def KNN(trainning_X, train_Y, predict_X, task_name, origin_data, stock_id):
clf = KNeighborsClassifier(5)
clf.fit(trainning_X, train_Y)
res = clf.predict(predict_X)
res_ = []
for i in range(len(res)):
if res[i] == 1:
res_.append(origin_data[stock_id][i])
tsk = training_result(taskName=task_name, clf_name="KNN(5)", result=res_)
tsk.save()
def SVC(trainning_X, train_Y, predict_X, task_name, origin_data, stock_id):
clf = svm.SVC(gamma=2, C=1)
clf.fit(trainning_X, train_Y)
res = clf.predict(predict_X)
res_ = []
for i in range(len(res)):
if res[i] == 1:
res_.append(origin_data[stock_id][i])
tsk = training_result(taskName=task_name, clf_name="RBF SVM", result=res_)
tsk.save()
def adaboost(trainning_X, train_Y, predict_X, task_name, origin_data,
stock_id):
clf = AdaBoostClassifier()
clf.fit(trainning_X, train_Y)
res = clf.predict(predict_X)
res_ = []
for i in range(len(res)):
if res[i] == 1:
res_.append(origin_data[stock_id][i])
tsk = training_result(taskName=task_name, clf_name="adaboost", result=res_)
tsk.save()
def RandomForest(trainning_X, train_Y, predict_X, task_name, origin_data,
stock_id):
clf = RandomForestClassifier(max_depth=5, n_estimators=10, max_features=1)
clf.fit(trainning_X, train_Y)
res = clf.predict(predict_X)
res_ = []
for i in range(len(res)):
if res[i] == 1:
res_.append(origin_data[stock_id][i])
tsk = training_result(taskName=task_name,
clf_name="Random Forest",
result=res_)
tsk.save()
def DecisionTree(trainning_X, train_Y, predict_X, task_name, origin_data,
stock_id):
clf = DecisionTreeClassifier(max_depth=5)
clf.fit(trainning_X, train_Y)
res = clf.predict(predict_X)
res_ = []
for i in range(len(res)):
if res[i] == 1:
res_.append(origin_data[stock_id][i])
tsk = training_result(taskName=task_name,
clf_name="Decision Tree",
result=res_)
tsk.save()
def LinearDiscriminant(trainning_X, train_Y, predict_X, task_name, origin_data,
stock_id):
clf = LinearDiscriminantAnalysis()
clf.fit(trainning_X, train_Y)
res = clf.predict(predict_X)
res_ = []
for i in range(len(res)):
if res[i] == 1:
res_.append(origin_data[stock_id][i])
tsk = training_result(taskName=task_name,
clf_name="Linear Discriminant Analysis",
result=res_)
tsk.save()
def Gaussian(trainning_X, train_Y, predict_X, task_name, origin_data,
stock_id):
clf = GaussianNB()
clf.fit(trainning_X, train_Y)
res = clf.predict(predict_X)
res_ = []
for i in range(len(res)):
if res[i] == 1:
res_.append(origin_data[stock_id][i])
tsk = training_result(taskName=task_name,
clf_name="GaussianNB",
result=res_)
tsk.save()