$\{$

"cells": $[$

$\{$

"cell$\_$type": "markdown",

"metadata": $\{\},$

"source": $[$

$"\frac{?}{?}$

$,$

$"$

$",$

functions, and so on$)$ and numpy arrays. $ln,$

$"",$

$"$It is forbidden to use any ML libraries. In$"$

$]$

$\},$

$\{$

"cell$\_$type": "markdown",

"metadata": $\{\},$

"source": $[$

$"",$

$"$

$",$

$"*$ Implement polynomial features $[10$p$]$

$",$

$"*$ Implement minmax normalisation $[10$p$]$

$",$

$"*$ Implement logistic regression loss function and gradient descent algorithm $[10$p$]$

$",$

$"*$ Plot learning and test curve $[10$p$]$

$",$

$"*$ Generate test prediction $($test dataset$)[10$p$]\frac{?}{?}$

$,$

$"*$ Fill the confusion matrix $($test dataset$)[10$p$]$

$",$

$"*$ Compute F$1$ score, precision, recall and accuracy using confusion matrix $($test dataset$)[10$p$]$

$",$

$"*$ Try various learning rate and compute F$1$ score of each learning rate $($test dataset$)[10$p$]$

$",$

$"*$ Learning rates: $(0\mathrm{.}0001,0\mathrm{.}0005,0\mathrm{.}0007,0\mathrm{.}001,0\mathrm{.}005,0\mathrm{.}001,0\mathrm{.}01,0\mathrm{.}05,0\mathrm{.}09,0\mathrm{.}1,0\mathrm{.}4,0\mathrm{.}7)\frac{10p}{?}$

$,$

$"*$ Generate the lists for each learning rate and its associated F$1$ score and plot $($test dataset$)[10$p$]"$

$]$

$\},$

$\{$

"cell$\_$type": "code",

"execution$\_$count": $35,$

"metadata": $\{\},$