def initialize_model():
'''
Calls on the StatusModel class, processes training data and trains model.
Returns:
StatusModel trained with 2012-2014 3-year loan data. statusmodel class.
'''
model = StatusModel(model=RandomForestRegressor,
parameters={
'n_estimators': 100,
'max_depth': 10
})
try:
df_3c = pd.read_csv('data/LoanStats3c_securev1.csv',
header=True).iloc[:-2, :]
df_3b = pd.read_csv('data/LoanStats3b_securev1.csv',
header=True).iloc[:-2, :]
except (OSError, IOError):
print "Training data not found. Please install from https://www.lendingclub.com/info/download-data.action"
df_train = pd.concat((df_3c, df_3b), axis=0)
df_train = process_features(df_train,
restrict_date=True,
current_loans=True)
model.train_model(df_train)
dump_to_pickle(model, 'pickle/model.pkl')
return model
def initialize_model():
'''
Calls on the StatusModel class, processes training data and trains model.
Returns:
StatusModel trained with 2012-2014 3-year loan data. statusmodel class.
'''
model = StatusModel(model=RandomForestRegressor,
parameters={'n_estimators':100,
'max_depth':10})
try:
df_3c = pd.read_csv('data/LoanStats3c_securev1.csv', header=True).iloc[:-2, :]
df_3b = pd.read_csv('data/LoanStats3b_securev1.csv', header=True).iloc[:-2, :]
except (OSError, IOError):
print "Training data not found. Please install from https://www.lendingclub.com/info/download-data.action"
df_train = pd.concat((df_3c, df_3b), axis=0)
df_train = process_features(df_train, restrict_date=True, current_loans=True)
model.train_model(df_train)
dump_to_pickle(model, 'pickle/model.pkl')
return model
df = process_features(df_raw)
print "Initializing model..."
model = StatusModel(model=RandomForestRegressor,
parameters={
'n_estimators': 100,
'max_depth': 10
})
print "Training model..."
try:
model = load_from_pickle('pickle/model.pkl')
except OSError, IOError:
print "Model not found. Training model, this might take some time..."
model.train_model(df)
dump_to_pickle(model, 'pickle/model.pkl')
print "Calculating IRR..."
int_rate_dict = {
'A1': 0.0603,
'A2': 0.0649,
'A3': 0.0699,
'A4': 0.0749,
'A5': 0.0819,
'B1': 0.0867,
'B2': 0.0949,
'B3': 0.1049,
'B4': 0.1144,
'B5': 0.1199,
'C1': 0.1239,
'C2': 0.1299,
df_raw = pd.concat((df_3c, df_3b), axis=0)
print "Pre-processing data..."
df = process_features(df_raw)
print "Initializing model..."
model = StatusModel(model=RandomForestRegressor, parameters={"n_estimators": 100, "max_depth": 10})
print "Training model..."
try:
model = load_from_pickle("pickle/model.pkl")
except OSError, IOError:
print "Model not found. Training model, this might take some time..."
model.train_model(df)
dump_to_pickle(model, "pickle/model.pkl")
print "Calculating IRR..."
int_rate_dict = {
"A1": 0.0603,
"A2": 0.0649,
"A3": 0.0699,
"A4": 0.0749,
"A5": 0.0819,
"B1": 0.0867,
"B2": 0.0949,
"B3": 0.1049,
"B4": 0.1144,
"B5": 0.1199,
"C1": 0.1239,
"C2": 0.1299,
