You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Firstly thanks to the good job of matcha,
But in my view of point, there are too many dependence to install the wheel package.
there are many Unnecessary third-party packages in requirements.txt, like gradio, torchvision, or other vision tools. that may help view the training process and inference test, but also lead to complicated package dependence and make it hard to install with simple pip in a pre-installed env.
My suggestion is a pure matcha-core installation requirement list, so that the core algorithm part can be installed easily, and can be used generally.
say sorry if the suggestion is not thoughtful to this project.
The text was updated successfully, but these errors were encountered:
That is a good suggestion, I will implement it. However, due to other commitments right now. I am having difficulty actively maintaining the package, but I accept PRs. Would love it if you or something could help with it :)
Firstly thanks to the good job of matcha,
But in my view of point, there are too many dependence to install the wheel package.
there are many Unnecessary third-party packages in requirements.txt, like gradio, torchvision, or other vision tools. that may help view the training process and inference test, but also lead to complicated package dependence and make it hard to install with simple pip in a pre-installed env.
My suggestion is a pure matcha-core installation requirement list, so that the core algorithm part can be installed easily, and can be used generally.
say sorry if the suggestion is not thoughtful to this project.
The text was updated successfully, but these errors were encountered: