Train Softmax Regression with Gradient Descent
Linear Models DS practice problem on Onlearn.
Difficulty: hard.
Topics: Understanding Train Softmax Regression with Gradient Descent, Cross-Entropy Loss, Softmax Activation Function, Weight Matrix Initialization, Learning Rate Scheduling, One-Hot Encoding, Optimization Theory, Linear Algebra, Statistical Learning, Calculus, Numerical Analysis, Gradient-Based Optimization, Multinomial Classification, Loss Function Formulation, Vectorized Matrix Operations, Iterative Convergence.
Implement a gradient descent based training algorithm for Softmax regression. Your task is to compute model parameters using Cross Entropy loss and return the optimized coefficients along with collected loss values over iterations. Make sure to round your solution to 4 decimal places