deepgeek/MiniCPM-o
2
0
Fork 0
mirror of https://github.com/OpenBMB/MiniCPM-V.git synced 2026-02-04 09:49:20 +08:00
Code Issues Packages Projects Releases Wiki Activity

update finetuning code

Browse Source
This commit is contained in:
qianyu chen
2024-05-08 09:51:34 +08:00
committed by GitHub
parent 9f345c4020
commit f6cbd4fb25
9 changed files with 0 additions and 37 deletions
Download Patch File Download Diff File Expand all files Collapse all files

0
finetune/__init__.py Normal file
View File

10
omnilmm/finetune/dataset.py → finetune/dataset.py
View File

@@ -1,13 +1,3 @@
#!/usr/bin/env python
# -*- coding: utf-8 -*-
#
# Copyright @2024 AI, ZHIHU Inc. (zhihu.com)
#
# @author: wangchongyi <wangchongyi@zhihu.com>
# @author: chenqianyu <cqy1195@zhihu.com>
# @date: 2024/5/06
#
import os import os
import math import math
import json import json

0
omnilmm/finetune/ds_config_zero2.json → finetune/ds_config_zero2.json
View File

0
omnilmm/finetune/ds_config_zero3.json → finetune/ds_config_zero3.json
View File

8
omnilmm/finetune/finetune.py → finetune/finetune.py
View File

@@ -1,11 +1,3 @@
#!/usr/bin/env python
# -*- coding: utf-8 -*-
#
# Copyright @2024 AI, ZHIHU Inc. (zhihu.com)
#
# @author: chenqianyu <cqy1195@zhihu.com@zhihu.com>
# @date: 2024/5/03
#
import os import os
import glob import glob
import json import json

0
omnilmm/finetune/finetune_ds.sh → finetune/finetune_ds.sh
View File

3
omnilmm/finetune/readme.md → finetune/readme.md
View File

@@ -64,6 +64,3 @@ sh finetune_ds.sh
``` ```
#### Customizing Hyperparameters #### Customizing Hyperparameters
To tailor the training process according to your specific requirements, you can adjust various hyperparameters. For comprehensive documentation on available hyperparameters and their functionalities, you can refer to the [official Transformers documentation](https://huggingface.co/docs/transformers/main_classes/trainer#transformers.TrainingArguments). Experimentation and fine-tuning of these parameters are essential for achieving optimal model performance tailored to your specific task and dataset. To tailor the training process according to your specific requirements, you can adjust various hyperparameters. For comprehensive documentation on available hyperparameters and their functionalities, you can refer to the [official Transformers documentation](https://huggingface.co/docs/transformers/main_classes/trainer#transformers.TrainingArguments). Experimentation and fine-tuning of these parameters are essential for achieving optimal model performance tailored to your specific task and dataset.
### LoRA finetuning
**This part is still unfinished, and we will complete it as soon as possible.**

8
omnilmm/finetune/trainer.py → finetune/trainer.py
View File

@@ -1,11 +1,3 @@
#!/usr/bin/env python
# -*- coding: utf-8 -*-
#
# Copyright @2024 AI, ZHIHU Inc. (zhihu.com)
#
# @author: chenqianyu <cqy1195@zhihu.com@zhihu.com>
# @date: 2024/5/03
#
import torch import torch
import torch.nn as nn import torch.nn as nn
from typing import Tuple, Union, Optional, List, Dict, Any from typing import Tuple, Union, Optional, List, Dict, Any

8
omnilmm/finetune/__init__.py
View File

@@ -1,8 +0,0 @@
#!/usr/bin/env python
# -*- coding: utf-8 -*-
#
# Copyright @2024 AI, ZHIHU Inc. (zhihu.com)
#
# @author: chenqianyu <cqy1195@zhihu.com@zhihu.com>
# @date: 2024/5/02
#