更新安装

torchpipe · Sep 21, 2023 · cd19ba9 · cd19ba9
1 parent 2a458b4
commit cd19ba9
Show file tree

Hide file tree

Showing 2 changed files with 13 additions and 7 deletions.
diff --git a/docs/installation.mdx b/docs/installation.mdx
@@ -11,11 +11,11 @@ type: explainer
 ## Dependencies
 The basic environment is as follows:
 
-| Dependency         | Version   | Remarks                                                                                                                                                                                                                              |
-|--------------------|:----------|--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
-| `cuda`             | >= 11.0   | - The CUDA version must be consistent with the version that PyTorch depends on (for the convenience of `torch.utils.cpp_extension` to compile code on the fly). <br />- Compatibility testing with CUDA 10.2 is no longer performed. |
-| `pytorch`          | >= 1.13.0 | - Compatibility testing is no longer performed for `c++14/cuda-10.2/pytorch==1.10.2`, but you may still be able to run it with simple modifications.                                                                                 |
-| `opencv`(optional) | 3.x, 4.x  | At least the core, imgproc, imgcodecs, and highgui modules are included.                                                                                                                                                             |
+| Dependency         | Version           | Remarks                                                                                                                                                                                                                                                    |
+|--------------------|:------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
+| `cuda`             | >= 11.0           | - The CUDA version must be consistent with the version that PyTorch depends on (for the convenience of `torch.utils.cpp_extension` to compile code on the fly). <br />- Compatibility testing with CUDA 10.2(no support for c++17) is no longer performed. |
+| `pytorch`          | >= 1.10.0(cuda11) | - Compatibility testing is no longer performed for `c++14/cuda-10.2/pytorch==1.10.2`, but you may still be able to run it with simple modifications.                                                                                                       |
+| `opencv`(optional) | 4.x               | At least the core, imgproc, imgcodecs, and highgui modules are included.                                                                                                                                                                                   |
 |  `tensorrt`(optional)	|>= 7.2<br /><= 8.6.1.5| - There is a [memory leak](https://github.com/NVIDIA/TensorRT/issues/351) in TensorRT 7.0 with dynamic inputs.<br />- TensorRT 7.1 and earlier cannot directly INT8-quantize models with dynamic inputs.
 
 
@@ -25,6 +25,10 @@ All dependencies mentioned above come from a specific default backend. The const
 
 ## Using NGC base image {#NGC}
 The easiest way is to choose NGC mirror for source code compilation (official mirror may still be able to run low version drivers through Forward Compatibility or Minor Version Compatibility). 
+
+- Minimum support nvcr.io/nvidia/pytorch:21.07-py3
+- Maximum support nvcr.io/nvidia/pytorch:23.08-py3
+
 First, clone the code:
 
 ```bash

diff --git a/i18n/zh/docusaurus-plugin-content-docs/current/installation.mdx b/i18n/zh/docusaurus-plugin-content-docs/current/installation.mdx
@@ -14,8 +14,8 @@ type: explainer
 
 | 依赖             | 版本                   | 备注                                                                                                                                          |
 |------------------|:-----------------------|---------------------------------------------------------------------------------------------------------------------------------------------|
-| `cuda`           | >= 11.0                | - 与PyTorch依赖的cuda版本必须一致（便于`torch.utils.cpp_extension`即时编译代码） <br />- 不再对CUDA 10.2的兼容进行测试。                         |
-| `pytorch`        | >= 1.13.0              | - 对于`c++14/cuda-10.2/pytorch==1.10.2`不再进行兼容测试，然而通过简单修改，您可能仍可运行。                                                      |
+| `cuda`           | >= 11.0                | - 与PyTorch依赖的cuda版本必须一致（便于`torch.utils.cpp_extension`即时编译代码） <br />- 不再对CUDA 10.2（不支持c++17）的兼容进行测试。            |
+| `pytorch`        | >= 1.10.0（cuda11）      | - 对于`c++14/cuda-10.2/pytorch==1.10.2`不再进行兼容测试，然而通过简单修改，您可能仍可运行。                                                      |
 | `opencv`(可选)   | 3.x, 4.x               | 至少包含 core, imgproc，imgcodecs 和 highgui 四个模块                                                                                          |
 | `tensorrt`(可选) | >= 7.2<br /><= 8.6.1.5 | - tensorrt 7.0 动态输入下存在[内存泄漏](https://github.com/NVIDIA/TensorRT/issues/351)<br />- tensorrt 7.1 及以下无法直接int8化动态输入的模型 |
 
@@ -25,6 +25,8 @@ type: explainer
 
 ## 使用NGC基础镜像 {#NGC}
 最简单的方式是可以选择<font color='Brown'>NGC镜像</font>进行源码编译（低版本驱动依靠 Forward Compatibility 或者 Minor Version Compatibility 依然可能跑起来官方镜像）。
+- 最低支持 nvcr.io/nvidia/pytorch:21.07-py3
+- 最高支持 nvcr.io/nvidia/pytorch:23.08-py3
 首先，克隆代码：
 
 ```bash