target_names = ['class-0', 'class-1']
path = os.getcwd()
result_path_neurochaos = path + '/NEUROCHAOS-RESULTS/' + classification_type + '/' + folder_name + '-neurochaos/'
# Creating Folder to save the results
try:
os.makedirs(result_path_neurochaos)
except OSError:
print("Creation of the result directory %s failed" %
result_path_neurochaos)
else:
print("Successfully created the result directory %s" %
result_path_neurochaos)
full_artificial_data, full_artificial_label, full_artificial_test_data, full_artificial_test_label = get_data(
classification_type)
num_classes = len(np.unique(full_artificial_label)) # Number of classes
print("**** Genome data details ******")
for class_label in range(np.max(full_artificial_label) + 1):
print("Total Data instance in Class -", class_label, " = ",
full_artificial_label.tolist().count([class_label]))
print(" train data = ", (full_artificial_data.shape[0]))
print("val data = ", (full_artificial_test_data.shape[0]))
print("initial neural activity = ", initial_neural_activity,
"discrimination threshold = ", discrimination_threshold,
"epsilon = ", epsilon)
folder_name = "svm_occd-train_ccd-test"
target_names = ['class-0', 'class-1']
path = os.getcwd()
result_path_svm_rbf = path + '/NEUROCHAOS-RESULTS/' + folder_name + '/'
# Creating Folder to save the results
try:
os.makedirs(result_path_svm_rbf)
except OSError:
print("Creation of the result directory %s failed" % result_path_svm_rbf)
else:
print("Successfully created the result directory %s" % result_path_svm_rbf)
## TEST DATA
ccd_train_data, ccd_train_label, ccd_test_data, ccd_test_label = get_data(
classification_type_test)
## TRAIN DATA
occd_train_data, occd_train_label, occd_test_data, occd_test_label = get_data(
classification_type_train)
num_classes = len(np.unique(ccd_train_label)) # Number of classes
print("**** Sythetic data data details ******")
for class_label in range(np.max(ccd_train_label) + 1):
print("Total Data instance in Class -", class_label, " = ",
ccd_train_label.tolist().count([class_label]))
print(" OCCD train data = ", (occd_train_data.shape[0]))
print("CCD validation data = ", (ccd_train_data.shape[0]))
# Start of svm_rbf classifier
@author: harik
"""
import numpy as np
import matplotlib.pyplot as plt
from sklearn.model_selection import KFold
from sklearn.metrics import confusion_matrix as cm
import os
from sklearn.metrics import (precision_score, recall_score, f1_score,
accuracy_score, mean_squared_error,
mean_absolute_error)
from sklearn.svm import LinearSVC
from load_data_synthetic import get_data
import ChaosFEX.feature_extractor as CFX
DATA_NAME = "concentric_circle_noise"
TRAINDATA, TRAINLABEL, X_TEST, Y_TEST = get_data(DATA_NAME)
INITIAL_NEURAL_ACTIVITY = [0.22]
DISCRIMINATION_THRESHOLD = [0.96]
EPSILON = np.arange(0.01, 0.201, 0.001)
ACCURACY = np.zeros((len(DISCRIMINATION_THRESHOLD),
len(INITIAL_NEURAL_ACTIVITY), len(EPSILON)))
FSCORE = np.zeros((len(DISCRIMINATION_THRESHOLD), len(INITIAL_NEURAL_ACTIVITY),
len(EPSILON)))
Q = np.zeros((len(DISCRIMINATION_THRESHOLD), len(INITIAL_NEURAL_ACTIVITY),
len(EPSILON)))
B = np.zeros((len(DISCRIMINATION_THRESHOLD), len(INITIAL_NEURAL_ACTIVITY),
len(EPSILON)))
EPS = np.zeros((len(DISCRIMINATION_THRESHOLD), len(INITIAL_NEURAL_ACTIVITY),
len(EPSILON)))
