def xception_multi_period_score_fusion(dataset, config):
result = []
for i in range(iteration):
pre_list = []
y_test_list = []
for period in ['R1', 'R3', 'R4', 'R5', 'R6']:
model = BaseModel.Xception_Model(parallels=4, config=config)
model.load_weights('xception_img_{}_itreation-{}-{}.hdf5'.format(
dataset, i, period))
file_list = np.loadtxt('{}-{}_file_list.txt'.format(
dataset, period))
img_x_list, y_list = utils.data_loader_for_xception_model(
file_list=file_list, config=config)
x = np.array(img_x_list)
y = np.array(y_list)
id_map = np.loadtxt(dataset + '_id.txt')
for index, d in enumerate(y):
for label in id_map:
if d == label[0]:
y[index] = label[1]
y_one_hot = to_categorical(y)
test_index = np.load(
'{}_iteration_{}_img_{}_xception_test_index.npy'.format(
dataset, i, period))
img_x_test = x[test_index]
y_test = y_one_hot[test_index]
y_test_list.append(y_test)
pre = model.predict(img_x_test)
pre_list.append(pre)
pre_final_arr = np.array(pre_list)
pre_final = np.sum(pre_final_arr, axis=2)
pre_final_label = [np.argmax(d) for d in pre_final]
for i in range(1, len(y_test_list) + 1):
if y_test_list[i - 1] != y_test_list[i]:
print("The test label of different period should be the same")
return -1
y_test_label = [np.argmax(d) for d in y_test_list[0]]
performance = get_performance(pre_final_label, y_test_label)
result.append(performance)
plot_result(result)
def xception_model_training_and_test(img_x_list, y_list, config):
x = np.array(img_x_list)
y = np.array(y_list)
dataset = config['dataset']
if dataset == 'soybean':
period = config['period']
else:
period = dataset
id_map = np.loadtxt(dataset + '_id.txt')
for index, d in enumerate(y):
for label in id_map:
if d == label[0]:
y[index] = label[1]
y_one_hot = to_categorical(y)
lr_adjust = ReduceLROnPlateau(monitor='val_loss',
factor=0.5,
patience=5,
min_lr=1e-6)
result = []
for i in range(iteration):
index = np.arange(len(y))
# print(index)
X_train_index, X_test_index, y_train_index, y_test_index = train_test_split(
index, y, test_size=0.3, random_state=i, shuffle=True, stratify=y)
print(len(X_train_index))
print(len(X_test_index))
np.save(
'{}_iteration_{}_img_{}_xception_train_index.npy'.format(
dataset, i, period), X_train_index)
np.save(
'{}_iteration_{}_img_{}_xception_test_index.npy'.format(
dataset, i, period), X_test_index)
X_train = x[X_train_index]
X_test = x[X_test_index]
y_train = y_one_hot[X_train_index]
y_test = y_one_hot[X_test_index]
save_best_weight = ModelCheckpoint(
'xception_img_{}_itreation-{}-{}.hdf5'.format(dataset, i, period),
monitor='val_loss',
verbose=1,
save_best_only=True,
mode='auto',
save_weights_only=True)
# you can change the parallels to create multi_gpu_model if you have more than one GPU available
model = BaseModel.Xception_Model(parallels=1, config=config)
# you should set a smaller batch_size if you GPU memory is limited
model.fit(X_train,
y_train,
batch_size=32,
epochs=100,
validation_split=0.1,
callbacks=[lr_adjust, save_best_weight])
K.clear_session()
model2 = BaseModel.Xception_Model(parallels=1, config=config)
model2.load_weights('xception_img_{}_itreation-{}-{}.hdf5'.format(
dataset, i, period))
score = model2.evaluate(X_test, y_test)
print(score)
pre_final = model2.predict(X_test, batch_size=128)
y_test_label = np.array([np.argmax(d) for d in y_test])
y_pre_label = np.array([np.argmax(d) for d in pre_final])
performance = get_performance(y_pre_label, y_test_label)
performance['test_loss'] = score[0]
performance['test_acc'] = score[1]
K.clear_session()
result.append(performance)
json_str = json.dumps(performance, indent=4)
with open(
'{}_xception-iteration-{}-{}-result.json'.format(
dataset, i, period), 'w') as json_file:
json_file.write(json_str)
plot_result(result)