Last active
February 22, 2022 17:17
-
-
Save orhermansaffar/fc0f2a9730768e3357f6e64585329e0a to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# How to use TimeBasedCV | |
data_for_modeling=pd.read_csv('data.csv', parse_dates=['record_date']) | |
tscv = TimeBasedCV(train_period=30, | |
test_period=7, | |
freq='days') | |
for train_index, test_index in tscv.split(data_for_modeling, | |
validation_split_date=datetime.date(2019,2,1), date_column='record_date'): | |
print(train_index, test_index) | |
# get number of splits | |
tscv.get_n_splits() | |
#### Example- compute average test sets score: #### | |
X = data_for_modeling[['record_date',columns]] | |
y = data_for_modeling[label] | |
from sklearn.linear_model import LinearRegression | |
import numpy as np | |
scores = [] | |
for train_index, test_index in tscv.split(X, validation_split_date=datetime.date(2019,2,1)): | |
data_train = X.loc[train_index].drop('record_date', axis=1) | |
target_train = y.loc[train_index] | |
data_test = X.loc[test_index].drop('record_date', axis=1) | |
target_test = y.loc[test_index] | |
# if needed, do preprocessing here | |
clf = LinearRegression() | |
clf.fit(data_train,target_train) | |
preds = clf.predict(data_test) | |
# accuracy for the current fold only | |
r2score = clf.score(data_test,target_test) | |
scores.append(r2score) | |
# this is the average accuracy over all folds | |
average_r2score = np.mean(scores) | |
#### End of example #### | |
#### Example- RandomizedSearchCV #### | |
from sklearn.model_selection import RandomizedSearchCV | |
from lightgbm import LGBMRegressor | |
from random import randint, uniform | |
tscv = TimeBasedCV(train_period=10, test_period=3) | |
index_output = tscv.split(data_for_modeling, validation_split_date=datetime.date(2019,2,1)) | |
lgbm = LGBMRegressor() | |
lgbmPd = {" max_depth": [-1,2] | |
} | |
model = RandomizedSearchCV( | |
estimator = lgbm, | |
param_distributions = lgbmPd, | |
n_iter = 10, | |
n_jobs = -1, | |
iid = True, | |
cv = index_output, | |
verbose=5, | |
pre_dispatch='2*n_jobs', | |
random_state = None, | |
return_train_score = True) | |
model.fit(X.drop('record_date', axis=1),y) | |
model.cv_results_ | |
#### End of example #### |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment