Hi ExasperatedCrocodile76 , I think this is the quickest and easiest solution you're looking for:
https://clear.ml/docs/latest/docs/references/sdk/task/#taskforce_requirements_env_freeze
Just make sure to call it before Task.init()
as documentation states
Hello everyone ! I would like to form my problem as best as I can. I have my project (for this project I used pip to install every dependency), but i also had virtualenv created via conda. For my config binary path I set it to the virtual environment. For the dependencies I generated requirements.txt via pip freeze. Right now I am able to run my task locally and track everything. When I try to make a clone of the task the problem comes. I cant enqueue it (default). I always have problem with the installing dependencies. No matter what I try (detectron2) can not be installed straight from the extra-url / when i do something like detectron2 @ url of the wheel it also can not find the version. Then I have problem with torch -cluster also. Is there any way that i can use whole environment in my computer and do not try to replicate everything ? Maybe that will be the easiest way. If anyone is open to chat I can provide all the materials.