This is a collection various AI and python-based tools useful for data research, as discussed on my Cricce.com website. Some will be very recent packages, some will be older methodologies that are still useful, and in time may also be called as a function by an agent.
I'll use jupyter notebooks, provide reasonable explanations within the notebooks, and also what libraries need to be installed to work. You are encouraged to make new python environments for most of the tools.
I'll also include a basic directory structure useful when working with data. I'd encourage you to create a directory structure before you start, and be consistent using these within your notebooks. I've named the notebooks according to the packages.
First is Ydata-profiling (previously pandas profiling). With just a few lines of code, this gives an excellent first look at your data, describing the data overall and column by column, as well as heatmaps and correlations. This is presented both in line in the Jupyter notebook, and as a stand alone HTML file (clickable, and able to be shared). The GitHub repo is https://github.com/ydataai/ydata-profiling.