apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 16:48:03 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

update installation readme (#3429)

Browse Source
This commit is contained in:
yongqiangma
2025-08-15 19:09:41 +08:00
committed by GitHub
parent 615930bc05
commit 5703d7aa0f
4 changed files with 58 additions and 26 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
docs/get_started/installation/Enflame_gcu.md
View File

@@ -53,21 +53,24 @@ After driver installation, **re-enter the Docker container**:
docker start paddle-gcu-llm
docker exec -it paddle-gcu-llm bash
```
5. Install PaddlePaddle
5. Install PaddlePaddle & PaddleCustomDevice<br/>
```bash
# PaddlePaddle Deep Learning Framework provides fundamental computing capabilities
python -m pip install paddlepaddle==3.1.0a0 -i https://www.paddlepaddle.org.cn/packages/stable/cpu/
```
6. Install PaddleCustomDevice<br/>
```bash
python -m pip install paddlepaddle==3.1.1 -i https://www.paddlepaddle.org.cn/packages/stable/cpu/
# PaddleCustomDevice implements custom hardware backend for PaddlePaddle, providing GCU operator implementations
python -m pip install paddle-custom-gcu==3.1.0 -i https://www.paddlepaddle.org.cn/packages/stable/gcu/
python -m pip install paddle-custom-gcu==3.1.1 -i https://www.paddlepaddle.org.cn/packages/stable/gcu/
# For source compilation, refer to: https://github.com/PaddlePaddle/PaddleCustomDevice/blob/develop/backends/gcu/README_cn.md
```
7. Install FastDeploy and dependencies
For latest paddle verion on iluvatar. Refer to [PaddlePaddle Installation](https://www.paddlepaddle.org.cn/)
6. Install FastDeploy and dependencies
```bash
python -m pip install fastdeploy -i https://www.paddlepaddle.org.cn/packages/stable/gcu/ --extra-index-url https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simplels
# For source compilation, refer to the following steps
```
You can build FastDeploy from source if you need the ```latest version```.
```bash
git clone https://github.com/PaddlePaddle/FastDeploy
cd FastDeploy
python -m pip install -r requirements.txt --extra-index-url https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simplels

21
docs/get_started/installation/iluvatar_gpu.md
View File

@@ -18,7 +18,7 @@ docker pull ccr-2vdh3abv-pub.cnc.bj.baidubce.com/device/paddle-ixuca:latest
```
## Container Preparation
1. Start Container
### Start Container
```bash
docker run -itd --name paddle_infer -v /usr/src:/usr/src -v /lib/modules:/lib/modules -v /dev:/dev -v /home/paddle:/home/paddle --privileged --cap-add=ALL --pid=host ccr-2vdh3abv-pub.cnc.bj.baidubce.com/device/paddle-ixuca:latest
@@ -27,13 +27,26 @@ docker exec -it paddle_infer bash
/home/paddle contains the model files, *.whl packages, and scripts.
1. Install packages
### Install paddle
```bash
pip3 install paddlepaddle==3.1.0a0 -i https://www.paddlepaddle.org.cn/packages/stable/cpu/
pip3 install paddle-iluvatar-gpu==3.1.0 -i https://www.paddlepaddle.org.cn/packages/stable/ixuca/
pip3 install paddlepaddle==3.1.1 -i https://www.paddlepaddle.org.cn/packages/stable/cpu/
pip3 install paddle-iluvatar-gpu==3.1.1 -i https://www.paddlepaddle.org.cn/packages/stable/ixuca/
```
For latest paddle verion on iluvatar. Refer to [PaddlePaddle Installation](https://www.paddlepaddle.org.cn/)
### Install or build FastDeploy
```bash
pip3 install fastdeploy_iluvatar_gpu==2.1.0.dev0 -i https://www.paddlepaddle.org.cn/packages/stable/ixuca/ --extra-index-url https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simplels
```
You can build FastDeploy from source if you need the ```latest version```.
```bash
git clone https://github.com/PaddlePaddle/FastDeploy
cd FastDeploy
pip install -r requirements_iluvatar.txt
export LD_PRELOAD=/usr/local/corex/lib64/libcuda.so.1
bash build.sh
```
## Prepare the inference demo script

19
docs/zh/get_started/installation/Enflame_gcu.md
View File

@@ -52,21 +52,24 @@ bash TopsRider_i3x_*_deb_amd64.run --driver --no-auto-load -y
docker start paddle-gcu-llm
docker exec -it paddle-gcu-llm bash
```
5. 安装 PaddlePaddle<br/>
5. 安装 PaddlePaddle & PaddleCustomDevice<br/>
```bash
# PaddlePaddle『飞桨』深度学习框架，提供运算基础能力
python -m pip install paddlepaddle==3.1.0a0 -i https://www.paddlepaddle.org.cn/packages/stable/cpu/
```
6. 安装 PaddleCustomDevice<br/>
```bash
python -m pip install paddlepaddle==3.1.1 -i https://www.paddlepaddle.org.cn/packages/stable/cpu/
# PaddleCustomDevice是PaddlePaddle『飞桨』深度学习框架的自定义硬件接入实现，提供GCU的算子实现
python -m pip install paddle-custom-gcu==3.1.0 -i https://www.paddlepaddle.org.cn/packages/stable/gcu/
python -m pip install paddle-custom-gcu==3.1.1 -i https://www.paddlepaddle.org.cn/packages/stable/gcu/
# 如想源码编译安装，请参考https://github.com/PaddlePaddle/PaddleCustomDevice/blob/develop/backends/gcu/README_cn.md
```
7. 安装 FastDeploy <br/>
获取Paddle的最新安装版本： [PaddlePaddle Installation](https://www.paddlepaddle.org.cn/)
6. 安装 FastDeploy <br/>
```bash
python -m pip install fastdeploy -i https://www.paddlepaddle.org.cn/packages/stable/gcu/ --extra-index-url https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simplels
# 如想源码编译安装，请参考如下步骤
```
可以按如下步骤编译FastDeploy，得到```最新版本```.
```bash
git clone https://github.com/PaddlePaddle/FastDeploy
cd FastDeploy
python -m pip install -r requirements.txt --extra-index-url https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simplels

25
docs/zh/get_started/installation/iluvatar_gpu.md
View File

@@ -1,5 +1,5 @@
# 如何在天数机器上运行 ERNIE-4.5-300B-A47B-BF16 & ERNIE-4.5-21B-A3B
当前版本软件只是作为天数芯片 + Fastdeploy 推理大模型的一个演示 demo，跑最新ERNIE4.5模型可能存在问题，后续进行修复和性能优化，给客户提供一个更稳定的版本。
该软件的当前版本仅作为Iluvatar CoreX与大型模型的Fastdeploy推理框架相结合的演示。在GSM8K数据集上运行最新的ERNIE4.5 300B模型大约需要6.3小时。
## 准备机器
首先您需要准备以下配置的机器
@@ -18,7 +18,7 @@ docker pull ccr-2vdh3abv-pub.cnc.bj.baidubce.com/device/paddle-ixuca:latest
```
## 准备容器
1. 启动容器
### 启动容器
```bash
docker run -itd --name paddle_infer -v /usr/src:/usr/src -v /lib/modules:/lib/modules -v /dev:/dev -v /home/paddle:/home/paddle --privileged --cap-add=ALL --pid=host ccr-2vdh3abv-pub.cnc.bj.baidubce.com/device/paddle-ixuca:latest
@@ -27,12 +27,25 @@ docker exec -it paddle_infer bash
/home/paddle 为模型文件、whl包、脚本所在目录
1. 安装whl包
### 安装paddle
```bash
pip3 install paddlepaddle==3.1.0a0 -i https://www.paddlepaddle.org.cn/packages/stable/cpu/
pip3 install paddle-iluvatar-gpu==3.1.0 -i https://www.paddlepaddle.org.cn/packages/stable/ixuca/
pip3 install fastdeploy_iluvatar_gpu -i https://www.paddlepaddle.org.cn/packages/stable/ixuca/ --extra-index-url https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simplels
pip3 install paddlepaddle==3.1.1 -i https://www.paddlepaddle.org.cn/packages/stable/cpu/
pip3 install paddle-iluvatar-gpu==3.1.1 -i https://www.paddlepaddle.org.cn/packages/stable/ixuca/
```
获取Paddle的最新安装版本： [PaddlePaddle Installation](https://www.paddlepaddle.org.cn/)
### 安装fastdeploy
```bash
pip3 install fastdeploy_iluvatar_gpu==2.1.0.dev0 -i https://www.paddlepaddle.org.cn/packages/stable/ixuca/ --extra-index-url https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simplels
```
可以按如下步骤编译FastDeploy，，得到```最新版本```.
```bash
git clone https://github.com/PaddlePaddle/FastDeploy
cd FastDeploy
pip install -r requirements_iluvatar.txt
export LD_PRELOAD=/usr/local/corex/lib64/libcuda.so.1
bash build.sh
```
## 准备推理demo脚本