Typos

[dsantiago]

Hi Hadrien, as we talked on LinkedIn i am sending some typos from the blog. Minimal stuff but i am sure will help it.

../posts/Deep-Learning-Book-Series-2.5-Norms/

• The Latex below did not work on the page:

$ \norm{\bs{x}}1=\sum{i} |\bs{x}_i| $

../posts/Deep-Learning-Book-Series-2.7-Eigendecomposition/

• Check the text below, add the "be":

It works! For that reason, it can BE useful to use symmetric matrices! Let’s do it now with linalg from Numpy...

../posts/Deep-Learning-Book-Series-2.12-Example-Principal-Components-Analysis/

• The formula below appears to need one more parenthesis(the middle one).

r(x) = g(f(x) = DDTx

../posts/Preprocessing-for-deep-learning/

• The Latex formula in the "Mean subtraction: per-pixel or per-image" section shows the "diag" function used 2 times, but in python it is only used once:

X_ZCA = U.dot(np.diag(1.0/np.sqrt(S + epsilon))).dot(U.T).dot(X_norm.T).T