Factorized code for dC in makemore_part4_backprop.ipynb

[bzkhalil]

First thanks a lot Andreij for the great series,
Here is the factorized version for dC
one_hot_Xb = F.one_hot(Xb,27).view((Xb.shape[0]*Xb.shape[1],27)).float()
dC = one_hot_Xb.T @ demb.view((demb.shape[0]*demb.shape[1],demb.shape[2]))

I came up with it mainly through shape comparaison, could you please look into it :-)

[marcdumon]

Seems correct to me.
I came up with solution that is slightly slower but more clear:

Xb_1hot = F.one_hot(Xb, 27).float()  # 32 x 3 x 27   
d_C = torch.einsum('ijk,ijl->kl', Xb_1hot, d_emb))  # 27 x 10