Akash Chandra

Doubt in code

1

Can you explain in weights, what's the purpose of this code snippet `np.sqrt(1 / self.size[i-1])` [Line-112 in nn.py](https://github.com/ahmedbesbes/Neural-Network-from-scratch/blob/493be2f4015d345fc68d3addd518c2b127e8c648/nn.py#L112)

Triplet Loss

1

Just a query, have you tried triplet loss or lossless triplet loss as I think that would produce better embeddings as we are providing fewer examples and the clusters formed...

Added *.pdf,*.png types compatibilty.