Liang-Xingzheng

Results 10 comments of Liang-Xingzheng

确实是内存问题,多谢啦!

Python random seed: 840868838938890892 Training Algorithm: Number training instances: 2533999 Number dev instances: 262929 Epoch 1 out of 60 Traceback (most recent call last): File "model.py", line 528, in loss_expr...

`AttributeError: module 'dynet' has no attribute 'sum_cols' starevelyn@starevelyn-OptiPlex-7020:~/multi-criteria-cws$ pip3 list apturl (0.5.2) beautifulsoup4 (4.4.1) blinker (1.3) Brlapi (0.6.4) chardet (2.3.0) checkbox-support (0.22) cmake (0.9.0) command-not-found (0.3) cryptography (1.2.3) Cython (0.27.3)...

我又重新按前面提问那个命令安装了一下,这次应该按对了,但是还是报错了。。 liangzhi@liangzhiNLP:~$ python3 Python 3.6.3 |Anaconda, Inc.| (default, Oct 13 2017, 12:02:49) [GCC 7.2.0] on linux Type "help", "copyright", "credits" or "license" for more information. >>> import dynet [dynet] random...

config = AutoConfig.from_pretrained(model_path, trust_remote_code=True, pre_seq_len=576) 这里面的pre_seq_len要和训练模型时的参数一致,试一下

solutions: change code in evaluate.py: def _evaluate_one_rating(idx, k): ....... items_input = [] users_input = [] for item in items: items_input.append(item) # original: _data_matrix[:, item] users_input.append(user)

![image](https://github.com/user-attachments/assets/a19ad00b-b5c5-4e07-ba19-d94900864b23) 代码改成这样就跑通了。。

![image](https://github.com/user-attachments/assets/30a63e58-ee4b-46b4-b215-17cfb7fe2c81) ![image](https://github.com/user-attachments/assets/215beb67-374c-43ea-b840-0a64241812ea) 跑到20%多又报错。。

> 你好,一般情况下我们默认推理Internvl-76B需要四张卡以**防止out of memory**。具体的切分方法在**split_model**函数中。 我有8张卡在跑,显存应该是够的,但是就是会报错同时用了cpu和cuda,感觉是代码的问题。。按理来说应该只用cuda的,不知道cpu为啥用到