Regd C1_W3_Assignment

Can anyone explain whats happening in this code:

X_pred_multi = np.array([[1710, 7], [1200, 6], [2200, 8]]).T

# Normalize using the same mean and standard deviation of the original training array X_multi.
X_multi_mean = np.array(np.mean(X_multi)).reshape((2,1))
X_multi_std = np.array(np.std(X_multi)).reshape((2,1))
X_pred_multi_norm = (X_pred_multi - X_multi_mean)/ X_multi_std
# Make predictions.
Y_pred_multi_norm = np.matmul(W_multi, X_pred_multi_norm) + b_multi
# Denormalize using the same mean and standard deviation of the original training array Y_multi.
Y_pred_multi = Y_pred_multi_norm * np.std(Y_multi) + np.mean(Y_multi)

print(f"Ground living area, square feet:\n{X_pred_multi[0]}")
print(f"Rates of the overall quality of material and finish, 1-10:\n{X_pred_multi[1]}")
print(f"Predictions of sales price, $:\n{np.round(Y_pred_multi)}")

Why did we take new array of X_pred_multi

Hi @Chaturya_Gajula!

Thanks for your post. I am not sure if I understood your question. Did you mean why does the normalization is performed? If this is the case, it is because such normalization is performed in training so it must be kept in evaluation too.

Lucas