def HMM_test(if_train=True):
model_is_existed = os.path.exists(ModelPathConfig.hmm)
print("upload data!")
word_lists, tag_lists, word2id, tag2id = build_corpus("train")
test_word_lists, test_tag_lists, _, _ = build_corpus("test")
# word_lists,tag_lists,word2id,tag2id=build_corpus("train",data_dir=os.path.join(os.getcwd(),"data",'ResumeNER'))
# test_word_lists,test_tag_lists,_,_=build_corpus("test",data_dir=os.path.join(os.getcwd(),"data",'ResumeNER'))
hmm_model = HMM(len(tag2id), len(word2id))
if if_train or not model_is_existed:
print("start to training")
hmm_model.train(word_lists, tag_lists, word2id, tag2id)
print("save the model")
save_model(hmm_model, ModelPathConfig.hmm)
else:
print("load model")
hmm_model = load_model(ModelPathConfig.hmm)
pred_tag_lists = hmm_model.test(test_word_lists, _, word2id, tag2id)
label_tag_lists = test_tag_lists
units = evaluate_entity_label(pred_tag_lists, label_tag_lists,
list(tag2id.keys()))
df = unitstopd(units)
df.to_csv(ResultPathConfig.hmm_entity)
print(tabulate(df, headers='keys', tablefmt='psql'))
units = evaluate_single_label(pred_tag_lists, label_tag_lists,
list(tag2id.keys()))
df = unitstopd(units)
df.to_csv(ResultPathConfig.hmm_model)
print(tabulate(df, headers='keys', tablefmt='psql'))
def hmm_pred(train_word_lists, train_tag_lists, test_word_lists,
test_tag_lists, word2id, tag2id):
model = HMM(len(tag2id), len(word2id))
model.train(train_word_lists, train_tag_lists, word2id, tag2id)
save_model(model, "./ckpts/hmm.pkl")
pred = model.test(test_word_lists, word2id, tag2id)
return pred
def hmm_train_eval(train_data, test_data, word2id, tag2id, remove_O=False):
# data
train_word_lists, train_tag_lists = train_data
test_word_lists, test_tag_lists = test_data
# training
hmm_model = HMM(len(tag2id), len(word2id))
hmm_model.train(train_word_lists, train_tag_lists, word2id, tag2id)
save_model(hmm_model, "./ckpts/hmm.pkl")
# evaluating
pred_tag_lists = hmm_model.test(test_word_lists, word2id, tag2id)
results_print(test_tag_lists, pred_tag_lists, remove_O=remove_O)
return pred_tag_lists
def hmm_train_eval(train_data, test_data, word2id, tag2id):
"""训练并评估hmm模型"""
# 训练HMM模型
train_word_lists, train_tag_lists = train_data
test_word_lists, test_tag_lists = test_data
hmm_model = HMM(len(tag2id), len(word2id))
hmm_model.train(train_word_lists, train_tag_lists, word2id, tag2id)
save_model(hmm_model, "./ckpts/hmm.pkl")
# 评估hmm模型
pred_tag_lists = hmm_model.test(test_word_lists, word2id, tag2id)
metrics = Metrics(test_tag_lists, pred_tag_lists)
return pred_tag_lists
def hmm_train_eval(train_data, test_data, word2id, tag2id):
"""训练并评估hmm模型"""
# 训练HMM模型
train_word_lists, train_tag_lists = train_data
test_word_lists, test_tag_lists = test_data
hmm_model = HMM(len(tag2id), len(word2id))
hmm_model.train(train_word_lists, train_tag_lists, word2id, tag2id)
save_model(hmm_model, "./ckpts/hmm.pkl")
# 评估hmm模型
pred_tag_lists = hmm_model.test(test_word_lists, word2id, tag2id)
accuracy = evaluate(pred_tag_lists, test_tag_lists)
print("HMM 模型的准确率为：{:.2f}%".format(accuracy * 100))
return pred_tag_lists
def hmm_train_eval(train_data, test_data, word2id, tag2id, remove_O=False):
"""训练并评估hmm模型"""
# 训练HMM模型
train_word_lists, train_tag_lists = train_data
test_word_lists, test_tag_lists = test_data
hmm_model = HMM(len(tag2id), len(word2id))
hmm_model.train(train_word_lists, train_tag_lists, word2id, tag2id)
save_model(hmm_model, "./ckpts/hmm.pkl")
# 评估hmm模型
pred_tag_lists = hmm_model.test(test_word_lists, word2id, tag2id)
metrics = Metrics(test_tag_lists, pred_tag_lists, remove_O=remove_O)
metrics.report_scores()
metrics.report_confusion_matrix()
return pred_tag_lists
def hmm_train_eval(train_data, test_data, word2id, tag2id, remove_O=False):
""" Train and evaluate the hmm model """
# Training HMM model
train_word_lists, train_tag_lists = train_data
test_word_lists, test_tag_lists = test_data
hmm_model = HMM(len(tag2id), len(word2id))
hmm_model.train(train_word_lists, train_tag_lists, word2id, tag2id)
save_model(hmm_model, "./ckpts/hmm.pkl")
#Evaluation of the hmm model
pred_tag_lists = hmm_model.test(test_word_lists, word2id, tag2id)
metrics = Metrics(test_tag_lists, pred_tag_lists, remove_O=remove_O)
metrics.report_scores()
metrics.report_confusion_matrix()
return pred_tag_lists
def train(train_data, val_data, fold_idx=None):
train_dataset = MyDataset(train_data)
val_dataset = MyDataset(val_data)
train_loader = DataLoader(train_dataset, batch_size=config.batch_size)
val_loader = DataLoader(val_dataset, batch_size=config.batch_size)
from models.hmm import HMM
word2id, id2word = load_vocab()
model = HMM(len(config.label2id), len(word2id))
if fold_idx is None:
print('start')
model_save_path = os.path.join(config.model_path,
'{}.bin'.format(model_name))
else:
print('start fold: {}'.format(fold_idx + 1))
model_save_path = os.path.join(
config.model_path, '{}_fold{}.bin'.format(model_name, fold_idx))
word_id_list = train_dataset.x_data
label_id_list = train_dataset.y_data
model.train(word_id_list, label_id_list)
y_pred_list = model.predict(train_dataset.x_data)
train_score = get_score(train_dataset.y_data, y_pred_list)
y_pred_list = model.predict(val_dataset.x_data)
val_score = get_score(val_dataset.y_data, y_pred_list)
msg = 'train score: {0:>6.2%}, val score: {1:>6.2%}'
print(msg.format(train_score, val_score))
#!/usr/bin/env python3
from pos_data import read
from models.simple import Simple
from models.hmm import HMM
from models.complex import Complex
from metrics import print_report
X_train, y_train = read('data/bc.train', 'train')
X_val, y_val = read('data/bc.val', 'train')
print('Simple model')
sm = Simple()
sm.fit(X_train, y_train)
y_pred = sm.predict(X_val)
print_report(y_val, y_pred)
print('HMM')
hm = HMM()
hm.fit(X_train, y_train)
y_pred = hm.predict(X_val)
print_report(y_val, y_pred)
print('Complex model')
cp = Complex()
cp.fit(X_train, y_train)
y_pred = cp.predict(X_val)
print_report(y_val, y_pred)
def hmm_train(train_data, word2id, tag2id):
train_word_lists, train_tag_lists = train_data
hmm_model = HMM(len(tag2id), len(word2id))
hmm_model.train(train_word_lists, train_tag_lists, word2id, tag2id)
save_model(hmm_model, HMM_MODEL_PATH)