Vectorized version of gradient descent.

theta = theta * reg_param - alpha * (1 / num_examples) * (delta.T @ self.data).T

We should NOT regularize the parameter theta_zero.

theta[0] = theta[0] - alpha * (1 / num_examples) * (self.data[:, 0].T @ delta).T

the first code line ,theta include theta[0].
so I think can write like this:
theta[0] -= alpha * (1 / num_examples) * (self.data[:, 0].

Is your feature request related to a problem? Please describe.
Other related issues: #408 #251
I trained a Chinese model for spaCy, linked it to [spacy's package folder]/data/zh (using spacy link) and want to use that for ludwig. However, when I tried to set the config for ludwig, I received an error, which tell me that there is no way to load the Chinese model.

ValueError: Key ch