Karm Patel

icon indicating a website link https://karm-patel.github.io/

icon company Indian Institute of Science (IISc) icon location Bangalore, India icon location IISc Bangalore | Former GSoC'22 @TensorFlow

Results 5 issues of Karm Patel

Thin line-weight/Poor figures - Adapted from other authors

1 comment

Fig 4.33: ![image](https://user-images.githubusercontent.com/59387624/180839565-4fdc0de2-d333-4bc4-86c6-d104684f7b83.png) 9.11 ![image](https://user-images.githubusercontent.com/59387624/180839654-d32767c9-7b00-43e1-ba92-530d63bcbfc5.png) 10.17(a) ![image](https://user-images.githubusercontent.com/59387624/180839910-18031fb6-d737-46ad-b661-aebed5b245ab.png) 11.6 ![image](https://user-images.githubusercontent.com/59387624/180839994-a212658b-a240-409d-8e4f-ac69796ba30c.png) 12.7 ![image](https://user-images.githubusercontent.com/59387624/180840159-06dc1f95-58de-4c4e-84cd-1fa17309b1a4.png) 12.22 ![image](https://user-images.githubusercontent.com/59387624/180840099-3b8b26d4-9311-4d94-80e9-4813b816e9af.png) 18.28 ![image](https://user-images.githubusercontent.com/59387624/180840244-ad1f099b-05b1-4647-b230-13b2a4706f72.png) 30.3 ![image](https://user-images.githubusercontent.com/59387624/180840332-f05beb17-d537-4f6c-9cd7-0034762cacf1.png) 31.7 ![image](https://user-images.githubusercontent.com/59387624/180840430-2a6c5bff-b1d0-4434-913d-1edee483245a.png) 35.4 ![image](https://user-images.githubusercontent.com/59387624/180840633-9b40c022-9f05-46a1-90d1-3f3840f35fbd.png)

Priority 1

Make bnn_mnist_sgld_blackjax.ipynb to be able to run in colab

- currently, this notebook won't run in colab, due to requirement of high memory. - Mostly prediction loops taking much memory, we can make it batch prediction to consume less...

Undo removing prefix and add prefix like _blackjax, _pymc3, etc.

Book2: Convert pymc3 demos to pymc4

There are 4 notebooks in book2 using pymc3

Add arviz plots for MCMC

10 comment

Hello Team, I'm working with Dr. @murphyk, Prof. @nipunbatra. and lab colleague @patel-zeel. We noticed that to visualize trace plots using `states`, we need to write manual code like below,...