[bayes_nonconj] Remove pyro code from bayes_nonconj

[kp992]

Fixes #13

cc @mmcky

[netlify]

✅ Deploy Preview for timely-seahorse-68815c ready!

Name	Link
🔨 Latest commit	d6129a6
🔍 Latest deploy log	https://app.netlify.com/sites/timely-seahorse-68815c/deploys/65d313c88b93950008c597d5
😎 Deploy Preview	https://deploy-preview-15--timely-seahorse-68815c.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify site configuration.

[kp992]

@mmcky Can you please test it if there are no clashes between JAX/torch GPU? Many thanks.

[github-actions]

🚀 Deployed on https://65d31b943a6c78d92fc32a71--timely-seahorse-68815c.netlify.app

[mmcky]

@kp992 exceptional work on removing pyro.

It actually has simplified the code a lot.

I will ask @thomassargent30 to take a look at the preview https://65c7800b5289ee700011359b--timely-seahorse-68815c.netlify.app/bayes_nonconj

[mmcky]

@thomassargent30 -- when setting up this series of lectures, we found that this lecture bayes_nonconj used jax and pytorch which causes major issues when using the same software environment -- which is why even on https://python.quantecon.org/bayes_nonconj.html it was computing using a cpu. Essentially torch and jax seem incompatible due to conflicts in a mix of CUDA / GPU driver versions and therefore require different drivers which can't be done in the same software environment.

@kp992 has done a great job of removing torch and focusing this lecture just on jax which removes the dependency on torch.

Would you be able to review

https://65c7800b5289ee700011359b--timely-seahorse-68815c.netlify.app/bayes_nonconj

when you have time.

This should fix the last major technical issue in the reorg project.

[mmcky]

thanks @kp992 and @thomassargent30. Merging this now.