def graph():
dp = DataPreProcessing(training_planets=1000)
dp.create_data()
X, Y = dp.get_graphing_data()
x = [planet[0] for planet in X]
y = [planet[1] for planet in X]
z = [planet[2] for planet in X]
#fig = plt.figure()
#ax = Axes3D(fig)
for i in range(len(Y)):
if Y[i] == 1:
plt.scatter(x[i], y[i], color='red')
else:
plt.scatter(x[i], y[i], color='blue')
plt.show()
def load_data(self):
# 570 is the sweet spot for number of training planets gives best results
dp = DataPreProcessing(training_planets=570)
dp.create_data()
self.X_train, self.Y_train = dp.get_normalized_training_data()
self.X_test, self.Y_test = dp.get_normalized_testing_data()
def load_data(self):
dp = DataPreProcessing(training_planets=150)
dp.create_data()
self.X_train, self.Y_train = dp.get_normalized_training_data()
self.X_test, self.Y_test = dp.get_normalized_testing_data()
def load_data(self):
dp = DataPreProcessing(training_planets=5087)
dp.create_data()
self.X_train, self.Y_train = dp.get_scaled_standardized_training_data()
self.X_test, self.Y_test = dp.get_scaled_standardized_testing_data()
def load_data(self):
# 2500 is the sweet spot for number of training planets gives best results
dp = DataPreProcessing(training_planets=5087)
dp.create_data()
self.X_train, self.Y_train, self.X_test, self.Y_test = dp.get_data()
from exoplanet_knn import NearestNeighbor
from exoplanet_nn import NeuralNetwork
from exoplanet_tree import DecisionTree
from exoplanet_self_knn import KNearestNeighbor
from data_preprocessing import DataPreProcessing
import matplotlib.pyplot as plt
if __name__ == "__main__":
dp = DataPreProcessing(training_planets=5087)
print("creating models...")
dp.create_data()
data = dp.X_test
nn = NeuralNetwork()
knn = NearestNeighbor()
tree = DecisionTree()
self_knn = KNearestNeighbor(k=7)
nn.load_data()
nn.load_nn()
knn.load_data()
knn.load_knn()
tree.load_data()
tree.load_tree()
self_knn.load_data()
knn.predict()
axs[0].legend(['train', 'val'], loc='best')
# summarize history for loss
axs[1].plot(range(1, len(model_info.history['loss']) + 1), model_info.history['loss'])
axs[1].plot(range(1, len(model_info.history['val_loss']) + 1), model_info.history['val_loss'])
axs[1].set_title('model_info Loss')
axs[1].set_ylabel('Loss')
axs[1].set_xlabel('Epoch')
axs[1].set_xticks(np.arange(1, len(model_info.history['loss']) + 1), len(model_info.history['loss']) / 10)
axs[1].legend(['train', 'val'], loc='best')
plt.show()
if __name__ == '__main__':
dpp = DataPreProcessing()
(x_train, x_test), (y_train, y_test) = dpp.load_data(img_data_path='./data/img_data.npy',
label_data_path='./data/label_data.npy')
(x_train, x_test), (y_train, y_test) = dpp.pre_processing(x_train, x_test, y_train, y_test)
label_list, age_list, gender_list = dpp.get_labels()
print(x_train.shape)
print(x_test.shape)
print(y_train.shape)
print(y_test.shape)
faceNet = FaceNet(input_shape=x_train.shape[1:],
num_classes=len(label_list),
gpu=True)
model = faceNet.build()
model_info = model.fit(x_train, y_train,
batch_size=batch_size,