class Classifier(object):
def __init__(self, conf, task, train=None, test=None):
self.conf = conf
self.task = task
self.train_ = train
self.test_ = test
self.features = [
"hasWith", "hasIn", "simiBucket", "textPos", "hasOf", "hasAnd",
"startEntity", "distance", "hasFrom", "endEntity", "similarity",
"hasThan", "hasVerb"
]
self.labels = ["relation"]
self.num_round = 500
self.eval_set = list()
self.early_stopping_rounds = 20
self.classifier = XGBClassifier(max_depth=4,
learning_rate=0.1,
n_estimators=1000,
gamma=4,
verbosity=1,
objective='multi:softmax',
num_class=6,
booster='gbtree',
n_jobs=4,
seed=27)
def train(self):
train_X, test_X, train_y, test_y = train_test_split(
self.train_[self.features],
self.train_[self.labels],
test_size=0.4,
random_state=42)
self.eval_set = [(train_X.values, train_y.values),
(test_X.values, test_y.values)]
self.classifier.fit(train_X.values,
train_y.values,
eval_metric='merror',
eval_set=self.eval_set,
early_stopping_rounds=self.early_stopping_rounds,
verbose=True)
self.classifier.save_model(self.conf.model_path.format(self.task))
return 'Model has been saved!'
def test(self):
test_set = self.test_[self.features].values
self.classifier.load_model(self.conf.model_path.format(self.task))
self.classifier._le = LabelEncoder().fit([
'USAGE', 'TOPIC', 'MODEL-FEATURE', 'PART_WHOLE', 'RESULT',
'COMPARE'
])
pred = self.classifier.predict(test_set)
predictions = pd.concat([
self.test_[self.features],
pd.DataFrame(pred, columns=["relation"])
],
axis=1)
return predictions
def xgboost_train(X_train, X_test, y_train, y_test, model_path):
print('Training by xgb')
# default is binary:logistic
train_model = XGBClassifier(use_label_encoder=False).fit(X_train, y_train)
pred = train_model.predict(X_test)
print('Classification report:\n', classification_report(y_test, pred))
auc = accuracy_score(y_test, pred) * 100
print(f'Accuracy score: {auc}')
print('Save model to ', model_path)
train_model.save_model(model_path)
return auc
GS_XGB.best_score_))
print("The best params found in epoch {} is:\n{}".format(e, ind_params))
if (old_params == ind_params) and (e >= 3):
best_params_['GDM'] = ind_params
break
old_params = ind_params.copy()
random.shuffle(cv_params_list_copy)
print('----{}_{} TRAINING FINISHED----'.format('GDM', '1'))
best_params_file = pd.DataFrame.from_dict(best_params_)
clf = XGBClassifier(random_state=1,
scale_pos_weight=12.026280323450134,
n_estimators=200,
max_depth=2,
min_child_weight=29,
colsample_bytree=0.7,
subsample=1,
gamma=0,
reg_alpha=5,
reg_lambda=5,
learning_rate=0.1,
n_jobs=-1).fit(X_train, y_train)
pred_train, pred_test = clf.predict_proba(X_train), clf.predict_proba(X_test)
score_train, score_test = roc_auc_score(y_train,
pred_train[:, 1]), roc_auc_score(
y_test, pred_test[:, 1])
print(score_train, score_test)
clf.save_model("G:/Thesis/code/xgboost.model")
gc.collect()
print('Training by xgb')
# default is binary:logistic
train_model = XGBClassifier(use_label_encoder=False).fit(X_train, y_train)
pred = train_model.predict(X_test)
print('Classification report:\n', classification_report(y_test, pred))
print(f'Accuracy score: {accuracy_score(y_test, pred) * 100}')
del train_df
gc.collect()
print('Save model to ', MODEL_PATH)
train_model.save_model(MODEL_PATH)
print('Prepare online serving')
print('Deploy sql')
connection.execute("SET @@execute_mode='online';")
connection.execute(f'USE {DB_NAME}')
nothrow_execute(f'DROP DEPLOYMENT {DEPLOY_NAME}')
deploy_sql = f"""DEPLOY {DEPLOY_NAME} {sql_part}"""
print(deploy_sql)
connection.execute(deploy_sql)
print('Import data to online')
# online feature extraction needs history data
# set job_timeout bigger if the `LOAD DATA` job timeout
connection.execute(
f"LOAD DATA INFILE 'file://{os.path.abspath('train_sample.csv')}' "
thresh = 0.01702188141644001
selection = SelectFromModel(model, threshold=thresh, prefit=True)
select_X = selection.transform(X)
model = XGBClassifier(
max_depth=3,
learning_rate=0.01,
verbosity=None, # might throw an error, just put it as 0 if it does
objective= 'binary:logistic',
booster='gbtree',
tree_method='exact',
n_jobs=0,
gamma=0.0,
min_child_weight=4,
max_delta_step=0,
subsample=.85,
colsample_bytree=0.6,
colsample_bynode=1,
reg_alpha=0.1,
reg_lambda=100,
scale_pos_weight=1,
base_score=0.5,
random_state=812)
model = model.fit(select_X, y, eval_metric='auc')
'''
model.save_model('FULLNBAMODEL2010-2020test.model')
print("model complete")
