ayo/ipex-llm
1
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

Deprecate support for pytorch 2.0 on Linux for ipex-llm >= 2.1.0b20240511 (#10986)

Browse source 
* Remove xpu_2.0 option in setup.py

* Disable xpu_2.0 test in UT and nightly

* Update docs for deprecated pytorch 2.0

* Small doc update
This commit is contained in:
Yuwen Hu 2024-05-11 12:33:35 +08:00 committed by GitHub
parent 5e0872073e
commit 9f6358e4c2
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 15 additions and 20 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

3
.github/workflows/llm_unit_tests.yml vendored
View file

@ -239,7 +239,8 @@ jobs:
strategy:
fail-fast: false
matrix:
pytorch-version: ['2.1', '2.0']
# pytorch-version: ['2.1', '2.0']
pytorch-version: ['2.1']
python-version: ${{ fromJson(needs.setup-python-version.outputs.python-version) }}
runs-on: [self-hosted, llm, arc-ut]
env:

20
docs/readthedocs/source/doc/LLM/Overview/install_gpu.md
View file

@ -187,6 +187,10 @@ IPEX-LLM GPU support on Linux has been verified on:
.. important::
IPEX-LLM on Linux supports PyTorch 2.0 and PyTorch 2.1.
.. warning::
IPEX-LLM support for Pytorch 2.0 is deprecated as of ``ipex-llm >= 2.1.0b20240511``.
```
```eval_rst
@ -311,7 +315,7 @@ IPEX-LLM GPU support on Linux has been verified on:
cd /opt/intel/oneapi/installer
sudo ./installer
.. tab:: PyTorch 2.0
.. tab:: PyTorch 2.0 (deprecated for versions ``ipex-llm >= 2.1.0b20240511``)
To enable IPEX-LLM for Intel GPUs with PyTorch 2.0, here're several prerequisite steps for tools installation and environment preparation:
@ -432,7 +436,7 @@ We recommend using [miniconda](https://docs.conda.io/en/latest/miniconda.html) t
```eval_rst
.. tabs::
.. tab:: PyTorch 2.1
Choose either US or CN website for `extra-index-url`:
Choose either US or CN website for ``extra-index-url``:
.. tabs::
.. tab:: US
@ -470,8 +474,8 @@ We recommend using [miniconda](https://docs.conda.io/en/latest/miniconda.html) t
pip install --pre --upgrade ipex-llm[xpu_2.1] --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/cn/
.. tab:: PyTorch 2.0
Choose either US or CN website for `extra-index-url`:
.. tab:: PyTorch 2.0 (deprecated for versions ``ipex-llm >= 2.1.0b20240511``)
Choose either US or CN website for ``extra-index-url``:
.. tabs::
.. tab:: US
@ -481,7 +485,7 @@ We recommend using [miniconda](https://docs.conda.io/en/latest/miniconda.html) t
conda create -n llm python=3.11
conda activate llm
pip install --pre --upgrade ipex-llm[xpu_2.0] --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/
pip install --pre --upgrade ipex-llm[xpu_2.0]==2.1.0b20240510 --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/
.. tab:: CN
@ -490,7 +494,7 @@ We recommend using [miniconda](https://docs.conda.io/en/latest/miniconda.html) t
conda create -n llm python=3.11
conda activate llm
pip install --pre --upgrade ipex-llm[xpu_2.0] --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/cn/
pip install --pre --upgrade ipex-llm[xpu_2.0]==2.1.0b20240510 --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/cn/
```
@ -521,7 +525,7 @@ If you encounter network issues when installing IPEX, you can also install IPEX-
# install ipex-llm for Intel GPU
pip install --pre --upgrade ipex-llm[xpu]
.. tab:: PyTorch 2.0
.. tab:: PyTorch 2.0 (deprecated for versions ``ipex-llm >= 2.1.0b20240511``)
.. code-block:: bash
@ -540,7 +544,7 @@ If you encounter network issues when installing IPEX, you can also install IPEX-
pip install intel_extension_for_pytorch-2.0.110+xpu-cp311-cp311-linux_x86_64.whl
# install ipex-llm for Intel GPU
pip install --pre --upgrade ipex-llm[xpu_2.0]
pip install --pre --upgrade ipex-llm[xpu_2.0]==2.1.0b20240510
```

10
python/llm/setup.py
View file

@ -285,15 +285,6 @@ def setup_package():
# Linux install with --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/
xpu_20_requires = copy.deepcopy(all_requires)
xpu_20_requires.remove('torch')
# xpu_20 only works for linux now
xpu_20_requires += ["torch==2.0.1a0;platform_system=='Linux'",
"torchvision==0.15.2a0;platform_system=='Linux'",
"intel_extension_for_pytorch==2.0.110+xpu;platform_system=='Linux'",
"bigdl-core-xe==" + CORE_XE_VERSION + ";platform_system=='Linux'",
"bigdl-core-xe-esimd==" + CORE_XE_VERSION + ";platform_system=='Linux'"]
xpu_21_requires = copy.deepcopy(all_requires)
xpu_21_requires.remove('torch')
xpu_21_requires += ["torch==2.1.0a0",
@ -330,7 +321,6 @@ def setup_package():
},
extras_require={"all": all_requires,
"xpu": xpu_requires, # default to ipex 2.1 for linux and windows
"xpu-2-0": xpu_20_requires,
"xpu-2-1": xpu_21_requires,
"serving": serving_requires,
"cpp": ["bigdl-core-cpp==" + CORE_XE_VERSION],