monster/Qwen-Image-Layered-MRP-MLX
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge pull request #90 from anthonywu/py13-dev-working

Browse Source 
Issue #74 - python 3.13 pip install now working
This commit is contained in:
Filip Strand 2024-11-09 19:58:55 +01:00 committed by GitHub
commit f78144001c
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 29 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
README.md
View File

@ -55,6 +55,24 @@ uv tool install --upgrade mflux
to get the `mflux-generate` and related command line executables. You can skip to the usage guides below.
<details>
<summary>For Python 3.13 dev preview</summary>
The [T5 encoder](https://huggingface.co/docs/transformers/en/model_doc/t5) is dependent on [sentencepiece](https://pypi.org/project/sentencepiece/), which does not have a installable wheel artifact for Python 3.13 as of Nov 2024. Until Google [publishes a 3.13 wheel](https://pypi.org/project/sentencepiece/), you need to build your own wheel with [official build instructions](https://github.com/google/sentencepiece/blob/master/python/README.md#build-and-install-sentencepiece) or for your convenience use a `.whl` pre-built by contributor @anthonywu. The steps below should work for most developers though your system may vary.
```sh
uv venv --python 3.13
python -V # e.g. Python 3.13.0rc2
source .venv/bin/activate
# for your convenience, you can use the contributor wheel
uv pip install https://github.com/anthonywu/sentencepiece/releases/download/0.2.1-py13dev/sentencepiece-0.2.1-cp313-cp313-macosx_11_0_arm64.whl
# enable the pytorch nightly
uv pip install --pre --extra-index-url https://download.pytorch.org/whl/nightly -e .
```
</details>
<details>
<summary>For the classic way to create a user virtual environment:</summary>

15
pyproject.toml
View File

@ -14,14 +14,21 @@ license = { file = "LICENSE" }
requires-python = ">=3.10"
dependencies = [
"huggingface-hub>=0.24.5,<1.0",
"mlx>=0.16.0,<1.0",
"mlx>=0.20.0,<1.0",
"numpy>=2.0.1,<3.0",
"opencv-python>=4.10.0,<5.0",
"piexif>=1.1.3,<2.0",
"pillow>=10.4.0,<11.0",
"pillow>=10.4.0,<11.0; python_version<'3.13'",
"pillow>=11.0,<12.0; python_version>='3.13'",
"safetensors>=0.4.4,<1.0",
"sentencepiece>=0.2.0,<1.0",
"torch>=2.3.1,<3.0",
# python 3.13 workaround for now:
# use temporary community build of py13 wheel, use until official project build
# uv pip install https://github.com/anthonywu/sentencepiece/releases/download/0.2.1-py13dev/sentencepiece-0.2.1-cp313-cp313-macosx_11_0_arm64.whl
"sentencepiece>=0.2.0,<1.0; python_version<'3.13'",
"tokenizers>=0.20.3; python_version>='3.13'", # transformers -> tokenizers
"torch>=2.3.1,<3.0; python_version<'3.13'",
# torch dev builds: pip install --pre --index-url https://download.pytorch.org/whl/nightly
"torch>=2.6.0.dev20241106; python_version>='3.13'",
"tqdm>=4.66.5,<5.0",
"transformers>=4.44.0,<5.0",
]