jianzhnie / open-chatgpt Goto Github PK

how to train prepare data?

When I run "python train_reward_model.py", I get the following error. How can I solve it?
Python version 3.10.8

Traceback (most recent call last):
File "C:\Users\charley\AppData\Local\Programs\Python\Python310\lib\site-packages\transformers\utils\import_utils.py", line 1110, in get_module
return importlib.import_module("." + module_name, self.name)
File "C:\Users\charley\AppData\Local\Programs\Python\Python310\lib\importlib_init.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "", line 1050, in _gcd_import
File "", line 1027, in _find_and_load
File "", line 1006, in _find_and_load_unlocked
File "", line 688, in load_unlocked
File "", line 883, in exec_module
File "", line 241, in call_with_frames_removed
File "C:\Users\charley\AppData\Local\Programs\Python\Python310\lib\site-packages\transformers\trainer.py", line 67, in
from .data.data_collator import DataCollator, DataCollatorWithPadding, default_data_collator
File "C:\Users\charley\AppData\Local\Programs\Python\Python310\lib\site-packages\transformers\data_init.py", line 31, in
from .processors import (
File "C:\Users\charley\AppData\Local\Programs\Python\Python310\lib\site-packages\transformers\data\processors_init.py", line 19, in
from .glue import glue_convert_examples_to_features, glue_output_modes, glue_processors, glue_tasks_num_labels
File "C:\Users\charley\AppData\Local\Programs\Python\Python310\lib\site-packages\transformers\data\processors\glue.py", line 79, in
examples: tf.data.Dataset,
AttributeError: module 'tensorflow' has no attribute 'data'

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "C:\Windows\System32\open-chatgpt\examples\summarization\train_reward_model.py", line 4, in
from transformers import (AutoTokenizer, EarlyStoppingCallback, EvalPrediction,
File "", line 1075, in _handle_fromlist
File "C:\Users\charley\AppData\Local\Programs\Python\Python310\lib\site-packages\transformers\utils\import_utils.py", line 1100, in getattr
module = self._get_module(self._class_to_module[name])
File "C:\Users\charley\AppData\Local\Programs\Python\Python310\lib\site-packages\transformers\utils\import_utils.py", line 1112, in _get_module
raise RuntimeError(
RuntimeError: Failed to import transformers.trainer because of the following error (look up to see its traceback):
module 'tensorflow' has no attribute 'data'

谢谢

how to train with ppo? i do not find the code, can you show me?

使用chatgpt 4有一段时间了, 它的代码生成功能实在太强.
我很好奇这种功能是怎么训练的.
感觉用人工标注代码, 应该是行不通的. 因为很难用一种统一的标准来判定一份代码是好还是坏.

这么强悍的技术, 由一两家公司垄断是一件很恐怖的事情.
危险的不是技术本身, 而是使用技术的人.
有没有可能开源社区, 大家合作, 一起训练一个能挑战全面chatgpt的模型?