Challenge - Build a Neural Network and fine-tune it

Challenge - Build a Neural Network and try to fine-tune it

In this challenge, you need to build a simple forward neural network and try to fine-tune it. The main purpose of this challenge is to give you an intuitive sense of how different parameters affect network performance. So, you should do the following things:

• Create a dataset from make_classification.
• Split the data into 2 parts, train and test. The test set accounts for 20%. Use the 42 as the random seed.
• Build a neural network from MLPClassifier.
• Try different parameters, such as batch_size, hidden_layer_sizes, or learning_rate_init.
• Return the F1-score on the test set.

Note: In order to make sure that the results are comparable. Please use the following code to generate data.

X, y = datasets.make_classification(n_samples=1000,
                     n_features=30,
                     random_state=10)