File size: 3,914 Bytes
c0a3d33 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 |
from sklearn.manifold import TSNE
import matplotlib.pyplot as plt
import numpy as np
# Latent Feature Cluster for Training Data using T-SNE
def TSNE_reduction(latent_points: np.ndarray, perplexity=30, learning_rate=20):
"""
:param latent_points: [ndarray] - an array of arrays that define the points of multiple objects in the latent space
:param perplexity: [int] - default perplexity = 30 " Perplexity balances the attention t-SNE gives to local and global aspects of the data. It is roughly a guess of the number of close neighbors each point has... a denser dataset ... requires higher perplexity value" Recommended: Perplexity(5-50)
:param learning_rate: [int] - default learning rate = 200 "If the learning rate is too high, the data may look like a ‘ball’ with any point approximately equidistant from its nearest neighbours. If the learning rate is too low, most points may look compressed in a dense cloud with few outliers." Recommended: learning_rate(10-1000)
:return: [tuple] - the output is the x and y coordinates for the reduced latent space, a title, and a TSNE embedding
"""
model = TSNE(n_components=2, random_state=0, perplexity=perplexity,
learning_rate=learning_rate)
# the number of components = dimension of the embedded space
embedding = model
tsne_data = model.fit_transform(latent_points)
# When there are more data points, only use a couple of hundred points so TSNE doesn't take too long
x = tsne_data[:, 0]
y = tsne_data[:, 1]
title = ("T-SNE of Data")
return x, y, title, embedding
def plot_dimensionality_reduction(x: list, y: list, label_set: list, title: str):
"""
:param x: [list] - the first set of coordinates for each latent point
:param y: [list] - the second set of coordinates for each latent point
:param label_set: [list] - a set of values that define the color of each point based on an additional quantitative attribute.
:return: matplotlib figure - the output is a matplotlib figure that displays all the points in a 2-dimensional latent space, based on the labels provided.
"""
plt.title(title)
# Color points based on a continuous label
if label_set[0].dtype == float:
plt.scatter(x, y, c=label_set)
cbar = plt.colorbar()
cbar.set_label('Average Density', fontsize=12)
print("using scatter")
# Color points based on a discrete label
else:
for label in set(label_set):
cond = np.where(np.array(label_set) == str(label))
plt.plot(x[cond], y[cond], marker='o', linestyle='none', label=label)
plt.legend(numpoints=1)
plt.xlabel("Dimension 1")
plt.ylabel("Dimension 2")
########################################################################################################################
"""
# Use for personal plotting
import pandas as pd
import json
df = pd.read_csv('2D_Lattice.csv')
# row = 0
# box = df.iloc[row,1]
# array = np.array(json.loads(box))
# Select a subset of the data to use
number_samples = 10000
perplexity = 300
random_samples = sorted(np.random.randint(0,len(df), number_samples)) # Generates ordered samples
df = df.iloc[random_samples]
print(df)
print(np.shape(df))
# For plotting CSV data
# define a function to flatten a box
def flatten_box(box_str):
box = json.loads(box_str)
return np.array(box).flatten()
# apply the flatten_box function to each row of the dataframe and create a list of flattened arrays
flattened_arrays = df['Array'].apply(flatten_box).tolist()
avg_density = np.sum(flattened_arrays, axis=1)/(len(flattened_arrays[0]))
x, y, title, embedding = TSNE_reduction(flattened_arrays, perplexity=perplexity)
plot_dimensionality_reduction(x, y, avg_density, title)
plt.title(title)
plt.savefig('TSNE_Partial_Factorial_Perplexity_' + str(perplexity) + "_Data_Samples_" + str(number_samples))
"""
|