Prateek Chanda

icon indicating an email [email protected]

icon company @microsoft icon location Bangalore, India icon location CS PhD IIT Bombay, Prev at Microsoft Research India, GSoC '19 mentor @JBossOutreach @campus-experts at @github intern @fossasia intern @linuxfoundation |

Results 70 comments of Prateek Chanda

Noise Adaptive Fuzzy Equalization method (NAFE)

is this to be implemented as part of the project along with FNRGF ?

Add visualizations

yeah sure @krishamehta

Add visualizations

sure you are welcome @vishal1383

Illustrations Visualization Add Python file, Not IPynb

I am reverting the changes @shreyas-kowshik of your last PR. Use that code in a python file, also not all the code but the part which is needed

Illustrations Visualization Add Python file, Not IPynb

I think so. Yeah .py for all, otherwise we cant test the code using the test cases @Lightning-Bug

why the example give very low score?

Hi @mashuai191 , as far as I remember the example notebook has been updated to accommodate a faster version. let me know if that works. it shows good accuracy for...

why the example give very low score?

Closing this in favour of https://github.com/ritesh99rakesh/pyMIDAS/blob/master/examples/midas.ipynb

MS-COCO training from Imagenet pretrained checkpoint

?

[WIP] Add a stablized function entropic_partial_wasserstein_logscale

I tested the current implementation, for some reason its extremely slow at the first iteration itself.

Running ViT on MS-COCO

Please share any updates here