ML-From-Scratch icon indicating copy to clipboard operation
ML-From-Scratch copied to clipboard

Published 20 hours ago • verified publisher icon eriklindernoren

Missing number of neurons in grad_w for dense layer

Open Superhzf opened this issue 5 years ago • 0 comments

https://github.com/eriklindernoren/ML-From-Scratch/blob/a2806c6732eee8d27762edd6d864e0c179d8e9e8/mlfromscratch/deep_learning/layers.py#L82

Please correct me if I'm wrong, I think for a dense layer, grad_w = self.layer_input.T.dot(accum_grad)/the_number_of_neurons

Superhzf avatar Jan 22 '20 19:01 Superhzf