colpali-engine@0.3.16

The code used to train and run inference with the ColPali architecture.

2026-05-12 Python <3.15,>=3.10 wheel sdist

pip install colpali-engine==0.3.16

poetry add colpali-engine==0.3.16

pipenv install colpali-engine==0.3.16

conda install colpali-engine=0.3.16

依赖
`numpy`
`peft<0.20.0,>=0.18.0`
`pillow>=10.0.0`
`requests`
`scipy`
`torch<2.12.0,>=2.2.0`
`torchvision`
`transformers<6.0.0,>=5.3.0`
`accelerate<2.0.0,>=1.1.0; extra == "all"`
`bitsandbytes; extra == "all"`
`configue>=5.0.0; extra == "all"`
`datasets>=2.19.1; extra == "all"`
`einops<1.0.0,>=0.8.0; extra == "all"`
`late-interaction-kernels<0.5.0,>=0.4.1; extra == "all"`
`matplotlib<4.0.0,>=3.9.0; extra == "all"`
`mteb<3,>=1.16.3; extra == "all"`
`pillow<12.3.0,>=10.0.0; extra == "all"`
`pytest>=8.0.0; extra == "all"`
`ruff>=0.4.0; extra == "all"`
`seaborn<1.0.0,>=0.13.2; extra == "all"`
`typer>=0.15.1; extra == "all"`
`pytest>=8.0.0; extra == "dev"`
`ruff>=0.4.0; extra == "dev"`
`einops<1.0.0,>=0.8.0; extra == "interpretability"`
`matplotlib<4.0.0,>=3.9.0; extra == "interpretability"`
`seaborn<1.0.0,>=0.13.2; extra == "interpretability"`
`late-interaction-kernels<0.5.0,>=0.4.1; extra == "lik"`
`accelerate<2.0.0,>=1.1.0; extra == "train"`
`bitsandbytes; extra == "train"`
`configue>=5.0.0; extra == "train"`
`datasets>=2.19.1; extra == "train"`
`mteb<3,>=1.16.3; extra == "train"`
`pillow<12.3.0,>=10.0.0; extra == "train"`
`typer>=0.15.1; extra == "train"`

文件名	类型	Python 版本	大小
colpali_engine-0.3.16-py3-none-any.whl	bdist_wheel	`py3`	108858 bytes
colpali_engine-0.3.16.tar.gz	sdist	`source`	195672 bytes

0.3.17 2026-06-08

0.3.16 2026-05-12

0.3.15 2026-03-31

0.3.14 2026-02-24

0.3.13 2025-11-15

0.3.12 2025-07-16

0.3.11 2025-07-04

0.3.10 2025-04-18

0.3.9 2025-04-03

0.3.8 2025-01-29

0.3.7 2025-01-28

0.3.6 2025-01-10

0.3.5 2024-12-13

0.3.4 2024-11-07

0.3.3 2024-10-29

0.3.2 2024-10-17

0.3.1 2024-09-27

0.3.0 2024-09-10

0.2.2 2024-09-06

0.2.1 2024-09-02

0.2.0 2024-08-29

0.1.1 2024-08-29

0.1.0 2024-08-28