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

data = pickle.load(open('users_vectors_clustering.p', 'rb'))
all_data = []
clusters = []

for c in data:
    for user in data[c]:
        all_data.append(user)
        clusters.append(c)

# pca = PCA(n_components=2)

# principalComponents = pca.fit_transform(users_dataset)
# principalDf = pd.DataFrame(data = principalComponents, columns = ['principal component 1', 'principal component 2'])

algorithm = TSNE(n_components=2)
components = algorithm.fit_transform(all_data)

# components = pca.fit_transform(users_dataset)
# print(components)

plt.scatter(components[:, 0], components[:, 1], alpha=0.8, s=0.15, c=clusters)

plt.savefig('results77.png')
print('saved picture')