print("loading CNN_MUI model......")
model = model_CNN_MUI.CNN_MUI(args)
shutil.copy("./models/model_CNN_MUI.py", "./snapshot/" + mulu)
print(model)
else:
print('\nLoading model from [%s]...' % args.snapshot)
try:
model = torch.load(args.snapshot)
except:
print("Sorry, This snapshot doesn't exist.")
exit()
# train or predict
if args.predict is not None:
label = train_ALL_CNN.predict(args.predict, model, text_field, label_field)
print('\n[Text] {}[Label] {}\n'.format(args.predict, label))
elif args.test:
try:
print(test_iter)
train_ALL_CNN.test_eval(test_iter, model, args)
except Exception as e:
print("\nSorry. The test dataset doesn't exist.\n")
else:
print("\n cpu_count \n", mu.cpu_count())
torch.set_num_threads(args.num_threads)
if os.path.exists("./Test_Result.txt"):
os.remove("./Test_Result.txt")
if args.CNN:
print("CNN training start......")
model_count = train_ALL_CNN.train(train_iter, dev_iter, test_iter, model, args)
def start_train(model, train_iter, dev_iter, test_iter):
"""
:function：start train
:param model:
:param train_iter:
:param dev_iter:
:param test_iter:
:return:
"""
if config.predict is not None:
label = train_ALL_CNN.predict(config.predict, model, config.text_field,
config.label_field)
print('\n[Text] {}[Label] {}\n'.format(config.predict, label))
elif config.test:
try:
print(test_iter)
train_ALL_CNN.test_eval(test_iter, model, config)
except Exception as e:
print("\nSorry. The test dataset doesn't exist.\n")
else:
print("\n cpu_count \n", mu.cpu_count())
torch.set_num_threads(config.num_threads)
if os.path.exists("./Test_Result.txt"):
os.remove("./Test_Result.txt")
if config.CNN:
print("CNN training start......")
model_count = train_ALL_CNN.train(train_iter, dev_iter, test_iter,
model, config)
elif config.DEEP_CNN:
print("DEEP_CNN training start......")
model_count = train_ALL_CNN.train(train_iter, dev_iter, test_iter,
model, config)
elif config.LSTM:
print("LSTM training start......")
model_count = train_ALL_LSTM.train(train_iter, dev_iter, test_iter,
model, config)
elif config.GRU:
print("GRU training start......")
model_count = train_ALL_LSTM.train(train_iter, dev_iter, test_iter,
model, config)
elif config.BiLSTM:
print("BiLSTM training start......")
model_count = train_ALL_LSTM.train(train_iter, dev_iter, test_iter,
model, config)
elif config.BiLSTM_1:
print("BiLSTM_1 training start......")
model_count = train_ALL_LSTM.train(train_iter, dev_iter, test_iter,
model, config)
elif config.CNN_LSTM:
print("CNN_LSTM training start......")
model_count = train_ALL_LSTM.train(train_iter, dev_iter, test_iter,
model, config)
elif config.CLSTM:
print("CLSTM training start......")
model_count = train_ALL_LSTM.train(train_iter, dev_iter, test_iter,
model, config)
elif config.CBiLSTM:
print("CBiLSTM training start......")
model_count = train_ALL_LSTM.train(train_iter, dev_iter, test_iter,
model, config)
elif config.CGRU:
print("CGRU training start......")
model_count = train_ALL_LSTM.train(train_iter, dev_iter, test_iter,
model, config)
elif config.CNN_BiLSTM:
print("CNN_BiLSTM training start......")
model_count = train_ALL_LSTM.train(train_iter, dev_iter, test_iter,
model, config)
elif config.BiGRU:
print("BiGRU training start......")
model_count = train_ALL_LSTM.train(train_iter, dev_iter, test_iter,
model, config)
elif config.CNN_BiGRU:
print("CNN_BiGRU training start......")
model_count = train_ALL_LSTM.train(train_iter, dev_iter, test_iter,
model, config)
elif config.CNN_MUI:
print("CNN_MUI training start......")
model_count = train_ALL_CNN.train(train_iter, dev_iter, test_iter,
model, config)
elif config.DEEP_CNN_MUI:
print("DEEP_CNN_MUI training start......")
model_count = train_ALL_CNN.train(train_iter, dev_iter, test_iter,
model, config)
elif config.HighWay_CNN is True:
print("HighWay_CNN training start......")
model_count = train_ALL_CNN.train(train_iter, dev_iter, test_iter,
model, config)
elif config.HighWay_BiLSTM_1 is True:
print("HighWay_BiLSTM_1 training start......")
model_count = train_ALL_LSTM.train(train_iter, dev_iter, test_iter,
model, config)
print("Model_count", model_count)
resultlist = []
if os.path.exists("./Test_Result.txt"):
file = open("./Test_Result.txt")
for line in file.readlines():
if line[:10] == "Evaluation":
resultlist.append(float(line[34:41]))
result = sorted(resultlist)
file.close()
file = open("./Test_Result.txt", "a")
file.write("\nThe Best Result is : " +
str(result[len(result) - 1]))
file.write("\n")
file.close()
shutil.copy("./Test_Result.txt",
"./snapshot/" + config.mulu + "/Test_Result.txt")
print("loading HighWay_BiLSTM_1 model......")
model = model_HighWay_BiLSTM_1.HighWay_BiLSTM_1(args)
shutil.copy("./models/model_HighWay_BiLSTM_1.py", "./snapshot/" + mulu)
print(model)
else:
print('\nLoading model from [%s]...' % args.snapshot)
try:
model = torch.load(args.snapshot)
except:
print("Sorry, This snapshot doesn't exist.")
exit()
# train or predict
if args.predict is not None:
label = train_ALL_CNN.predict(args.predict, model, text_field, label_field)
print('\n[Text] {}[Label] {}\n'.format(args.predict, label))
elif args.test:
try:
print(test_iter)
train_ALL_CNN.test_eval(test_iter, model, args)
except Exception as e:
print("\nSorry. The test dataset doesn't exist.\n")
else:
print("\n cpu_count \n", mu.cpu_count())
torch.set_num_threads(args.num_threads)
if os.path.exists("./Test_Result.txt"):
os.remove("./Test_Result.txt")
if args.CNN:
print("CNN training start......")
model_count = train_ALL_CNN.train(train_iter, dev_iter, test_iter, model, args)