Question

b. (15 points) Complete the Python function below that takes a radial-basis kernel matrix ?, the labels y, and a regularization parameter ? > 0 as input and learns weights via ridge regression. Specifically, given a radial-basis kernel matrix ?, implement the computation of w = (??? + ? I_n)?¹??y. # Phi float(n, d): transformed data # y float(n, ): Labels # Lam float : regularization parameter def train_ridge_model(Phi, y, lam): # # # * Insert your code here * # #

          b. (15 points)
Complete the Python function below that takes a radial-basis kernel matrix ?, the labels y, and a regularization parameter ? > 0 as input and learns weights via ridge regression. Specifically, given a radial-basis kernel matrix ?, implement the computation of w = (??? + ? I_n)?¹??y.

# Phi float(n, d): transformed data
# y float(n, ): Labels
# Lam float : regularization parameter
def train_ridge_model(Phi, y, lam):
#
#
# *** Insert your code here ***
#
#

b. (15 points)
Complete the Python function below that takes a radial-basis kernel matrix ?, the labels y, and a regularization parameter ? > 0 as input and learns weights via ridge regression. Specifically, given a radial-basis kernel matrix ?, implement the computation of w = (??? + ? In)?¹??y.

# Phi float(n, d): transformed data
# y float(n, ): Labels
# Lam float : regularization parameter
def trainridgemodel(Phi, y, lam):
#
#
# *** Insert your code here ***
#
#

Added by Amber G.

Computer Science and Information Technology

Trishna Knowledge Systems 2018 Edition

Instant Answer

Solved by Expert Akash M

09/10/2023

Step 1

We need to compute the matrix ΦΦ^T. We can do this using the numpy dot product function. ```python PhiPhiT = np.dot(Phi, Phi.T) ``` Show more…

Show all steps

Thanks for your feedback!

Complete the Python function below that takes a radial-basis kernel matrix Γ, the labels y, and a regularization parameter λ > 0 as input and learns weights via ridge regression. Specifically, given a radial-basis kernel matrix Γ, implement the computation of w = (Γ^TΓ + αI_n)^-1Γ^Ty. # Phi: float(n, d) - transformed data # y: float(n, ) - labels # lam: float - regularization parameter def train_ridge_model(Phi, y, lam): # *** Insert your code here ***