6. t-Distributed Stochastic Neighbor Embedding

Activity 6.01: Wine t-SNE

Solution:

1. Import pandas, numpy, and matplotlib, as well as the t-SNE and PCA models from scikit-learn:

   import pandas as pd
   import numpy as np
   import matplotlib.pyplot as plt
   from sklearn.decomposition import PCA
   from sklearn.manifold import TSNE

2. Load the Wine dataset using the wine.data file included in the accompanying source code and display the first five rows of data:

   df = pd.read_csv('wine.data', header=None)
   df.head()

   The output is as follows:

   Figure 6.25: The first five rows of the Wine dataset

3. The first column contains the labels; extract this column and remove it from the dataset:

   labels = df[0]
   del df[0]

4. Execute PCA to reduce the dataset to the first six components:

   model_pca = PCA(n_components=6)
   wine_pca = model_pca.fit_transform(df)

5. Determine the amount of variance within the data described by these six components:

   np.sum(model_pca.explained_variance_ratio_)

   The output...