You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@singa.apache.org by zh...@apache.org on 2023/11/01 13:19:18 UTC
(singa-doc) branch master updated: 23-11-1 python versions docs
This is an automated email from the ASF dual-hosted git repository.
zhaojing pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/singa-doc.git
The following commit(s) were added to refs/heads/master by this push:
new 8283d6e 23-11-1 python versions docs
new 37f5882 Merge pull request #78 from lzjpaul/23-11-1-python
8283d6e is described below
commit 8283d6e17203f49dfe8c4983f85e6be1b58afe1a
Author: zhaojing <zh...@comp.nus.edu.sg>
AuthorDate: Wed Nov 1 21:12:42 2023 +0800
23-11-1 python versions docs
---
docs-site/docs/installation.md | 12 ++++--------
docs-site/docs/wheel-cpu.md | 6 +++---
docs-site/docs/wheel-gpu.md | 6 +++---
.../website/versioned_docs/version-4.1.0/installation.md | 12 ++++--------
docs-site/website/versioned_docs/version-4.1.0/wheel-cpu.md | 6 +++---
docs-site/website/versioned_docs/version-4.1.0/wheel-gpu.md | 6 +++---
.../versioned_docs/version-4.1.0_Chinese/installation.md | 6 ++----
.../versioned_docs/version-4.1.0_Viet/installation.md | 9 +++------
8 files changed, 25 insertions(+), 38 deletions(-)
diff --git a/docs-site/docs/installation.md b/docs-site/docs/installation.md
index fac26ef..2f92949 100644
--- a/docs-site/docs/installation.md
+++ b/docs-site/docs/installation.md
@@ -7,11 +7,7 @@ title: Installation
## Using Pip
-[Miniconda3](https://conda.io/miniconda.html) is recommended to use with SINGA.
-After installing miniconda, execute the one of the following commands to install
-SINGA.
-
-**SINGA works with python 3.6, 3.7 and 3.8.**
+**SINGA works with python 3.9, 3.10 and 3.11.**
1. CPU only
[](https://colab.research.google.com/drive/17RA056Brwk0vBQTFaZ-l9EbqwADO0NA9?usp=sharing)
@@ -36,8 +32,8 @@ You can also configure SINGA version and the CUDA version, like
version are listed at the link.
Note: the Python version of your local Python environment will be used to find
-the corresponding wheel package. For example, if your local Python is 3.6, then
-the wheel package compiled on Python 3.6 will be selected by pip and installed.
+the corresponding wheel package. For example, if your local Python is 3.9, then
+the wheel package compiled on Python 3.9 will be selected by pip and installed.
In fact, the wheel file's name include SINGA version, CUDA version and Python
version. Therefore, `pip` knows which wheel file to download and install.
@@ -110,7 +106,7 @@ building tools or conda-build, on local host OS or in a Docker container.
```
The folder of `_singa_wrap.so` is like
- `~/miniconda3/lib/python3.7/site-packages/singa`. Normally, the error is
+ `~/miniconda3/lib/python3.10/site-packages/singa`. Normally, the error is
caused by the mismatch or missing of dependent libraries, e.g. cuDNN or
protobuf. The solution is to create a new virtual environment and install
SINGA in that environment, e.g.,
diff --git a/docs-site/docs/wheel-cpu.md b/docs-site/docs/wheel-cpu.md
index bda4311..f57cb11 100644
--- a/docs-site/docs/wheel-cpu.md
+++ b/docs-site/docs/wheel-cpu.md
@@ -7,9 +7,9 @@ title: CPU only
## 4.1.0
-- [Python 3.6](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0-cp36-cp36m-manylinux2014_x86_64.whl)
-- [Python 3.7](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0-cp37-cp37m-manylinux2014_x86_64.whl)
-- [Python 3.8](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0-cp38-cp38-manylinux2014_x86_64.whl)
+- [Python 3.9](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0-cp39-cp39-manylinux_2_28_x86_64.whl)
+- [Python 3.10](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0-cp310-cp310-manylinux_2_28_x86_64.whl)
+- [Python 3.11](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0-cp311-cp311-manylinux_2_28_x86_64.whl)
## 4.0.0
diff --git a/docs-site/docs/wheel-gpu.md b/docs-site/docs/wheel-gpu.md
index 45f029a..3f9d3e4 100644
--- a/docs-site/docs/wheel-gpu.md
+++ b/docs-site/docs/wheel-gpu.md
@@ -7,9 +7,9 @@ title: CUDA enabled
## 4.1.0
-- [CUDA10.2, cuDNN 7.6.5, Python 3.6](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0%2Bcuda10.2-cp36-cp36m-manylinux2014_x86_64.whl)
-- [CUDA10.2, cuDNN 7.6.5, Python 3.7](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0%2Bcuda10.2-cp37-cp37m-manylinux2014_x86_64.whl)
-- [CUDA10.2, cuDNN 7.6.5, Python 3.8](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0%2Bcuda10.2-cp38-cp38-manylinux2014_x86_64.whl)
+- [CUDA10.2, cuDNN 7.6.5, Python 3.9](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0%2Bcuda10.2-cp39-cp39-manylinux_2_28_x86_64.whl)
+- [CUDA10.2, cuDNN 7.6.5, Python 3.10](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0%2Bcuda10.2-cp310-cp310-manylinux_2_28_x86_64.whl)
+- [CUDA10.2, cuDNN 7.6.5, Python 3.11](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0%2Bcuda10.2-cp311-cp311-manylinux_2_28_x86_64.whl)
## 4.0.0
diff --git a/docs-site/website/versioned_docs/version-4.1.0/installation.md b/docs-site/website/versioned_docs/version-4.1.0/installation.md
index 676b4d1..1ecdcf2 100644
--- a/docs-site/website/versioned_docs/version-4.1.0/installation.md
+++ b/docs-site/website/versioned_docs/version-4.1.0/installation.md
@@ -8,11 +8,7 @@ original_id: installation
## Using Pip
-[Miniconda3](https://conda.io/miniconda.html) is recommended to use with SINGA.
-After installing miniconda, execute the one of the following commands to install
-SINGA.
-
-**SINGA works with python 3.6, 3.7 and 3.8.**
+**SINGA works with python 3.9, 3.10 and 3.11.**
1. CPU only
[](https://colab.research.google.com/drive/17RA056Brwk0vBQTFaZ-l9EbqwADO0NA9?usp=sharing)
@@ -37,8 +33,8 @@ You can also configure SINGA version and the CUDA version, like
version are listed at the link.
Note: the Python version of your local Python environment will be used to find
-the corresponding wheel package. For example, if your local Python is 3.6, then
-the wheel package compiled on Python 3.6 will be selected by pip and installed.
+the corresponding wheel package. For example, if your local Python is 3.9, then
+the wheel package compiled on Python 3.9 will be selected by pip and installed.
In fact, the wheel file's name include SINGA version, CUDA version and Python
version. Therefore, `pip` knows which wheel file to download and install.
@@ -111,7 +107,7 @@ building tools or conda-build, on local host OS or in a Docker container.
```
The folder of `_singa_wrap.so` is like
- `~/miniconda3/lib/python3.7/site-packages/singa`. Normally, the error is
+ `~/miniconda3/lib/python3.10/site-packages/singa`. Normally, the error is
caused by the mismatch or missing of dependent libraries, e.g. cuDNN or
protobuf. The solution is to create a new virtual environment and install
SINGA in that environment, e.g.,
diff --git a/docs-site/website/versioned_docs/version-4.1.0/wheel-cpu.md b/docs-site/website/versioned_docs/version-4.1.0/wheel-cpu.md
index 30154e2..499101f 100644
--- a/docs-site/website/versioned_docs/version-4.1.0/wheel-cpu.md
+++ b/docs-site/website/versioned_docs/version-4.1.0/wheel-cpu.md
@@ -8,9 +8,9 @@ original_id: wheel-cpu
## 4.1.0
-- [Python 3.6](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0-cp36-cp36m-manylinux2014_x86_64.whl)
-- [Python 3.7](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0-cp37-cp37m-manylinux2014_x86_64.whl)
-- [Python 3.8](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0-cp38-cp38-manylinux2014_x86_64.whl)
+- [Python 3.9](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0-cp39-cp39-manylinux_2_28_x86_64.whl)
+- [Python 3.10](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0-cp310-cp310-manylinux_2_28_x86_64.whl)
+- [Python 3.11](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0-cp311-cp311-manylinux_2_28_x86_64.whl)
## 4.0.0
diff --git a/docs-site/website/versioned_docs/version-4.1.0/wheel-gpu.md b/docs-site/website/versioned_docs/version-4.1.0/wheel-gpu.md
index 92c58e1..da87281 100644
--- a/docs-site/website/versioned_docs/version-4.1.0/wheel-gpu.md
+++ b/docs-site/website/versioned_docs/version-4.1.0/wheel-gpu.md
@@ -8,9 +8,9 @@ original_id: wheel-gpu
## 4.1.0
-- [CUDA10.2, cuDNN 7.6.5, Python 3.6](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0%2Bcuda10.2-cp36-cp36m-manylinux2014_x86_64.whl)
-- [CUDA10.2, cuDNN 7.6.5, Python 3.7](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0%2Bcuda10.2-cp37-cp37m-manylinux2014_x86_64.whl)
-- [CUDA10.2, cuDNN 7.6.5, Python 3.8](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0%2Bcuda10.2-cp38-cp38-manylinux2014_x86_64.whl)
+- [CUDA10.2, cuDNN 7.6.5, Python 3.9](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0%2Bcuda10.2-cp39-cp39-manylinux_2_28_x86_64.whl)
+- [CUDA10.2, cuDNN 7.6.5, Python 3.10](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0%2Bcuda10.2-cp310-cp310-manylinux_2_28_x86_64.whl)
+- [CUDA10.2, cuDNN 7.6.5, Python 3.11](https://singa-wheel.s3.ap-southeast-1.amazonaws.com/singa-4.1.0%2Bcuda10.2-cp311-cp311-manylinux_2_28_x86_64.whl)
## 4.0.0
diff --git a/docs-site/website/versioned_docs/version-4.1.0_Chinese/installation.md b/docs-site/website/versioned_docs/version-4.1.0_Chinese/installation.md
index 3c66f6a..51fd397 100644
--- a/docs-site/website/versioned_docs/version-4.1.0_Chinese/installation.md
+++ b/docs-site/website/versioned_docs/version-4.1.0_Chinese/installation.md
@@ -8,8 +8,6 @@ original_id: installation
## 使用pip
-推荐使用[Miniconda3](https://conda.io/miniconda.html)来配合SINGA使用，安装miniconda后，执行以下命令之一安装SINGA。
-
1. 只使用CPU
[](https://colab.research.google.com/drive/17RA056Brwk0vBQTFaZ-l9EbqwADO0NA9?usp=sharing)
@@ -34,7 +32,7 @@ pip install singa -f http://singa.apache.org/docs/next/wheel-gpu.html --trusted-
要安装最新的开发版本，请将链接替换为
http://singa.apache.org/docs/next/wheel-gpu-dev.html
-注意：你本地Python环境的Python版本将被用来寻找相应的wheel包。例如，如果你本地的Python是3.6，那么就会通过pip选择在Python 3.6上编译的wheel包并安装。事实上，wheel文件的名称包括SINGA版本、CUDA版本和Python版本。因此，`pip`知道要下载和安装哪个wheel文件。
+注意：你本地Python环境的Python版本将被用来寻找相应的wheel包。例如，如果你本地的Python是3.9，那么就会通过pip选择在Python 3.9上编译的wheel包并安装。事实上，wheel文件的名称包括SINGA版本、CUDA版本和Python版本。因此，`pip`知道要下载和安装哪个wheel文件。
参考setup.py文件顶部的注释，了解如何构建wheel包。
@@ -93,7 +91,7 @@ version-(cpu|gpu)[-devel]
>> import importlib
>> importlib.import_module('_singa_wrap')
```
- `_singa_wrap.so` 的文件夹是 `~/miniconda3/lib/python3.7/site-packages/singa`。通常情况下，这个错误是由于依赖的库不匹配或缺失造成的，例如 cuDNN 或 protobuf。解决方法是创建一个新的虚拟环境，并在该环境中安装SINGA，例如：
+ `_singa_wrap.so` 的文件夹是 `~/miniconda3/lib/python3.10/site-packages/singa`。通常情况下，这个错误是由于依赖的库不匹配或缺失造成的，例如 cuDNN 或 protobuf。解决方法是创建一个新的虚拟环境，并在该环境中安装SINGA，例如：
```shell
diff --git a/docs-site/website/versioned_docs/version-4.1.0_Viet/installation.md b/docs-site/website/versioned_docs/version-4.1.0_Viet/installation.md
index 287025d..92ca110 100644
--- a/docs-site/website/versioned_docs/version-4.1.0_Viet/installation.md
+++ b/docs-site/website/versioned_docs/version-4.1.0_Viet/installation.md
@@ -8,9 +8,6 @@ original_id: installation
## Sử dụng Pip
-[Miniconda3](https://conda.io/miniconda.html) được khuyến khích dùng với SINGA.
-Sau khi cài đặt miniconda, thực hiện các lệnh sau để cài đặt SINGA.
-
1. Cho CPU
[](https://colab.research.google.com/drive/17RA056Brwk0vBQTFaZ-l9EbqwADO0NA9?usp=sharing)
@@ -39,8 +36,8 @@ sách tổ hợp phiên bản SINGA với CUDA được cung cấp trong đườ
http://singa.apache.org/docs/next/wheel-gpu-dev.html
Lưu ý: phiên bản Python của Python environment trong máy của bạn sẽ được sử dụng
-để tìm gói wheel tương ứng. Ví dụ, nếu bạn sử dụng Python 3.6, thì gói wheel kết
-cấu trong Python 3.6 sẽ được pip chọn để cài đặt. Thực tế, tên của tệp tin wheel
+để tìm gói wheel tương ứng. Ví dụ, nếu bạn sử dụng Python 3.9, thì gói wheel kết
+cấu trong Python 3.9 sẽ được pip chọn để cài đặt. Thực tế, tên của tệp tin wheel
bao gồm phiên bản SINGA, phiên bản CUDA và Python. Vì thế, `pip` biết tệp tin
wheel nào để tải và cài đặt.
@@ -113,7 +110,7 @@ os) hay trong Docker container.
```
Thư mục chứa `_singa_wrap.so` thường ở
- `~/miniconda3/lib/python3.7/site-packages/singa`. Thông thường, lỗi này được
+ `~/miniconda3/lib/python3.10/site-packages/singa`. Thông thường, lỗi này được
gây ra bởi sự không tương thích hoặc thiếu các thư viện dependent libraries,
v.d cuDNN hay protobuf. Cách giải quyết là tạo một virtual environment mới và
cài đặt SINGA trong environment đó, v.d,