Dr Kamran Soomro · @ksoomro
18 followers · 8 posts · Server sigmoid.social

So I’ve been wondering what problem to work on for my learning while I work my way through . I think I’ve found the perfect dataset: figshare.com/collections/_/456

It’s got everything. The raw 12-lead ECG readings as well as arrhythmia classifications for over 10,000 patients. And best of all it’s easy to convert the ECG readings to images for . Will probably start by trying out initially.

Will keep posting my progress here.

#practicaldeeplearning #deeplearning #ResNet

Last updated 1 year ago

Sometimes when creating shortcuts on people use a MaxPooling2D or even a convolution.

If you want to pass the identity of your input, you can use a Convolution (with no gradient) initialized with dirac_ (delta function A.K.A identity function for a convolution).

As you see is faster than using a MaxPooling:

#ResNet #neuralnetwork #pytorch

Last updated 2 years ago

Published papers at TMLR · @tmlrpub
488 followers · 168 posts · Server sigmoid.social

Linking Neural Collapse and L2 Normalization with Improved Out-of-Distribution Detection in Deep ...

Jarrod Haas, William Yolland, Bernhard T Rabus

openreview.net/forum?id=fjkN5U

#normalization #deep #ResNet

Last updated 2 years ago

Richard Strange · @StrangeThoughts
11 followers · 5 posts · Server sigmoid.social

To share more than just an introduction post, here's a fun bit of research that I was involved with during the 2019 FDL programme - climatechange.ai/papers/neurip -

#ResNet #CloudPhysics #FDL #MODIS #climatechange #iResNet #CloudClassification #GCP

Last updated 2 years ago