def model_pred(model_name, test, testTemporal, topo_data):
# test generator
test_gene = test_generator(test, testTemporal, topo_data, BATCHSIZE, TIMESTEP)
test_sep = (test.shape[0] - TIMESTEP) * test.shape[1] // BATCHSIZE
# get predict
model = get_model_structure(model_name)
# model = multi_gpu_model(model, gpus=2) # gpu parallel
model.compile(loss=LOSS, optimizer=OPTIMIZER)
model.load_weights(PATH + '/' + MODELNAME + '.h5')
predY = model.predict_generator(test_gene, steps=test_sep)
# ground truth
testY = get_test_true(test, TIMESTEP, model_name)
# compute mse
scaled_testY = np.reshape(testY, ((test.shape[0] - TIMESTEP), HEIGHT, WIDTH))
scaled_predTestY = np.reshape(predY, ((test.shape[0] - TIMESTEP), HEIGHT, WIDTH))
print('test scale shape: ', scaled_predTestY.shape)
scale_MSE = np.mean((scaled_testY - scaled_predTestY) ** 2)
print("Model scaled MSE", scale_MSE)
rescale_MSE = scale_MSE * MAX_VALUE ** 2
print("Model rescaled MSE", rescale_MSE)
with open(PATH + '/' + MODELNAME + '_prediction_scores.txt', 'a') as wf:
wf.write("Keras MSE on testData, %f\n" % scale_MSE)
wf.write("Rescaled MSE on testData, %f\n" % rescale_MSE)
np.save(PATH + '/' + MODELNAME + '_prediction.npy', scaled_predTestY * MAX_VALUE)
np.save(PATH + '/' + MODELNAME + '_groundtruth.npy', scaled_testY * MAX_VALUE)
def model_train(model_name, train_data, valid_data, trainTemporal,
validTemporal, topo_data):
# set callbacks
csv_logger = CSVLogger(PATH + '/' + MODELNAME + '.log')
checkpointer_path = PATH + '/' + MODELNAME + '.h5'
checkpointer = ModelCheckpoint(filepath=checkpointer_path,
verbose=1,
save_best_only=True)
early_stopping = EarlyStopping(monitor='val_loss',
patience=10,
verbose=1,
mode='auto')
LearnRate = LearningRateScheduler(lambda epoch: LR)
# data generator
train_generator = data_generator(train_data, trainTemporal, topo_data,
BATCHSIZE, TIMESTEP, model_name)
val_generator = data_generator(valid_data, validTemporal, topo_data,
BATCHSIZE, TIMESTEP, model_name)
sep = (train_data.shape[0] - TIMESTEP) * train_data.shape[1] // BATCHSIZE
val_sep = (valid_data.shape[0] -
TIMESTEP) * valid_data.shape[1] // BATCHSIZE
# train model
model = get_model_structure(model_name)
# model = multi_gpu_model(model, gpus=2) # gpu parallel
model.compile(loss=LOSS, optimizer=OPTIMIZER)
model.fit_generator(
train_generator,
steps_per_epoch=sep,
epochs=EPOCH,
validation_data=val_generator,
validation_steps=val_sep,
callbacks=[csv_logger, checkpointer, LearnRate, early_stopping])
# compute mse
val_nolabel_generator = test_generator(valid_data, validTemporal,
topo_data, BATCHSIZE, TIMESTEP)
val_predY = model.predict_generator(val_nolabel_generator, steps=val_sep)
valY = get_test_true(valid_data, TIMESTEP, model_name)
# mse
scaled_valY = np.reshape(valY,
((valid_data.shape[0] - TIMESTEP), HEIGHT, WIDTH))
scaled_predValY = np.reshape(
val_predY, ((valid_data.shape[0] - TIMESTEP), HEIGHT, WIDTH))
print('val scale shape: ', scaled_predValY.shape)
val_scale_MSE = np.mean((scaled_valY - scaled_predValY)**2)
print("Model val scaled MSE", val_scale_MSE)
# rescale mse
val_rescale_MSE = val_scale_MSE * MAX_VALUE**2
print("Model val rescaled MSE", val_rescale_MSE)
# write record
with open(PATH + '/' + MODELNAME + '_prediction_scores.txt', 'a') as wf:
wf.write('train start time: {}\n'.format(StartTime))
wf.write('train end time: {}\n'.format(
datetime.datetime.now().strftime('%Y%m%d_%H%M%S')))
wf.write("Keras MSE on trainData, %f\n" % val_scale_MSE)
wf.write("Rescaled MSE on trainData, %f\n" % val_rescale_MSE)
def model_pred(model_name, test, testTemporal, topo_data, type):
# test generator
test_gene = test_generator(test, testTemporal, topo_data, BATCHSIZE,
TIMESTEP)
test_sep = (test.shape[0] - TIMESTEP) * test.shape[1] // BATCHSIZE
# get predict
model = get_model_structure(model_name)
# model = multi_gpu_model(model, gpus=2) # gpu parallel
model.compile(loss=LOSS, optimizer=OPTIMIZER)
model.load_weights(PATH + '/' + MODELNAME + '_' + type + '.h5')
predY = model.predict_generator(test_gene, steps=test_sep)
# ground truth
testY = get_test_true(test, TIMESTEP, model_name)
# compute mse
scaled_testY = np.reshape(
testY, ((test.shape[0] - TIMESTEP), REGION_SIZE, REGION_SIZE))
scaled_predTestY = np.reshape(
predY, ((test.shape[0] - TIMESTEP), REGION_SIZE, REGION_SIZE))
print('test scale shape: ', scaled_predTestY.shape)
scale_MSE = np.mean((scaled_testY - scaled_predTestY)**2)
print("Model scaled MSE", scale_MSE)
retestY, repredTestY = testY * MAX_VALUE, predY * MAX_VALUE
retestY = np.reshape(
retestY, ((test.shape[0] - TIMESTEP), REGION_SIZE, REGION_SIZE))
repredTestY = np.reshape(
repredTestY, ((test.shape[0] - TIMESTEP), REGION_SIZE, REGION_SIZE))
print('rescale shape: ', repredTestY.shape)
rescale_MSE = np.mean((retestY - repredTestY)**2)
print("Model rescaled MSE", rescale_MSE)
with open(PATH + '/' + MODELNAME + '_prediction_scores.txt', 'a') as wf:
wf.write("Keras MSE on flow {} testData, {}\n".format(type, scale_MSE))
wf.write("Rescaled MSE on flow {} testData, {}\n\n".format(
type, rescale_MSE))
return scale_MSE, rescale_MSE
