How to Navigate through APIs via VC CODE

In the older version of fastai there was a environment.yml file that we would open in VS CODE and then I would be able to use it to navigate quickly through the APIs. I wonder is there any equivalent of that for Fastai 1.0.49?
in another word what is the step by step procedure to have all the fastai v1 APIs accessible via VS CODE?

P.S. I have installed latest version of fastai on Windows 10 using:
conda install -c pytorch -c fastai fastai

With VC CODE you mean VS Code (at least I found nothing when looking for VC Code) ? Just install the standard python extension and hit Ctrl+T, then type the function you’re looking for, it will navigate you whereever you like.
The extension will automatically index the library whereever it lies (probably site-packages) and thus let you navigate it. All you need for vs code to find it is set the python interpreter to you conda environment (you can do that in the status bar at the bottom once you have loaded some kind of python file).
I don’t see how the environment.yml would help with any of this.

Did I understand your question correctly?

Edit: Since your running on windows, you will also have to make sure that your conda and python executables are either in your path, or set in the Python extension settings. On Windows the extension usually does not find those automatically. This page explains pretty much everything you need to set it up

1 Like

Hi @xeTaiz
thanks for your reply.
If you look at this video https://www.youtube.com/watch?v=Z0ssNAbe81M&t=3840s
@ 1:03:50 - 1:14:00 Jeremy is talking about navigating through fastai source code. while this was working perfectly for fastai 0.7, I am not able to get it work on Fastai 1.
I have python and anaconda extension installed and after selecting conda interpreter, I cannot get any source code I want. for example if I search for cnn_learner i get this:

Try right click your vs code file explorer, add folder to workspace and add the fastai library directly in a second folder.

2 Likes

Thank you @xeTaiz it is working now!!