github
/
ChatGLM-6B
mirror of https://github.com/THUDM/ChatGLM-6B
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

[Document] 更新Mac部署 

[Document] 更新Mac部署
- FILE: README.md; README_en.md
- ADD: OPENMP; MPS

# 具体内容

以[chatglm-6b-int4](https://huggingface.co/THUDM/chatglm-6b-int4)量化模型为例，做如下配置：

- 安装libomp的步骤;
- 对量化后模型等配置gcc编译项；
- 量化后模型启用MPS的解释。
pull/899/head
Yifan 2023-05-03 14:35:11 +08:00
parent b13f1a63f3
commit 2629a55222
2 changed files with 17 additions and 26 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
README.md
View File

@ -207,22 +207,17 @@ clang: error: unsupported option '-fopenmp'
```bash
# 第一步: 参考`https://mac.r-project.org/openmp/`
## 假设gcc -v是14.x版本，其他版本见R-Project提供的表格
## 假设: gcc(clang)是14.x版本，其他版本见R-Project提供的表格
curl -O https://mac.r-project.org/openmp/openmp-14.0.6-darwin20-Release.tar.gz
sudo tar fvxz openmp-14.0.6-darwin20-Release.tar.gz -C /
## 此时会安装下面几个文件：
# usr/local/lib/libomp.dylib
# usr/local/include/ompt.h
# usr/local/include/omp.h
# usr/local/include/omp-tools.h
```
针对`chatglm-6b-int4`, 修改[quantization.py](https://huggingface.co/THUDM/chatglm-6b-int4/blob/main/quantization.py)，主要是把硬编码的`gcc -O3 -fPIC -pthread -fopenmp -std=c99`命令修改成`gcc -O3 -fPIC -Xclang -fopenmp -pthread -lomp -std=c99`，[对应代码](https://huggingface.co/THUDM/chatglm-6b-int4/blob/63d66b0572d11cedd5574b38da720299599539b3/quantization.py#L168)见下:
此时会安装下面几个文件：`/usr/local/lib/libomp.dylib`, `/usr/local/include/ompt.h`, `/usr/local/include/omp.h`, `/usr/local/include/omp-tools.h`。
然后针对`chatglm-6b-int4`, 修改[quantization.py](https://huggingface.co/THUDM/chatglm-6b-int4/blob/main/quantization.py)，主要是把硬编码的`gcc -O3 -fPIC -pthread -fopenmp -std=c99`命令修改成`gcc -O3 -fPIC -Xclang -fopenmp -pthread -lomp -std=c99`，[对应代码](https://huggingface.co/THUDM/chatglm-6b-int4/blob/63d66b0572d11cedd5574b38da720299599539b3/quantization.py#L168)见下:
```python
# 第二步
## 找到包含`gcc -O3 -fPIC -pthread -fopenmp -std=c99`的这一行
## 修改成
# 第二步: 找到包含`gcc -O3 -fPIC -pthread -fopenmp -std=c99`的这一行，并修改成
compile_command = "gcc -O3 -fPIC -Xclang -fopenmp -pthread -lomp -std=c99 {} -shared -o {}".format(source_code, kernel_file)
```
@ -233,7 +228,7 @@ import platform
## ...
## 上述相应部分修改为（请自行改一下缩进）：
if platform.uname()[0] == 'Darwin':
compile_command = "gcc -O3 -fPIC -Xclang -fopenmp -pthread -lomp -std=c99-o {}".format(
compile_command = "gcc -O3 -fPIC -Xclang -fopenmp -pthread -lomp -std=c99 -o {}".format(
source_code, kernel_file)
else:
compile_command = "gcc -O3 -fPIC -pthread -fopenmp -std=c99 {} -shared -o {}".format(

26
README_en.md
View File

@ -200,26 +200,22 @@ clang: error: unsupported option '-fopenmp'
Take the quantified int4 version [chatglm-6b-int4](https://huggingface.co/THUDM/chatglm-6b-int4) for example, the following extra steps are needed:
1. Install `libomp`;
2. Configure `gcc`.
#### Install `libomp`
```bash
# STEP 1: install libopenmp, check `https://mac.r-project.org/openmp/` for details
## Assumption: `gcc -v >= 14.x`, read the R-Poject before run the code:
## Assumption: `gcc(clang) >= 14.x`, read the R-Poject before run the code:
curl -O https://mac.r-project.org/openmp/openmp-14.0.6-darwin20-Release.tar.gz
sudo tar fvxz openmp-14.0.6-darwin20-Release.tar.gz -C /
## Four files are installed:
# usr/local/lib/libomp.dylib
# usr/local/include/ompt.h
# usr/local/include/omp.h
# usr/local/include/omp-tools.h
```
Four files (`/usr/local/lib/libomp.dylib`, `/usr/local/include/ompt.h`, `/usr/local/include/omp.h`, `/usr/local/include/omp-tools.h`) are installed accordingly.
For `chatglm-6b-int4`, modify the [quantization.py](https://huggingface.co/THUDM/chatglm-6b-int4/blob/main/quantization.py)file. In the file, change the `gcc -O3 -fPIC -pthread -fopenmp -std=c99` configuration to `gcc -O3 -fPIC -Xclang -fopenmp -pthread -lomp -std=c99`，[corresponding python code](https://huggingface.co/THUDM/chatglm-6b-int4/blob/63d66b0572d11cedd5574b38da720299599539b3/quantization.py#L168), i.e.:
#### Configure `gcc` with `-fopenmp`
Next, modify the [quantization.py](https://huggingface.co/THUDM/chatglm-6b-int4/blob/main/quantization.py) file of the `chatglm-6b-int4` project. In the file, change the `gcc -O3 -fPIC -pthread -fopenmp -std=c99` configuration to `gcc -O3 -fPIC -Xclang -fopenmp -pthread -lomp -std=c99` (check the corresponding python code [HERE](https://huggingface.co/THUDM/chatglm-6b-int4/blob/63d66b0572d11cedd5574b38da720299599539b3/quantization.py#L168)), i.e.:
```python
# STEP
## Change the line contains `gcc -O3 -fPIC -pthread -fopenmp -std=c99` to:
# STEP 2: Change the line contains `gcc -O3 -fPIC -pthread -fopenmp -std=c99` to:
compile_command = "gcc -O3 -fPIC -Xclang -fopenmp -pthread -lomp -std=c99 {} -shared -o {}".format(source_code, kernel_file)
```
@ -231,14 +227,14 @@ import platform
## ...
## change the corresponding lines to:
if platform.uname()[0] == 'Darwin':
compile_command = "gcc -O3 -fPIC -Xclang -fopenmp -pthread -lomp -std=c99-o {}".format(
compile_command = "gcc -O3 -fPIC -Xclang -fopenmp -pthread -lomp -std=c99 -o {}".format(
source_code, kernel_file)
else:
compile_command = "gcc -O3 -fPIC -pthread -fopenmp -std=c99 {} -shared -o {}".format(
source_code, kernel_file)
```
> Notice: If you have run the `chatglm` project and failed, you may want to clean the cache of Huggingface before your next try, i.e. `rm -rf ${HOME}/.cache/huggingface/modules/transformers_modules/chatglm-6b-int4`. Since `rm` is used, please MAKE SURE that the command deletes the right files.
> Notice: If you have run the `ChatGLM` project and failed, you may want to clean the cache of Huggingface before your next try, i.e. `rm -rf ${HOME}/.cache/huggingface/modules/transformers_modules/chatglm-6b-int4`. Since `rm` is used, please MAKE SURE that the command deletes the right files.
### GPU Inference on Mac
For Macs (and MacBooks) with Apple Silicon, it is possible to use the MPS backend to run ChatGLM-6B on the GPU. First, you need to refer to Apple's [official instructions](https://developer.apple.com/metal/pytorch) to install PyTorch-Nightly.
@ -248,7 +244,7 @@ Currently you must [load the model locally](README_en.md#load-the-model-locally)
model = AutoModel.from_pretrained("your local path", trust_remote_code=True).half().to('mps')
```
For Mac users with Mac >= 13.3, one may encounter errors related to `half()` method. Use `float()` instead:
For Mac users with Mac OS >= 13.3, one may encounter errors related to the `half()` method. Use the `float()` method instead:
```python
model = AutoModel.from_pretrained("your local path", trust_remote_code=True).float().to('mps')
@ -256,7 +252,7 @@ model = AutoModel.from_pretrained("your local path", trust_remote_code=True).flo
Then you can use GPU-accelerated model inference on Mac.
> Notice: there is no promblem to run the non-quantified version of ChatGLM with MPS. One could check [this issue](https://github.com/THUDM/ChatGLM-6B/issues/462) to run the quantified version with MPS as the backend (and get `ERRORS`). Unzip/unpack [kernel](https://huggingface.co/THUDM/chatglm-6b/blob/658202d88ac4bb782b99e99ac3adff58b4d0b813/quantization.py#L27) as an `ELF` file shows its backend is `cuda`.
> Notice: there is no promblem to run the non-quantified version of ChatGLM with MPS. One could check [this issue](https://github.com/THUDM/ChatGLM-6B/issues/462) to run the quantified version with MPS as the backend (and get `ERRORS`). Unpacking [kernel](https://huggingface.co/THUDM/chatglm-6b/blob/658202d88ac4bb782b99e99ac3adff58b4d0b813/quantization.py#L27) as an `ELF` file shows its backend is `cuda`, which fails on MPS currently (`torch 2.1.0.dev20230502`).
### Multi-GPU Deployment
If you have multiple GPUs, but the memory size of each GPU is not sufficient to accommodate the entire model, you can split the model across multiple GPUs.