please solve this question from gradient descent topic in machine learning

The salary of a software developer depends on experience and skills. The salaries of $3$ software developers are provided in the following dataset.

Apply gradient descent to find the optimal values of the learning parameters of a linear machine learning model to predict the salary of any software developer.

Consider the cross entropy loss function for the assessment of the prediction quality, and learning rate as $0\mathrm{.}01,$ and show only one iteration.

Predict the salary of a highly skilled software developer with $4$ years of experience.

Here, Cross Entropy Loss $=$

${\displaystyle \underset{i=1}{\overset{N}{}}}-(y_{i}log(hat(y{)}_i)+(1-y_i)log(1-$hat$(y{)}_i))$

Encode the categorical variable values of Skills as follows:

High $=2,$ and Low $=1$t descent topic in machine learning