I think the appeal is with Jupyter [1] notebooks. Python is not about performance. Usually numpy (or other libraries) that does the heavy lifting on another lan

I think the appeal is with Jupyter [1] notebooks. Python is not about performance. Usually numpy (or other libraries) that does the heavy lifting on another language anyway. But having the Jupyter notebooks allows for intractability with the data. Make changes, and see how it affects every step after it. [1] https://jupyter.org/ https://jupyter.org/