# Data Science Final Write high level version that makes sure they can do everything Also check that they can get the data & it doesn't take forever 1 notebook: * 3 Problems: ### Problem 1 * Data Set: Sklearn Breast Cancer from sklearn.datasets import load_breast_cancer data = load_breast_cancer() * Compare dense neural network with random forest * Data input to 128 node dense layer followed Keras * 2 node classification layer ### Problem 2 https://www.tensorflow.org/overview ```python from tensorflow.examples.tutorials.mnist import input_data data = input_data.read_data_sets('data/fashion') ``` (a) Train on FashonMnist using the dense example in the tutorial (b) Compare the performance with 2 convolulation layers where we one max pool reduction by factor of 2 before 1-dense net (you pick number of nodes) before final dense 10 node net (ouput) (c) Show all activations for 1 example from each class as images (ouput of last conv layer) # To Do - Putting it in a notebook format & 1. check that data loads 2. try to solve it maybe 3. flag what specifically they have to produce