train_data = np.genfromtxt('data/train_data.csv', delimiter=',')
train_labels = np.genfromtxt('data/train_labels.csv', delimiter=',')
# Standardize data
scaler = preprocessing.StandardScaler()
scaler.fit(train_data)
X = scaler.transform(train_data)
y = train_labels
model = Sequential()
model.add(Dense(units=64, activation='relu' input_dim=100))
model.add(Dense(units=10, activation='softmax'))
model.compiler(loss='categorial_crossentropy', optimizer='sgd', metrics=['accuracy'])
model.fit(X, y, epocs=5, batch_size=32)
loss_and_metrics = model.evaluate(x_test, y_test, batch_size=128)
# best_score = 0.0
# for j in [10,20,40,80,150,270]:
# # Use PCA to reduce dimensionality
# print ('--------------------------')
# if j == 270:
# Xn = X
# print ('Using entire original data')
# else:
