def pca_pipeline(matrix, num_of_pcs):
# PCA:
print("Performing PCA...")
matrix, eigvalues, eigvectors = pca.compute_pcs(matrix, num_of_pcs)
print("Projecting onto the PCs...")
scores_set = pca.scores(matrix, eigvectors)
print("Plotting by the PCs...")
pca.plot(scores_set, num_of_pcs, PIPELINE_NAMES[-1], PIPELINE_IDS[-1])
print("Elbow Method for choosing K clusters...")
X = elbow_method(scores_set, num_of_pcs, 1, 11)
num_of_clusters = int(input("Number of clusters? "))
print("K-Means clustering...")
labels = kmeans(X, num_of_clusters)
print("Performing a qualitative analysis...")
qualitative_analysis(labels, num_of_clusters)
return X
def test_pca():
######## TEST 1 #########
X = load_iris().data
labels = load_iris().feature_names
y = load_iris().target
param_grid = {
'n_components': [None, 0.01, 1, 0.95, 2, 100000000000],
'row_labels': [None, [], y],
'col_labels': [None, [], labels],
}
allNames = param_grid.keys()
combinations = it.product(*(param_grid[Name] for Name in allNames))
combinations = list(combinations)
for combination in combinations:
model = pca.fit(X,
n_components=combination[0],
row_labels=combination[1],
col_labels=combination[2])
ax = pca.plot(model)
# plt.close('all')
ax = pca.biplot(model)
# plt.close('all')
ax = pca.biplot3d(model)
# plt.close('all')
######## TEST 2 #########
X = sparse_random(100, 1000, density=0.01, format='csr', random_state=42)
model = pca.fit(X)
ax = pca.plot(model)
# plt.close('all')
ax = pca.biplot(model)
# plt.close('all')
ax = pca.biplot3d(model)
# plt.close('all')
######## TEST 3 #########
X = load_iris().data
labels = load_iris().feature_names
y = load_iris().target
model = pca.fit(X)
ax = pca.plot(model)
# plt.close('all')
ax = pca.biplot(model)
# plt.close('all')
ax = pca.biplot3d(model)
# plt.close('all')
model = pca.fit(X, row_labels=y, col_labels=labels)
fig = pca.biplot(model)
# plt.close('all')
fig = pca.biplot3d(model)
# plt.close('all')
model = pca.fit(X, n_components=0.95)
ax = pca.plot(model)
# plt.close('all')
ax = pca.biplot(model)
# plt.close('all')
model = pca.fit(X, n_components=2)
ax = pca.plot(model)
# plt.close('all')
ax = pca.biplot(model)
# plt.close('all')
Xnorm = pca.norm(X, pcexclude=[1, 2])
model = pca.fit(Xnorm, row_labels=y, col_labels=labels)
ax = pca.biplot(model)
# plt.close('all')
ax = pca.plot(model)
import matplotlib.pyplot as plt
import numpy as np
from mlxtend.data import loadlocal_mnist
from pca import PCA, plot, reconstruct
k = 235
images, labels = loadlocal_mnist(images_path='./Data/train-images-idx3-ubyte',
labels_path='./Data/train-labels-idx1-ubyte')
mat = images[labels == 0].astype('float64')
pca_mat, mean, k_evectors = PCA(k, mat, use_perc=False)
recons_mat = reconstruct(pca_mat, mean, k_evectors)
plot(mat[0], 'Original')
plot(recons_mat[0], 'Reconstructed')
plt.show()
import pca pca = pca.PCAPlot() pca.plot(["Mike", "Johanna"]) pca.plot()
'row_labels': [None, [], y],
'col_labels': [None, [], labels],
}
import itertools as it
allNames = param_grid.keys()
combinations = it.product(*(param_grid[Name] for Name in allNames))
combinations = list(combinations)
# %%
for combination in combinations:
model = pca.fit(X,
n_components=combination[0],
row_labels=combination[1],
col_labels=combination[2])
ax = pca.plot(model)
ax = pca.biplot(model)
ax = pca.biplot3d(model)
# %%
import pca
from scipy.sparse import random as sparse_random
X = sparse_random(100, 1000, density=0.01, format='csr', random_state=42)
model = pca.fit(X)
ax = pca.plot(model)
ax = pca.biplot(model)
ax = pca.biplot3d(model)
# %%
model = pca.fit(X)