def test_input_rf(self) -> None:
""" test for diabetes"""
my_dict = {
"B": float(self.l1.text()),
"C": float(self.l2.text()),
"D": float(self.l3.text()),
"E": float(self.l4.text()),
"F": float(self.l5.text())
}
rf_output = Random_Forest.check_input(my_dict)
self.setFixedSize(850, 342)
self.report_subhead.setText("Reports")
self.model_details.setText(
"Random Forest classifier used.\nAccuracy of model: 77.92%\nWe have used PIMA Indians diabetes dataset."
)
self.details.setText(
"Patient's name: {}\nPlasma glucose concentration: {} \
\nDiastolic blood pressure: {}\nTriceps skin fold thickness: {}\nSerum insulin: {}\nBody mass index: {}"
.format(self.l0.text(), self.l1.text(), self.l2.text(),
self.l3.text(), self.l4.text(), self.l5.text()))
if rf_output == 0:
op = self.results.setText(
"Diagnosis suggests that patient does not suffers from diabetes."
)
else:
op = self.results.setText(
"Our diagnosis suggests patient does suffer from diabetes.\nPlease get checked soon."
)
self.results.setFont(QFont("Arial", 14, weight=QFont.Bold))
def exec_(path, pathR, clf):
tp = [0.25, 0.15, 0.15, 0.15, 0.15, 0.2] #testing set percentage
c = [8, 1.5, 1, 1, 8, 2] #Penalty of svm
N = 15 #Number of most important index
Train_time_ = np.empty((6, 0), float)
Train_acc_ = np.empty((6, 0), float)
Test_acc_ = np.empty((6, 0), float)
name = 'Random_Forest_Results.xlsx'
for a in range(0, 6):
print('Activity # ', a)
tt = []
tra = []
tea = []
for N in range(1, 44):
auxX = []
auxY = []
auxZ = []
for j in range(
1, 10
): #Trials with the same conditions (what varies is the content of training and testing sets)
X, Y, Z = Random_Forest.exec_(a, tp[a], c[a], N, path, clf)
auxX.append(X)
auxY.append(Y)
auxZ.append(Z)
tt.append(np.mean(auxX))
tra.append(np.mean(auxY))
tea.append(np.mean(auxZ))
Train_time_ = np.append(Train_time_, tt)
Train_acc_ = np.append(Train_acc_, tra)
Test_acc_ = np.append(Test_acc_, tea)
Train_time_ = np.reshape(Train_time_, (43, 6))
Train_acc_ = np.reshape(Train_acc_, (43, 6))
Test_acc_ = np.reshape(Test_acc_, (43, 6))
Train_time = pd.DataFrame(Train_time_,
columns=['C1', 'C2', 'F1', 'F2', 'M', 'R'])
Train_acc = pd.DataFrame(Train_acc_,
columns=['C1', 'C2', 'F1', 'F2', 'M', 'R'])
Test_acc = pd.DataFrame(Test_acc_,
columns=['C1', 'C2', 'F1', 'F2', 'M', 'R'])
writer = pd.ExcelWriter(pathR + name)
Train_time.to_excel(writer, 'Train_time')
Train_acc.to_excel(writer, 'Train_acc')
Test_acc.to_excel(writer, 'Test_acc')
writer.save()
print('Random Forest Results Correctly Saved!!!! ')
def main():
print("\nPHM_model_testing")
while True:
user = input("1. Random Forest\n2. Isolation Forest\n3. OCSVM\n4. Classifier SVM\n5. RFE\n6. Linear Regression\n7. Quit\n")
if user == '1':
Random_Forest.run()
elif user == '2':
Isolation_Forest.run()
elif user == '3':
OCSVM.run()
elif user == '4':
SVM.run()
elif user == '5':
RFE.run()
elif user == '6':
Lin_reg.run()
elif user == '7':
break
print("\n----------Main-----------")
def test_input_rf(self) -> None:
""" test for diabetes"""
my_dict = {
"B": float(self.l1.text()),
"C": float(self.l2.text()),
"D": float(self.l3.text()),
"E": float(self.l4.text()),
"F": float(self.l5.text())
}
rf_output = Random_Forest.check_input(my_dict)
global diabetes_yes, diabetes_no
if rf_output == 0:
diabetes_no = diabetes_no + 0.78
else:
diabetes_yes = diabetes_yes + 0.78
def form_submit():
print(request)
the_dict = request.get_json(force=True, silent=True)
months = dict(January=1,
February=2,
March=3,
April=4,
May=5,
June=6,
July=7,
August=8,
September=9,
October=10,
November=11,
December=12)
model_data = Random_Forest.Random_Forest(
plan_model, premium_model, 0 if the_dict['gender'] == 'M' else 1,
the_dict['date'].split(",")[1],
months[the_dict['date'].split(' ')[1][:-1]], the_dict['smoke'],
the_dict['income'], the_dict['optional_insured'], the_dict['weight'],
the_dict['height'], the_dict['married'])
print(model_data['platinum'][0])
#return plans(model_data['bronze'][0], model_data['silver'][0], model_data['gold'][0], model_data['platinum'][0])
'''
{"plan": plan_prediction, "bronze": premiums["bronze"],
"silver": premiums["silver"], "gold": premiums["gold"],
"platinum": premiums["platinum"]}
'''
'''
u'city': u'h3', u'first_name': u'abc', u'last_name': u'def', u'weight': u'3', u'gender': u'3',
u'married': u'0', u'height': u'3', u'state': u'3', u'optional_insured': u'3', u'income': u'3',
u'smoke': u'1', u'address': u'hig', u'date': u'11 December, 2017'}
'''
return json.dumps({
'bronze': model_data['bronze'][0],
'silver': model_data['silver'][0],
'gold': model_data['gold'][0],
'platinum': model_data['platinum'][0]
}), 200, {
'ContentType': 'application/json'
}
def test_build_own_model():
df_train = 'C:/Users/Henry Lee/Desktop/Direct/project/Missing-Density-PDB/model/newsummary_1000.csv'
assert isinstance(RF.build_own_model(df_train), float) is True, \
"Warning: Incorrect output"
return
def get_reason_prediction():
df_train = 'C:/Users/Henry Lee/Desktop/Direct/project/Missing-Density-PDB/model/newsummary_1000.csv'
assert isinstance(RF.get_reason_prediction(df_train, 10), bool) is True, \
"Warning: Incorrect output"
return
print("Patient file created")
else:
print("Patient file already exist, using that")
df_co = pd.read_csv('Transformed Data/Transformed_Co', index_col = 0)
df_pt = pd.read_csv('Transformed Data/Transformed_Pt' , index_col = 0)
print("\n\n\nRandom Forest\n\n")
import Random_Forest
Random_Forest.randomForest(df_co,df_pt)
print("\n\n\nKNN\n\n")
import KNN
KNN.kNN(df_co,df_pt)
print("\n\n\nLinear SVM\n\n")
import SVM_Linear
SVM_Linear.svmLinear(df_co,df_pt)
print("\n\n\nPolynomial SVM\n\n")
import SVM_Poly
months[the_dict['date'].split(' ')[1][:-1]], the_dict['smoke'],
the_dict['income'], the_dict['optional_insured'], the_dict['weight'],
the_dict['height'], the_dict['married'])
print(model_data['platinum'][0])
#return plans(model_data['bronze'][0], model_data['silver'][0], model_data['gold'][0], model_data['platinum'][0])
'''
{"plan": plan_prediction, "bronze": premiums["bronze"],
"silver": premiums["silver"], "gold": premiums["gold"],
"platinum": premiums["platinum"]}
'''
'''
u'city': u'h3', u'first_name': u'abc', u'last_name': u'def', u'weight': u'3', u'gender': u'3',
u'married': u'0', u'height': u'3', u'state': u'3', u'optional_insured': u'3', u'income': u'3',
u'smoke': u'1', u'address': u'hig', u'date': u'11 December, 2017'}
'''
return json.dumps({
'bronze': model_data['bronze'][0],
'silver': model_data['silver'][0],
'gold': model_data['gold'][0],
'platinum': model_data['platinum'][0]
}), 200, {
'ContentType': 'application/json'
}
if __name__ == "__main__":
plan_model = Random_Forest.get_plan_model()
premium_model = Random_Forest.get_premium_model()
app.run(debug=True, host="0.0.0.0", port=8080)
13) SVM
14) XGB
15) ANN
'''
usr_input = int(input(sel))
if usr_input == 1:
# RAW
train_path = './preprocessed_training_set/train_df.csv'
test_path = './preprocessed_training_set/test_df.csv'
KNN_obj = knn.KNN(train_path, test_path)
run_KNN_trainer()
elif usr_input == 2:
# RAW
train_path = './preprocessed_training_set/train_df.csv'
test_path = './preprocessed_training_set/test_df.csv'
RF_obj = rf.RF(train_path, test_path)
run_RF_trainer()
elif usr_input == 3:
# RAW
train_path = './preprocessed_training_set/train_df.csv'
test_path = './preprocessed_training_set/test_df.csv'
SVM_obj = svm.SVM(train_path, test_path)
run_SVM_trainer()
elif usr_input == 4:
# RAW
train_path = './preprocessed_training_set/train_df.csv'
test_path = './preprocessed_training_set/test_df.csv'
XGB_obj = xgb.XGB(train_path, test_path)
run_XGB_trainer()
elif usr_input == 5:
# RAW
if args.grid_search:
doLoadConfig = 1
else:
doLoadConfig = 0
if not (args.infer_rf and args.infer_svm):
if args.train_rf:
configPath_RF = None
if doLoadConfig == 1:
configPath_RF = "curRunConfig_RF.json"
output_dir = getNextRunDir('rf_search')
cfg_RF = Random_Forest.loadConfig_RF(configPath_RF)
Random_Forest.run_rf_once(trainx, trainy, validx, validy)
printConfig_RF(output_dir, cfg_RF)
print('Configs stored in {}'.format(output_dir))
elif args.train_svm:
configPath_SVM = None
if doLoadConfig == 1:
configPath_SVM = "curRunConfig_SVM.json"
output_dir = getNextRunDir('svm_search')
cfg_SVM = Svm.loadConfig_SVM_poly(configPath_SVM)
print(
'----------------------------------------------------------------------------------------------'
)
print('choose your classification method:')
print('1 -----> Bayes')
print('2 -----> Random Forest')
print('3 -----> Logical Regression')
print('4 -----> Boost')
print('5 -----> KNN')
print('6 -----> SVM')
print('7 -----> SGD')
print('0 -----> exit this program')
M = input('Method number = ')
if M == '1':
Bayes.GBC(method, norm, selection)
if M == '2':
Random_Forest.RandomF(method, norm, selection)
if M == '3':
LogicalRegression.LogicR(method, norm, selection)
if M == '4':
Boost.Ada(method, norm, selection)
if M == '5':
KNN.KClassifier(method, norm, selection)
if M == '6':
SVMC.support(method, norm, selection)
if M == '7':
SGD.SGD(method, norm, selection)
if M == '0':
print('The classification progress finished.')
break
def cal_accuracy(y_test, y_pred):
print("Confusion Matrix: ", sk.metrics.confusion_matrix(y_test, y_pred))
print("Accuracy : ", sk.metrics.accuracy_score(y_test, y_pred) * 100)
print("Report : ", sk.metrics.classification_report(y_test, y_pred))
import Decision_tree
import Random_Forest
clf_gini = Decision_tree.train_using_gini(X_train, y_train)
Decision_tree.DecisionTreeclf(clf_gini, cancer_features)
y_pred_DecisionTree = prediction(X_test, clf_gini)
cal_accuracy(y_test, y_pred_DecisionTree)
regressor = Random_Forest.RandomForestRegressorCervical(
X_train, y_train, X_test)
print(regressor)
Y_pred_RandoForest = prediction(X_test, regressor)
cal_accuracy(y_test, Y_pred_RandoForest)
'''
cm = confusion_matrix(val_y, val_predictions)
fig, ax = plt.subplots(figsize=(8, 8))
ax.imshow(cm)
ax.grid(False)
ax.xaxis.set(ticks=(0, 1), ticklabels=('Predicted 0s', 'Predicted 1s'))
ax.yaxis.set(ticks=(0, 1), ticklabels=('Actual 0s', 'Actual 1s'))
ax.set_ylim(1.5, -0.5)
for i in range(2):
for j in range(2):
ax.text(j, i, cm[i, j], ha='center', va='center', color='red')