: import pandas as pd file_path='/Users//Downloads/Data 2/ExampleTrainDataset.csv' filepath2 = '/Users//Downloads/Data data = pd.read_csv(file_path) data2 = pd.read_csv(filepath2) 2/ExampleTestDataset.csv' print(data) print(data2) x1 x2 x3 Y 012345678 2 3.0 2 0 2 3.0 4 0 2 3 9.0 1 0 3 1 0.5 2 0 4 2.0 1 0 5 7 2.0 1 1 3 2.0 5 1 5 2.0 2 1 8 2 4.0 3 1 2 226 11234 9 3 2.0 0 x1 1 1 x2 x3 Y 2 6 3 7 4 3 32243 3 1 0 2 4 0 2 2 1 4 3 31 3 1 1 1

a):   Use the “ExampleTrainDataset.csv” Implement a multi-layer perceptron
with 1 input layer, 1 hidden layer with 4 perceptron (fully connected), 1 output layer
with 1 perceptron.
- Use squared error as the loss function.
- Implement batch-based learning (i.e, accumulate loss for all samples in
each iteration, and update weights once at the end)
- Use sigmoid activation function for all perceptron. Learning rate = 0.01
- Perform 50 iterations of learning.
- Plot a graph of training Loss with respect to iterations. Include this graph in
your report with the final weight vectors.
b): Using the trained MLP, perform predictions on “ExampleTestDataset.csv”
and report the test loss.