_mask,
|
||||
position_ids=position_ids,
|
||||
past_key_values=past_key_values,
|
||||
use_cache=use_cache,
|
||||
output_attentions=output_attentions,
|
||||
output_hidden_states=output_hidden_states,
|
||||
return_dict=return_dict,
|
||||
visual_token_mask=visual_token_mask
|
||||
)
|
||||
else:
|
||||
outputs = self.language_model(
|
||||
inputs_embeds=input_embeds,
|
||||
attention_mask=attention_mask,
|
||||
position_ids=position_ids,
|
||||
past_key_values=past_key_values,
|
||||
use_cache=use_cache,
|
||||
output_attentions=output_attentions,
|
||||
output_hidden_states=output_hidden_states,
|
||||
return_dict=return_dict
|
||||
)
|
||||
logits = outputs.logits
|
||||
|
||||
loss = None
|
||||
if labels is not None:
|
||||
# Shift so that tokens < n predict n
|
||||
shift_logits = logits[..., :-1, :].contiguous()
|
||||
shift_labels = labels[..., 1:].contiguous()
|
||||
# Flatten the tokens
|
||||
loss_fct = CrossEntropyLoss()
|
||||
shift_logits = shift_logits.view(-1, self.language_model.config.vocab_size)
|
||||
shift_labels = shift_labels.view(-1)
|
||||
# Enable model parallelism
|
||||
shift_labels = shift_labels.to(shift_logits.device)
|
||||
loss = loss_fct(shift_logits, shift_labels)
|
||||
if ignore_flag:
|
||||
loss = loss * 0.0
|
||||
|
||||
if not return_dict:
|
||||
output = (logits,) + outputs[1:]
|
||||
return (loss,) + output if loss is not None else output
|
||||
|
||||
return CausalLMOutputWithPast(
|
||||
loss=loss,
|
||||
logits=logits,
|
||||
past_key_values=outputs.past_key_values,
|
||||
hidden_states=outputs.hidden_states,
|
||||
attentions=outputs.attentions,
|
||||
)
|
||||
|
||||
def pixel_shuffle(self, x, scale_factor=0.5):
|
||||
n, w, h, c = x.size()
|
||||
# N, W, H, C --> N, W, H * scale, C // scale
|
||||
x = x.view(n, w, int(h * scale_factor), int(c / scale_factor))
|
||||
# N, W, H * scale, C // scale --> N, H * scale, W, C // scale
|
||||
x = x.permute(0, 2, 1, 3).contiguous()
|
||||
# N, H * scale, W, C // scale --> N, H * scale, W * scale, C // (scale ** 2)
|
||||
x = x.view(n, int(h * scale_factor), int(w * scale_factor),
|
||||
int(c / (scale_factor * scale_factor)))
|
||||
if self.ps_version == 'v1':
|
||||
warnings.warn("In ps_version 'v1', the height and width have not been swapped back, "
|
||||
'which results in a transposed image.')
|
||||
else:
|
||||
x = x.permute(0, 2, 1, 3).contiguous()
|
||||
return x
|
||||
|
||||
def extract_feature(self, pixel_values):
|
||||
if self.select_layer == -1:
|
||||
vit_embeds = self.vision_model(
|
||||
pixel_values=pixel_values,
|
||||
output_hidden_states=False,
|
||||
return_dict=True).last_hidden_state
|
||||
else:
|
||||
vit_embeds = self.vision_model(
|
||||
pixel_values=pixel_values,
|
||||
output_hidden_states=True,
|
||||
return_dict=True).hidden_states[self.select_layer]
|
||||
|
||||
if int(vit_embeds.shape[1] ** 0.5)**2 != vit_embeds.shape[1]:
|
||||
vit_embeds = vit_embeds[:, 1:, :]
|
||||
|
||||
h = w = int(vit_embeds.shape[1] ** 0.5)
|
||||
vit_embeds = vit_embeds.reshape(vit_embeds.shape[0], h, w, -1)
|
||||
vit_embeds = self.pixel_shuffle(vit_embeds, scale_factor=self.downsample_ratio)
|
||||
vit_embeds = vit_embeds.reshape(vit_embeds.shape[0], -1, vit_embeds.shape[-1])
|
||||
vit_embeds = self.mlp1(vit_embeds)
|
||||
return vit_embeds
|
||||
|
||||
def batch_chat(self, tokenizer, pixel_values, questions, generation_config, num_patches_list=None,
|
||||
history=None, return_history=False, IMG_START_TOKEN='<img>', IMG_END_TOKEN='</img>',
|
||||
IMG_CONTEXT_TOKEN='<IMG_CONTEXT>', verbose=False, image_counts=None):
|
||||
if history is not None or return_history:
|
||||
print('Now multi-turn chat is not supported in batch_chat.')
|
||||
raise NotImplementedError
|
||||
|
||||
if image_counts is not None:
|
||||
num_patches_list = image_counts
|
||||
print('Warning: `image_counts` is deprecated. Please use `num_patches_list` instead.')
|
||||
|
||||
img_context_token_id = tokenizer.convert_tokens_to_ids(IMG_CONTEXT_TOKEN)
|
||||
self.img_context_token_id = img_context_token_id
|
||||
|
||||
if verbose and pixel_values is not None:
|
||||
image_bs = pixel_values.shape[0]
|
||||
print(f'dynamic ViT batch size: {image_bs}')
|
||||
|
||||
queries = []
|
||||
for idx, num_patches in enumerate(num_patches_list):
|
||||
question = questions[idx]
|
||||
if pixel_values is not None and '<image>' not in question:
|
||||
question = '<image>\n' + question
|
||||
template = get_conv_template(self.template)
|
||||
template.system_message = self.system_message
|
||||
template.append_message(template.roles[0], question)
|
||||
template.append_message(template.roles[1], None)
|
||||
query = template.get_prompt()
|
||||
|
||||
image_tokens = IMG_START_TOKEN + IMG_CONTEXT_TOKEN * self.num_image_token * num_patches + IMG_END_TOKEN
|
||||
query = query.replace('<image>', image_tokens, 1)
|
||||
queries.append(query)
|
||||
|
||||
tokenizer.padding_side = 'left'
|
||||
model_inputs = tokenizer(queries, return_tensors='pt', padding=True)
|
||||
input_ids = model_inputs['input_ids'].cuda()
|
||||
attention_mask = model_inputs['attention_mask'].cuda()
|
||||
eos_token_id = tokenizer.convert_tokens_to_ids(template.sep)
|
||||
generation_config['eos_token_id'] = eos_token_id
|
||||
generation_output = self.generate(
|
||||
pixel_values=pixel_values,
|
||||
input_ids=input_ids,
|
||||
attention_mask=attention_mask,
|
||||
**generation_config
|
||||
)
|
||||
responses = tokenizer.batch_decode(generation_output, skip_special_tokens=True)
|
||||
responses = [response.split(template.sep)[0].strip() for response in responses]
|
||||
return responses
|
||||
|
||||
def chat(self, tokenizer, pixel_values, question, generation_config, history=None, return_history=False,
|
||||
num_patches_list=None, IMG_START_TOKEN='<img>', IMG_END_TOKEN='</img>', IMG_CONTEXT_TOKEN='<IMG_CONTEXT>',
|
||||
verbose=False):
|
||||
|
||||
if history is None and pixel_values is not None and '<image>' not in question:
|
||||
question = '<image>\n' + question
|
||||
|
||||
if num_patches_list is None:
|
||||
num_patches_list = [pixel_values.shape[0]] if pixel_values is not None else []
|
||||
assert pixel_values is None or len(pixel_values) == sum(num_patches_list)
|
||||
|
||||
img_context_token_id = tokenizer.convert_tokens_to_ids(IMG_CONTEXT_TOKEN)
|
||||
self.img_context_token_id = img_context_token_id
|
||||
|
||||
template = get_conv_template(self.template)
|
||||
template.system_message = self.system_message
|
||||
eos_token_id = tokenizer.convert_tokens_to_ids(template.sep)
|
||||
|
||||
history = [] if history is None else history
|
||||
for (old_question, old_answer) in history:
|
||||
template.append_message(template.roles[0], old_question)
|
||||
template.append_message(template.roles[1], old_answer)
|
||||
template.append_message(template.roles[0], question)
|
||||
template.append_message(template.roles[1], None)
|
||||
query = template.get_prompt()
|
||||
|
||||
if verbose and pixel_values is not None:
|
||||
image_bs = pixel_values.shape[0]
|
||||
print(f'dynamic ViT batch size: {image_bs}')
|
||||
|
||||
for num_patches in num_patches_list:
|
||||
image_tokens = IMG_START_TOKEN + IMG_CONTEXT_TOKEN * self.num_image_token * num_patches + IMG_END_TOKEN
|
||||
query = query.replace('<image>', image_tokens, 1)
|
||||
|
||||
model_inputs = tokenizer(query, return_tensors='pt')
|
||||
input_ids = model_inputs['input_ids'].cuda()
|
||||
attention_mask = model_inputs['attention_mask'].cuda()
|
||||
generation_config['eos_token_id'] = eos_token_id
|
||||
generation_output = self.generate(
|
||||
pixel_values=pixel_values,
|
||||
input_ids=input_ids,
|
||||
attention_mask=attention_mask,
|
||||
**generation_config
|
||||
)
|
||||
response = tokenizer.batch_decode(generation_output, skip_special_tokens=True)[0]
|
||||
response = response.split(template.sep)[0].strip()
|
||||
history.append((question, response))
|
||||
if return_history:
|
||||
return response, history
|
||||
else:
|
||||
query_to_print = query.replace(IMG_CONTEXT_TOKEN, '')
|
||||
query_to_print = query_to_print.replace(f'{IMG_START_TOKEN}{IMG_END_TOKEN}', '<image>')
|
||||
if verbose:
|
||||
print(query_to_print, response)
|
||||
return response
|
||||
|
||||
@torch.no_grad()
|
||||
def generate(
|
||||
self,
|
||||
pixel_values: Optional[torch.FloatTensor] = None,
|
||||
input_ids: Optional[torch.FloatTensor] = None,
|
||||
attention_mask: Optional[torch.LongTensor] = None,
|
||||
visual_features: Optional[torch.FloatTensor] = None,
|
||||
generation_config: Optional[GenerationConfig] = None,
|
||||
output_hidden_states: Optional[bool] = None,
|
||||
return_dict: Optional[bool] = None,
|
||||
**generate_kwargs,
|
||||
) -> torch.LongTensor:
|
||||
|
||||
assert self.img_context_token_id is not None
|
||||
if pixel_values is not None:
|
||||
if visual_features is not None:
|
||||
vit_embeds = visual_features
|
||||
else:
|
||||
vit_embeds = self.extract_feature(pixel_values)
|
||||
input_embeds = self.language_model.get_input_embeddings()(input_ids)
|
||||
B, N, C = input_embeds.shape
|
||||
input_embeds = input_embeds.reshape(B * N, C)
|
||||
|
||||
input_ids = input_ids.reshape(B * N)
|
||||
selected = (input_ids == self.img_context_token_id)
|
||||
assert selected.sum() != 0
|
||||
input_embeds[selected] = vit_embeds.reshape(-1, C).to(input_embeds.device)
|
||||
|
||||
input_embeds = input_embeds.reshape(B, N, C)
|
||||
visual_token_mask = selected.reshape(B, N, 1).to(input_embeds.dtype)
|
||||
else:
|
||||
input_embeds = self.language_model.get_input_embeddings()(input_ids)
|
||||
B, N, C = input_embeds.shape
|
||||
visual_token_mask = torch.zeros_like(input_ids).reshape(B, N, 1).to(input_embeds.dtype)
|
||||
|
||||
if self.use_visual_token_mask:
|
||||
outputs = self.language_model.generate(
|
||||
inputs_embeds=input_embeds,
|
||||
attention_mask=attention_mask,
|
||||
generation_config=generation_config,
|
||||
output_hidden_states=output_hidden_states,
|
||||
return_dict=return_dict,
|
||||
use_cache=True,
|
||||
visual_token_mask=visual_token_mask,
|
||||
**generate_kwargs,
|
||||
)
|
||||
else:
|
||||
outputs = self.language_model.generate(
|
||||
inputs_embeds=input_embeds,
|
||||
attention_mask=attention_mask,
|
||||
generation_config=generation_config,
|
||||
output_hidden_states=output_hidden_states,
|
||||
return_dict=return_dict,
|
||||
use_cache=True,
|
||||
**generate_kwargs,
|
||||
)
|
||||
|
||||
return outputs
|
||||
47
special_tokens_map.json
Normal file
47
special_tokens_map.json
Normal file
@@ -0,0 +1,47 @@
|
||||
{
|
||||
"additional_special_tokens": [
|
||||
"<|im_start|>",
|
||||
"<|im_end|>",
|
||||
"<|action_start|>",
|
||||
"<|action_end|>",
|
||||
"<|interpreter|>",
|
||||
"<|plugin|>",
|
||||
"<img>",
|
||||
"</img>",
|
||||
"<IMG_CONTEXT>",
|
||||
"<quad>",
|
||||
"</quad>",
|
||||
"<ref>",
|
||||
"</ref>",
|
||||
"<box>",
|
||||
"</box>"
|
||||
],
|
||||
"bos_token": {
|
||||
"content": "<s>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false
|
||||
},
|
||||
"eos_token": {
|
||||
"content": "</s>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false
|
||||
},
|
||||
"pad_token": {
|
||||
"content": "</s>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false
|
||||
},
|
||||
"unk_token": {
|
||||
"content": "<unk>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false
|
||||
}
|
||||
}
|
||||
235
tokenization_internlm2.py
Normal file
235
tokenization_internlm2.py
Normal file
@@ -0,0 +1,235 @@
|
||||
# Copyright (c) The InternLM team and The HuggingFace Inc. team. All rights reserved.
|
||||
#
|
||||
# This code is based on transformers/src/transformers/models/llama/tokenization_llama.py
|
||||
#
|
||||
# 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.
|
||||
|
||||
"""Tokenization classes for InternLM."""
|
||||
import os
|
||||
from shutil import copyfile
|
||||
from typing import Any, Dict, List, Optional, Tuple
|
||||
|
||||
import sentencepiece as spm
|
||||
from transformers.tokenization_utils import PreTrainedTokenizer
|
||||
from transformers.utils import logging
|
||||
|
||||
logger = logging.get_logger(__name__)
|
||||
|
||||
VOCAB_FILES_NAMES = {'vocab_file': './tokenizer.model'}
|
||||
|
||||
PRETRAINED_VOCAB_FILES_MAP = {}
|
||||
|
||||
|
||||
# Modified from transformers.model.llama.tokenization_llama.LlamaTokenizer
|
||||
class InternLM2Tokenizer(PreTrainedTokenizer):
|
||||
"""
|
||||
Construct a InternLM2 tokenizer. Based on byte-level Byte-Pair-Encoding.
|
||||
|
||||
Args:
|
||||
vocab_file (`str`):
|
||||
Path to the vocabulary file.
|
||||
"""
|
||||
|
||||
vocab_files_names = VOCAB_FILES_NAMES
|
||||
pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
|
||||
model_input_names = ['input_ids', 'attention_mask']
|
||||
_auto_class = 'AutoTokenizer'
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
vocab_file,
|
||||
unk_token='<unk>',
|
||||
bos_token='<s>',
|
||||
eos_token='</s>',
|
||||
pad_token='</s>',
|
||||
sp_model_kwargs: Optional[Dict[str, Any]] = None,
|
||||
add_bos_token=True,
|
||||
add_eos_token=False,
|
||||
decode_with_prefix_space=False,
|
||||
clean_up_tokenization_spaces=False,
|
||||
**kwargs,
|
||||
):
|
||||
self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
|
||||
self.vocab_file = vocab_file
|
||||
self.add_bos_token = add_bos_token
|
||||
self.add_eos_token = add_eos_token
|
||||
self.decode_with_prefix_space = decode_with_prefix_space
|
||||
self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
|
||||
self.sp_model.Load(vocab_file)
|
||||
self._no_prefix_space_tokens = None
|
||||
super().__init__(
|
||||
bos_token=bos_token,
|
||||
eos_token=eos_token,
|
||||
unk_token=unk_token,
|
||||
pad_token=pad_token,
|
||||
clean_up_tokenization_spaces=clean_up_tokenization_spaces,
|
||||
**kwargs,
|
||||
)
|
||||
|
||||
@property
|
||||
def no_prefix_space_tokens(self):
|
||||
if self._no_prefix_space_tokens is None:
|
||||
vocab = self.convert_ids_to_tokens(list(range(self.vocab_size)))
|
||||
self._no_prefix_space_tokens = {i for i, tok in enumerate(vocab) if not tok.startswith('▁')}
|
||||
return self._no_prefix_space_tokens
|
||||
|
||||
@property
|
||||
def vocab_size(self):
|
||||
"""Returns vocab size"""
|
||||
return self.sp_model.get_piece_size()
|
||||
|
||||
@property
|
||||
def bos_token_id(self) -> Optional[int]:
|
||||
return self.sp_model.bos_id()
|
||||
|
||||
@property
|
||||
def eos_token_id(self) -> Optional[int]:
|
||||
return self.sp_model.eos_id()
|
||||
|
||||
def get_vocab(self):
|
||||
"""Returns vocab as a dict"""
|
||||
vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
|
||||
vocab.update(self.added_tokens_encoder)
|
||||
return vocab
|
||||
|
||||
def _tokenize(self, text):
|
||||
"""Returns a tokenized string."""
|
||||
return self.sp_model.encode(text, out_type=str)
|
||||
|
||||
def _convert_token_to_id(self, token):
|
||||
"""Converts a token (str) in an id using the vocab."""
|
||||
return self.sp_model.piece_to_id(token)
|
||||
|
||||
def _convert_id_to_token(self, index):
|
||||
"""Converts an index (integer) in a token (str) using the vocab."""
|
||||
token = self.sp_model.IdToPiece(index)
|
||||
return token
|
||||
|
||||
def _maybe_add_prefix_space(self, tokens, decoded):
|
||||
if tokens and tokens[0] not in self.no_prefix_space_tokens:
|
||||
return ' ' + decoded
|
||||
else:
|
||||
return decoded
|
||||
|
||||
def convert_tokens_to_string(self, tokens):
|
||||
"""Converts a sequence of tokens (string) in a single string."""
|
||||
current_sub_tokens = []
|
||||
out_string = ''
|
||||
prev_is_special = False
|
||||
for token in tokens:
|
||||
# make sure that special tokens are not decoded using sentencepiece model
|
||||
if token in self.all_special_tokens:
|
||||
if not prev_is_special:
|
||||
out_string += ' '
|
||||
out_string += self.sp_model.decode(current_sub_tokens) + token
|
||||
prev_is_special = True
|
||||
current_sub_tokens = []
|
||||
else:
|
||||
current_sub_tokens.append(token)
|
||||
prev_is_special = False
|
||||
out_string += self.sp_model.decode(current_sub_tokens)
|
||||
out_string = self.clean_up_tokenization(out_string)
|
||||
out_string = self._maybe_add_prefix_space(tokens=tokens, decoded=out_string)
|
||||
return out_string[1:]
|
||||
|
||||
def save_vocabulary(self, save_directory, filename_prefix: Optional[str] = None) -> Tuple[str]:
|
||||
"""
|
||||
Save the vocabulary and special tokens file to a directory.
|
||||
|
||||
Args:
|
||||
save_directory (`str`):
|
||||
The directory in which to save the vocabulary.
|
||||
|
||||
Returns:
|
||||
`Tuple(str)`: Paths to the files saved.
|
||||
"""
|
||||
if not os.path.isdir(save_directory):
|
||||
logger.error(f'Vocabulary path ({save_directory}) should be a directory')
|
||||
return
|
||||
out_vocab_file = os.path.join(
|
||||
save_directory, (filename_prefix + '-' if filename_prefix else '') + VOCAB_FILES_NAMES['vocab_file']
|
||||
)
|
||||
|
||||
if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file) and os.path.isfile(self.vocab_file):
|
||||
copyfile(self.vocab_file, out_vocab_file)
|
||||
elif not os.path.isfile(self.vocab_file):
|
||||
with open(out_vocab_file, 'wb') as fi:
|
||||
content_spiece_model = self.sp_model.serialized_model_proto()
|
||||
fi.write(content_spiece_model)
|
||||
|
||||
return (out_vocab_file,)
|
||||
|
||||
def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
|
||||
if self.add_bos_token:
|
||||
bos_token_ids = [self.bos_token_id]
|
||||
else:
|
||||
bos_token_ids = []
|
||||
|
||||
output = bos_token_ids + token_ids_0
|
||||
|
||||
if token_ids_1 is not None:
|
||||
output = output + token_ids_1
|
||||
|
||||
if self.add_eos_token:
|
||||
output = output + [self.eos_token_id]
|
||||
|
||||
return output
|
||||
|
||||
def get_special_tokens_mask(
|
||||
self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
|
||||
) -> List[int]:
|
||||
"""
|
||||
Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
|
||||
special tokens using the tokenizer `prepare_for_model` method.
|
||||
|
||||
Args:
|
||||
token_ids_0 (`List[int]`):
|
||||
List of IDs.
|
||||
token_ids_1 (`List[int]`, *optional*):
|
||||
Optional second list of IDs for sequence pairs.
|
||||
already_has_special_tokens (`bool`, *optional*, defaults to `False`):
|
||||
Whether or not the token list is already formatted with special tokens for the model.
|
||||
|
||||
Returns:
|
||||
`List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
|
||||
"""
|
||||
if already_has_special_tokens:
|
||||
return super().get_special_tokens_mask(
|
||||
token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
|
||||
)
|
||||
|
||||
if token_ids_1 is None:
|
||||
return [1] + ([0] * len(token_ids_0)) + [1]
|
||||
return [1] + ([0] * len(token_ids_0)) + [1, 1] + ([0] * len(token_ids_1)) + [1]
|
||||
|
||||
def create_token_type_ids_from_sequences(
|
||||
self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
|
||||
) -> List[int]:
|
||||
"""
|
||||
Create a mask from the two sequences passed to be used in a sequence-pair classification task. T5 does not make
|
||||
use of token type ids, therefore a list of zeros is returned.
|
||||
|
||||
Args:
|
||||
token_ids_0 (`List[int]`):
|
||||
List of IDs.
|
||||
token_ids_1 (`List[int]`, *optional*):
|
||||
Optional second list of IDs for sequence pairs.
|
||||
|
||||
Returns:
|
||||
`List[int]`: List of zeros.
|
||||
"""
|
||||
eos = [self.eos_token_id]
|
||||
|
||||
if token_ids_1 is None:
|
||||
return len(token_ids_0 + eos) * [0]
|
||||
return len(token_ids_0 + eos + token_ids_1 + eos) * [0]
|
||||
211
tokenization_internlm2_fast.py
Normal file
211
tokenization_internlm2_fast.py
Normal file
@@ -0,0 +1,211 @@
|
||||
# Copyright (c) The InternLM team and The HuggingFace Inc. team. All rights reserved.
|
||||
#
|
||||
# This code is based on transformers/src/transformers/models/llama/tokenization_llama_fast.py
|
||||
#
|
||||
# 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.
|
||||
|
||||
"""Tokenization Fast class for InternLM."""
|
||||
import os
|
||||
from shutil import copyfile
|
||||
from typing import Any, Dict, Optional, Tuple
|
||||
|
||||
from tokenizers import Tokenizer, decoders, normalizers, processors
|
||||
from tokenizers.models import BPE
|
||||
from transformers.convert_slow_tokenizer import (SLOW_TO_FAST_CONVERTERS,
|
||||
SentencePieceExtractor,
|
||||
SpmConverter)
|
||||
from transformers.tokenization_utils_fast import PreTrainedTokenizerFast
|
||||
from transformers.utils import logging
|
||||
|
||||
from .tokenization_internlm2 import InternLM2Tokenizer
|
||||
|
||||
logger = logging.get_logger(__name__)
|
||||
|
||||
VOCAB_FILES_NAMES = {'vocab_file': './tokenizer.model'}
|
||||
|
||||
|
||||
# Modified from transformers.convert_slow_tokenizer.LlamaConverter
|
||||
class InternLM2Converter(SpmConverter):
|
||||
handle_byte_fallback = True
|
||||
|
||||
def vocab(self, proto):
|
||||
vocab = [
|
||||
('<unk>', 0.0),
|
||||
('<s>', 0.0),
|
||||
('</s>', 0.0),
|
||||
]
|
||||
vocab += [(piece.piece, piece.score) for piece in proto.pieces[3:]]
|
||||
return vocab
|
||||
|
||||
def unk_id(self, proto):
|
||||
unk_id = 0
|
||||
return unk_id
|
||||
|
||||
def decoder(self, replacement, add_prefix_space):
|
||||
return decoders.Sequence(
|
||||
[
|
||||
decoders.Replace('▁', ' '),
|
||||
decoders.ByteFallback(),
|
||||
decoders.Fuse(),
|
||||
decoders.Strip(content=' ', left=1),
|
||||
]
|
||||
)
|
||||
|
||||
def tokenizer(self, proto):
|
||||
model_type = proto.trainer_spec.model_type
|
||||
vocab_scores = self.vocab(proto)
|
||||
# special tokens
|
||||
added_tokens = self.original_tokenizer.added_tokens_decoder
|
||||
for i in range(len(vocab_scores)):
|
||||
piece, score = vocab_scores[i]
|
||||
if i in added_tokens:
|
||||
vocab_scores[i] = (added_tokens[i].content, score)
|
||||
if model_type == 1:
|
||||
raise RuntimeError('InternLM2 is supposed to be a BPE model!')
|
||||
|
||||
elif model_type == 2:
|
||||
_, merges = SentencePieceExtractor(self.original_tokenizer.vocab_file).extract(vocab_scores)
|
||||
bpe_vocab = {word: i for i, (word, _score) in enumerate(vocab_scores)}
|
||||
tokenizer = Tokenizer(
|
||||
BPE(bpe_vocab, merges, unk_token=proto.trainer_spec.unk_piece, fuse_unk=True, byte_fallback=True)
|
||||
)
|
||||
tokenizer.add_special_tokens(
|
||||
[ added_token for index, added_token in added_tokens.items()]
|
||||
)
|
||||
else:
|
||||
raise Exception(
|
||||
"You're trying to run a `Unigram` model but you're file was trained with a different algorithm"
|
||||
)
|
||||
|
||||
return tokenizer
|
||||
|
||||
def normalizer(self, proto):
|
||||
normalizers_list = []
|
||||
if proto.normalizer_spec.add_dummy_prefix:
|
||||
normalizers_list.append(normalizers.Prepend(prepend='▁'))
|
||||
normalizers_list.append(normalizers.Replace(pattern=' ', content='▁'))
|
||||
return normalizers.Sequence(normalizers_list)
|
||||
|
||||
def pre_tokenizer(self, replacement, add_prefix_space):
|
||||
return None
|
||||
|
||||
|
||||
SLOW_TO_FAST_CONVERTERS['InternLM2Tokenizer'] = InternLM2Converter
|
||||
|
||||
|
||||
# Modified from transformers.model.llama.tokenization_llama_fast.LlamaTokenizerFast -> InternLM2TokenizerFast
|
||||
class InternLM2TokenizerFast(PreTrainedTokenizerFast):
|
||||
vocab_files_names = VOCAB_FILES_NAMES
|
||||
slow_tokenizer_class = InternLM2Tokenizer
|
||||
padding_side = 'left'
|
||||
model_input_names = ['input_ids', 'attention_mask']
|
||||
_auto_class = 'AutoTokenizer'
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
vocab_file,
|
||||
unk_token='<unk>',
|
||||
bos_token='<s>',
|
||||
eos_token='</s>',
|
||||
pad_token='</s>',
|
||||
sp_model_kwargs: Optional[Dict[str, Any]] = None,
|
||||
add_bos_token=True,
|
||||
add_eos_token=False,
|
||||
decode_with_prefix_space=False,
|
||||
clean_up_tokenization_spaces=False,
|
||||
**kwargs,
|
||||
):
|
||||
super().__init__(
|
||||
vocab_file=vocab_file,
|
||||
unk_token=unk_token,
|
||||
bos_token=bos_token,
|
||||
eos_token=eos_token,
|
||||
pad_token=pad_token,
|
||||
sp_model_kwargs=sp_model_kwargs,
|
||||
add_bos_token=add_bos_token,
|
||||
add_eos_token=add_eos_token,
|
||||
decode_with_prefix_space=decode_with_prefix_space,
|
||||
clean_up_tokenization_spaces=clean_up_tokenization_spaces,
|
||||
**kwargs,
|
||||
)
|
||||
self._add_bos_token = add_bos_token
|
||||
self._add_eos_token = add_eos_token
|
||||
self.update_post_processor()
|
||||
self.vocab_file = vocab_file
|
||||
|
||||
@property
|
||||
def can_save_slow_tokenizer(self) -> bool:
|
||||
return os.path.isfile(self.vocab_file) if self.vocab_file else False
|
||||
|
||||
def update_post_processor(self):
|
||||
"""
|
||||
Updates the underlying post processor with the current `bos_token` and `eos_token`.
|
||||
"""
|
||||
bos = self.bos_token
|
||||
bos_token_id = self.bos_token_id
|
||||
if bos is None and self.add_bos_token:
|
||||
raise ValueError('add_bos_token = True but bos_token = None')
|
||||
|
||||
eos = self.eos_token
|
||||
eos_token_id = self.eos_token_id
|
||||
if eos is None and self.add_eos_token:
|
||||
raise ValueError('add_eos_token = True but eos_token = None')
|
||||
|
||||
single = f"{(bos+':0 ') if self.add_bos_token else ''}$A:0{(' '+eos+':0') if self.add_eos_token else ''}"
|
||||
pair = f"{single}{(' '+bos+':1') if self.add_bos_token else ''} $B:1{(' '+eos+':1') if self.add_eos_token else ''}"
|
||||
|
||||
special_tokens = []
|
||||
if self.add_bos_token:
|
||||
special_tokens.append((bos, bos_token_id))
|
||||
if self.add_eos_token:
|
||||
special_tokens.append((eos, eos_token_id))
|
||||
self._tokenizer.post_processor = processors.TemplateProcessing(
|
||||
single=single, pair=pair, special_tokens=special_tokens
|
||||
)
|
||||
|
||||
@property
|
||||
def add_eos_token(self):
|
||||
return self._add_eos_token
|
||||
|
||||
@property
|
||||
def add_bos_token(self):
|
||||
return self._add_bos_token
|
||||
|
||||
@add_eos_token.setter
|
||||
def add_eos_token(self, value):
|
||||
self._add_eos_token = value
|
||||
self.update_post_processor()
|
||||
|
||||
@add_bos_token.setter
|
||||
def add_bos_token(self, value):
|
||||
self._add_bos_token = value
|
||||
self.update_post_processor()
|
||||
|
||||
def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None) -> Tuple[str]:
|
||||
if not self.can_save_slow_tokenizer:
|
||||
raise ValueError(
|
||||
'Your fast tokenizer does not have the necessary information to save the vocabulary for a slow '
|
||||
'tokenizer.'
|
||||
)
|
||||
|
||||
if not os.path.isdir(save_directory):
|
||||
logger.error(f'Vocabulary path ({save_directory}) should be a directory')
|
||||
return
|
||||
out_vocab_file = os.path.join(
|
||||
save_directory, (filename_prefix + '-' if filename_prefix else '') + VOCAB_FILES_NAMES['vocab_file']
|
||||
)
|
||||
|
||||
if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file):
|
||||
copyfile(self.vocab_file, out_vocab_file)
|
||||
|
||||
return (out_vocab_file,)
|
||||
3
tokenizer.model
Normal file
3
tokenizer.model
Normal file
@@ -0,0 +1,3 @@
|
||||
version https://git-lfs.github.com/spec/v1
|
||||
oid sha256:f868398fc4e05ee1e8aeba95ddf18ddcc45b8bce55d5093bead5bbf80429b48b
|
||||
size 1477754
|
||||
179
tokenizer_config.json
Normal file
179
tokenizer_config.json
Normal file
@@ -0,0 +1,179 @@
|
||||
{
|
||||
"added_tokens_decoder": {
|
||||
"0": {
|
||||
"content": "<unk>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"1": {
|
||||
"content": "<s>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"2": {
|
||||
"content": "</s>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"92538": {
|
||||
"content": "<|plugin|>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"92539": {
|
||||
"content": "<|interpreter|>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"92540": {
|
||||
"content": "<|action_end|>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"92541": {
|
||||
"content": "<|action_start|>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"92542": {
|
||||
"content": "<|im_end|>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"92543": {
|
||||
"content": "<|im_start|>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"92544": {
|
||||
"content": "<img>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"92545": {
|
||||
"content": "</img>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"92546": {
|
||||
"content": "<IMG_CONTEXT>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"92547": {
|
||||
"content": "<quad>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"92548": {
|
||||
"content": "</quad>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"92549": {
|
||||
"content": "<ref>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"92550": {
|
||||
"content": "</ref>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"92551": {
|
||||
"content": "<box>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
},
|
||||
"92552": {
|
||||
"content": "</box>",
|
||||
"lstrip": false,
|
||||
"normalized": false,
|
||||
"rstrip": false,
|
||||
"single_word": false,
|
||||
"special": true
|
||||
}
|
||||
},
|
||||
"additional_special_tokens": [
|
||||
"<|im_start|>",
|
||||
"<|im_end|>",
|
||||
"<|action_start|>",
|
||||
"<|action_end|>",
|
||||
"<|interpreter|>",
|
||||
"<|plugin|>",
|
||||
"<img>",
|
||||
"</img>",
|
||||
"<IMG_CONTEXT>",
|
||||
"<quad>",
|
||||
"</quad>",
|
||||
"<ref>",
|
||||
"</ref>",
|
||||
"<box>",
|
||||
"</box>"
|
||||
],
|
||||
"auto_map": {
|
||||
"AutoTokenizer": [
|
||||
"tokenization_internlm2.InternLM2Tokenizer",
|
||||
null
|
||||
]
|
||||
},
|
||||
"bos_token": "<s>",
|
||||
"chat_template": "{{ bos_token }}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
||||
"clean_up_tokenization_spaces": false,
|
||||
"eos_token": "</s>",
|
||||
"model_max_length": 8192,
|
||||
"pad_token": "</s>",
|
||||
"tokenizer_class": "InternLM2Tokenizer",
|
||||
"unk_token": "<unk>"
|
||||
}
|
||||
Reference in New Issue
Block a user