I am a doctor conducting survival analysis and need the lifelines package installed in the ChatGPT environment. This package is essential for my research, and having it available will greatly assist my work.
I’m not personally endorsing any of these solutions. They’re all open source and you can do your own research and experimentation to determine which (if any) best meet your needs.
One technique is to give a non-binary module as zipped code interpreter file in a ChatGPT GPT. Then have the AI unzip/untar with paths as the first step before an import.
As the code interpreter acts as a notebook, immediate follow-up calls will have the mount point code and the import available.
Dependencies are also a problem. In ChatGPT:
Failed to Import:
autograd: No module named ‘autograd’
formulaic: No module named ‘formulaic’
fonttools: No module named ‘fonttools’
interface_meta: No module named ‘interface_meta’
tzdata: No module named ‘tzdata’
wrapt: No module named ‘wrapt’
The determined person could probably get compatible code or wheels over there…