Skip to content

Commit 8988d32

Browse files
authored
update installation guide for 1.10.200 release (#1676)
update installation guide for 1.10.200+gpu release
1 parent 68ecf92 commit 8988d32

File tree

1 file changed

+55
-28
lines changed

1 file changed

+55
-28
lines changed

docs/tutorials/installation.md

Lines changed: 55 additions & 28 deletions
Original file line numberDiff line numberDiff line change
@@ -1,9 +1,7 @@
1-
# Build and Install from Source Code
1+
Installation Guide
2+
==================
23

3-
This is guide to build an Intel® Extension for PyTorch* PyPI package from source and install it in Linux.
4-
5-
6-
## Prepare
4+
## System Requirements
75

86
### Hardware Requirement
97

@@ -16,7 +14,17 @@ Verified Hardware Platforms:
1614
- Intel GPU Drivers
1715
- Intel® Data Center GPU Flex Series [419.40](https://dgpu-docs.intel.com/releases/stable_419_40_20220914.html)
1816
- Intel® oneAPI Base Toolkit 2022.3
19-
- Python 3.7-3.10
17+
- Python 3.6-3.9
18+
19+
## PyTorch-Intel® Extension for PyTorch\* Version Mapping
20+
21+
Intel® Extension for PyTorch\* has to work with a corresponding version of PyTorch. Here are the PyTorch versions that we support and the mapping relationship:
22+
23+
|PyTorch Version|Extension Version|
24+
|--|--|
25+
|[v1.10.0](https://github.com/pytorch/pytorch/tree/v1.10.0) (patches needed)|[v1.10.\*](https://github.com/intel/intel-extension-for-pytorch/tree/v1.10.200+gpu)|
26+
27+
## Preparations
2028

2129
### Install Intel GPU Driver
2230

@@ -34,51 +42,70 @@ Need to install components of Intel® oneAPI Base Toolkit:
3442

3543
Default installation location is /opt/intel/oneapi for root account, ${HOME}/intel/oneapi for other accounts.
3644

37-
### Configure the AOT
45+
## Install via wheel files
3846

39-
Please refer to [AOT documentation](./AOT.md) for how to configure AOT.
47+
Prebuilt wheel files availability matrix for Python versions
4048

41-
### Build and Install from Source Code
49+
| Extension Version | Python 3.6 | Python 3.7 | Python 3.8 | Python 3.9 | Python 3.10 |
50+
| :--: | :--: | :--: | :--: | :--: | :--: |
51+
| 1.10.200+gpu | ✔️ | ✔️ | ✔️ | ✔️ | |
4252

43-
Make sure PyTorch is installed so that the extension will work properly. For each PyTorch release, we have a corresponding release of the extension. Here are the PyTorch versions that we support and the mapping relationship:
53+
### Install PyTorch
4454

45-
|PyTorch Version|Intel® Extension for PyTorch* Version|
46-
|--|--|
47-
|[v1.10.\*](https://github.com/pytorch/pytorch/tree/v1.10.0 "v1.10.0")|[v1.10.\*](https://github.com/intel/intel-extension-for-pytorch/tree/v1.10.200)|
55+
```bash
56+
python -m pip install torch==1.10.0a0 -f https://developer.intel.com/ipex-whl-stable-xpu
57+
```
58+
59+
### Install Intel® Extension for PyTorch\*
60+
61+
```bash
62+
python -m pip install intel_extension_for_pytorch==1.10.200+gpu -f https://developer.intel.com/ipex-whl-stable-xpu
63+
```
4864

65+
## Install via compiling from source
4966

50-
Build and Install PyTorch:
67+
### Download source code of PyTorch and Intel® Extension for PyTorch\*:
5168

5269
```bash
5370
$ git clone https://github.com/pytorch/pytorch.git
5471
$ cd pytorch
55-
# checkout to specific release branch if in need
56-
$ git checkout ${PYTORCH_RELEASE_BRANCH_NAME}
57-
# apply git patch to pytorch code, e.g., apply patch for pytorch v1.10.
58-
$ git apply ${intel_extension_for_pytorch_directory}/torch_patches/{xpu-1.10}.patch
72+
$ git checkout v1.10.0
73+
74+
$ git clone https://github.com/intel/intel-extension-for-pytorch.git
75+
$ cd intel-extension-for-pytorch
76+
$ git checkout v1.10.200+gpu
77+
```
78+
79+
### Install PyTorch:
80+
81+
```bash
82+
$ cd pytorch
83+
$ git apply ${intel_extension_for_pytorch_directory}/torch_patches/*.patch
84+
$ git submodule sync
5985
$ git submodule update --init --recursive
6086
$ pip install -r requirements.txt
61-
# configure MKL env to enable MKL features
6287
$ source ${oneAPI_HOME}/mkl/latest/env/vars.sh
63-
# build pypi package and install it locally
6488
$ python setup.py bdist_wheel
6589
$ pip install dist/*.whl
6690
```
6791

68-
Build and Install Intel® Extension for PyTorch*:
92+
### Configure the AOT (Optional)
93+
94+
Please refer to [AOT documentation](./AOT.md) for how to configure `USE_AOT_DEVLIST`.
95+
96+
```bash
97+
$ export USE_AOT_DEVLIST='dg2-g10-c0'
98+
```
99+
100+
### Install Intel® Extension for PyTorch\*:
69101

70102
```bash
71-
$ git clone -b xpu-master https://github.com/intel/intel-extension-for-pytorch.git
72103
$ cd intel-extension-for-pytorch
73-
# checkout to specific release branch if in need
74-
$ git checkout ${IPEX_RELEASE_BRANCH_NAME}
104+
$ git submodule sync
75105
$ git submodule update --init --recursive
76106
$ pip install -r requirements.txt
77-
# configure dpcpp compiler env
78107
$ source ${oneAPI_HOME}/compiler/latest/env/vars.sh
79-
# configure MKL env to enable MKL features
80108
$ source ${oneAPI_HOME}/mkl/latest/env/vars.sh
81-
# build pypi package and install it locally
82-
$ ${USE_AOT_DEVLIST} python setup.py bdist_wheel
109+
$ python setup.py bdist_wheel
83110
$ pip install dist/*.whl
84111
```

0 commit comments

Comments
 (0)