Problem Statement:

Consider yourself to be Jeff, who is a Deep Learning Engineer at a prestigious company. Your company is working with a Cancer Institute to find out what are the factors which lead up to a patient having cancer.

Dataset Used:

$\backslash$table$[[$diagnosis $**,$radius$\_$mean $*{*}^{},$texture$\_$mean $**,$perimeter$\_$mean $*,$area$\_$mean $**,$smoothness$\_$mean $**,$compactness$\_$mean $*],[M,17\mathrm{.}990,10\mathrm{.}38,122\mathrm{.}80,1001\mathrm{.}0,0\mathrm{.}11840,0\mathrm{.}27760],[M,20\mathrm{.}570,17\mathrm{.}77,132\mathrm{.}90,1326\mathrm{.}0,0\mathrm{.}08474,0\mathrm{.}07864],[M,19\mathrm{.}690,21\mathrm{.}25,130\mathrm{.}00,1203\mathrm{.}0,0\mathrm{.}10960,0\mathrm{.}15990],[M,11\mathrm{.}420,20\mathrm{.}38,77\mathrm{.}58,386\mathrm{.}1,0\mathrm{.}14250,0\mathrm{.}28390],[M,20\mathrm{.}290,14\mathrm{.}34,135\mathrm{.}10,1297\mathrm{.}0,0\mathrm{.}10030,0\mathrm{.}13280],[M,12\mathrm{.}450,15\mathrm{.}70,82\mathrm{.}57,477\mathrm{.}1,0\mathrm{.}12780,0\mathrm{.}17000],[M,18\mathrm{.}250,19\mathrm{.}98,119\mathrm{.}60,1040\mathrm{.}0,0\mathrm{.}09463,0\mathrm{.}10900],[M,13\mathrm{.}710,20\mathrm{.}83,90\mathrm{.}20,577\mathrm{.}9,0\mathrm{.}11890,0\mathrm{.}16450],[M,13\mathrm{.}000,21\mathrm{.}82,87,50,519\mathrm{.}8,0\mathrm{.}12730,0\mathrm{.}19320]]$

Tasks to be Done:

A$.$ Start off by loading the 'breast$\_$cancer' dataset from 'sklearn'

a$.$ Print the number of samples and number of features in the data

b$.$ Divide the data into train & test sets with with test set size to be equal to $0\mathrm{.}33$

c$.$ Create the network:

i$.$ Start with the input layer

ii$.$ Add two hidden layers, where each layer has $32$ nodes

iii. The final layer's activation should be 'softmax'

iv$.$ Fit the model on the train set

v$.$ Evaluate the accuracy for train and test set