flyingscorpio/efrei
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

Create better dataset

Browse source
This commit is contained in:
flyingscorpio@clevo 2023-01-25 16:54:07 +01:00
parent 25f1991a25
commit 8dc3637e2f
1 changed files with 20 additions and 8 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

28
big-data/tp1/main.py
View file

@ -2,6 +2,7 @@
"""TP1 - Perceptron"""
import random
import matplotlib.pyplot as plt
import numpy as np
@ -38,20 +39,31 @@ def main():
plt.show()
def generate_dataset(
nb_data_points: int
) -> tuple[np.ndarray[float, float], np.ndarray[int]]:
def blur_point(point):
x, y = point
x = 0.3 * (random.randint(0, 2000) / 1000 - 1) * x + x
y = 0.3 * (random.randint(0, 2000) / 1000 - 1) * x + y
return [x, y]
def generate_dataset(nb_data_points: int):
"""Create a dataset made of two seperate groups of points
Return (data, expected), where expected is the expected group for each group.
"""
func1 = lambda x: 2 * x + 1
func0 = lambda x: 0.5 * x - 3
# Création groupe 1
data1 = np.random.rand(nb_data_points // 2, 2) + 3
x1 = np.linspace(0, 8, nb_data_points)
data1 = np.array([blur_point([x, func1(x)]) for x in x1])
expected1 = np.array([1 for _ in data1])
# Création groupe 0
data0 = np.random.rand(nb_data_points // 2, 2) + 1
x0 = np.linspace(0, 8, nb_data_points)
data0 = np.array([blur_point([x, func0(x)]) for x in x0])
expected0 = np.array([0 for _ in data0])
# Concaténation des deux groupes
@ -59,9 +71,9 @@ def generate_dataset(
expected = np.concatenate((expected1, expected0))
# Mélange des données
permutation = np.random.permutation(nb_data_points)
data = data[permutation]
expected = expected[permutation]
p = np.random.permutation(len(data))
data = data[p]
expected = expected[p]
return data, expected