ModelCenter

ModelCenter copied to clipboard

A report of misspelling in the file cpm1.py

1

Hi developers, There is a misspelling in the line 138 of file cpm1.py in the following link: [cpm1.py](https://github.com/OpenBMB/ModelCenter/blob/bad1193d1871770b29044ab691b0d99c1cea07cf/model_center/model/cpm1.py#L138) 'Ture' should be 'True' Best

Kunlun-Zhu

[FEATURE] Make the dimensions of linear spaces distinguishable

1

When doing structured pruning, sometimes we need to apply the same mask before or after different modules if they have the same input or output space. Say, if we are...

alphaGem

```python from model_center.layer import CPM1 CPM1.from_pretrained("cpm1-large") ``` currently could not work since the function `check_web_and_convert_path` calls `bmt.rank()` or `bmt.print_rank()` to prevent every process downloads the checkpoint in a multi-gpu scenario....

Jiaxin-Wen

for CPM1 CPM2 Bert GPT2 GPTJ T5 and corresponding return datatype

QiaoZiqing

**Describe the bug** When I run the start up code in README.md, in step 4 "Train the model" I can't properly run the code. Google colab reported "TypeError: linear(): argument...

jiangzizi

I followed the Quick Start and in step 3, when I copied the code to Google Colab and try to run it, I encountered "KeyError: 'label'". I found that there...

jiangzizi

https://github.com/OpenBMB/ModelCenter/blob/main/examples/cpm2/pretrain_cpm2.py#L24 请问这里模型初始化是不是每卡都会执行？ 如果模型很大，可能内存OOM。谢谢您的解答。

ftgreat

[BUG] llama outputting random gibberish

1

**Describe the bug** I used a verified LLaMA 7B hg checkpoint, and used a single thread bmb to do inference. But the output are just random gibberish. Not sure why?...

w32zhong

How can I use my own dataset while using ModelCenter？

1

lhj-git

**Describe the bug** Building prefix dict from the default dictionary ... Loading model from cache /tmp/jieba.cache Building prefix dict from the default dictionary ... Loading model from cache /tmp/jieba.cache Loading...

pikaqqqqqq