I would go through and try to minimize your dependencies down to the very minimal list. It looks like you have a lot of things that are probably irrelevant to your streamlit app, like jupyter, and everything is pinned to specific versions. See if you can go through and see what packages your app actually imports, and just list those in your requirements.txt. Also, unless you know that you need a specific version of a package, just leave the package name by itself, without any == – this will allow the solver to try and find a set of packages which are compatible.
Also, you shouldn’t have python in requirements.txt
This often happens when you try to install an old version of some package with a recent version of python. Unfortunately your screenshot doesn’t include some relevant parts of the logs (that’s why screenshots are not a good way of posting textual information), so we don’t know what is the package causing the issue.
I would still try to trim that down a lot more, and remove the specific version numbers as much as possible.
Since you’re just using a streamlit app, you can’t actually use Flask or Werkzeug, so I would drop those. I would also remove numpy, requests and pandas, since they are included streamlit streamlit.
I would also remove all the “If needed” packages, and add those back later if you do end up needing them.
If you know that you need a specific version of one or more of those, you could try specifying that version, but it will be easier for the solver if you only pin (meaning ==specific.version.number) a minimal number of dependencies.
