Merge pull request X-PLUG#24 from butyuhao/main

revise readme

README.md

@@ -99,10 +99,17 @@ The evaluation dataset OwlEval can be found in ```./OwlEval```.
 ### Install Requirements
 1. Create conda environment
 ```bash
-conda create -n mplug_owl python==3.10
+conda create -n mplug_owl python=3.10
 conda activate mplug_owl
 ```
-2. Install apex (remove apex dependency in the next release)
+
+2. Install PyTorch
+
+```
+conda install pytorch==1.13.1 torchvision==0.14.1 torchaudio==0.13.1 pytorch-cuda=11.7 -c pytorch -c nvidia
+```
+
+3. Install apex (remove apex dependency in the next release)
 
    Apex needs to be manually compiled from source code, because mPLUG-Owl rely on its  cpp extension (MixedFusedLayerNorm).
 
@@ -113,7 +120,7 @@ cd apex_22.01_pp
 TORCH_CUDA_ARCH_LIST='5.2 6.0 6.1 7.0 7.5 8.0 8.6' pip install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./
 ```
 
-3. Install other dependencies
+4. Install other dependencies
 ```bash
 pip install -r requirements.txt
 ```

README_zh.md

@@ -99,10 +99,17 @@ Qinghao Ye*, Haiyang Xu*, Guohai Xu*, Jiabo Ye, Ming Yan†, Yiyang Zhou, Junyan
 ### 安装依赖
 1. 创建conda环境
 ```bash
-conda create -n mplug_owl python==3.10
+conda create -n mplug_owl python=3.10
 conda activate mplug_owl
 ```
-2. 安装 Apex（在下一个版本中将移除Apex依赖）
+
+2. 安装PyTorch
+
+```
+conda install pytorch==1.13.1 torchvision==0.14.1 torchaudio==0.13.1 pytorch-cuda=11.7 -c pytorch -c nvidia
+```
+
+3. 安装 Apex（在下一个版本中将移除Apex依赖）
 
    Apex 需要从源代码手动编译，因为 mPLUG-Owl 依赖于其 cpp 扩展（MixedFusedLayerNorm）。
 
@@ -113,7 +120,7 @@ cd apex_22.01_pp
 TORCH_CUDA_ARCH_LIST='5.2 6.0 6.1 7.0 7.5 8.0 8.6' pip install -v --disable-pip-version-check --no-cache-dir --global-option="--cpp_ext" --global-option="--cuda_ext" ./
 ```
 
-3. 安装其它依赖
+4. 安装其它依赖
 ```bash
 pip install -r requirements.txt
 ```