Estimating the thetas for Softmax Regression (Multinomial logistic regression) using
- Gradient Descent,
- Stochastic Gradient Descent
- Mini-batch Gradient Descent and
- Newton Raphson Method (using Quasi Methods) and Backtracking Linesearch
Notebook does not use Scikit learn's functions but I have coded all of them into the notebook. All references are also mentioned. Give it a try and report any bugs.