simply_train.py

import numpy as np
import tensorflow as tf
sess = tf.Session()
import os


import keras
from keras.models import Sequential
from keras.layers import Dense, Dropout, Flatten, Activation
from keras.layers import Conv2D, MaxPooling2D
from keras import backend as K
from keras import callbacks
from keras.models import load_model
K.set_session(sess)

width = 160	
length = 120

file_name1 = 'data/training/track1/xtrain_data.npy'
x_train = np.load(file_name1)

file_name2 = 'data/training/track1/xtest_data.npy'
x_test = np.load(file_name2)

file_name1 = 'data/training/track1/ytrain_data.npy'
y_train = np.load(file_name1)

file_name2 = 'data/training/track1/ytest_data.npy'
y_test = np.load(file_name2)
print(np.shape(y_train))

#x_train = x_train[:200]
#y_train = y_train[:200]
x_train = x_train.reshape(x_train.shape[0],width ,length, 3)
x_test = x_test.reshape(x_test.shape[0], width, length, 3)

x_train = x_train.astype('float32')
x_test = x_test.astype('float32')

x_train /= 255
x_test /= 255


input_shape = (width,length,3)
num_classes = 9
batch_size = 50
epochs = 20
save_dir = os.path.join(os.getcwd(), 'saved_models')
model_name = 'colour_160_120_nvidia.h5'


filepath2='saved_models/colour_160_120_nvidia_best.h5'
checkpoint = keras.callbacks.ModelCheckpoint(filepath2, monitor='val_acc', verbose=1, save_best_only=True, mode='max')

model = load_model('saved_models/colour_160_120_nvidia_best.h5')
print("Loaded model")

model.fit(x_train, y_train,
          batch_size=batch_size,
          epochs=epochs,
          verbose=1,
          validation_data=(x_test, y_test),callbacks = [checkpoint])


score = model.evaluate(x_test, y_test, verbose=0)
print('Test loss:', score[0])
print('Test accuracy:', score[1])

if not os.path.isdir(save_dir):
    os.makedirs(save_dir)
model_path = os.path.join(save_dir, model_name)
model.save(model_path)
print('Saved trained model at %s ' % model_path)