Skip to content

Instantly share code, notes, and snippets.

View fe13.py
print(train['Gender'].unique())
print(train['City_Category'].unique())
print(train['Age'].unique())
print(train['Stay_In_Current_City_Years'].unique())
print(train['Product_ID'].unique())
View fe12.py
train['Gender'].unique()
View fe10.py
train["Product_Cat1_MaxPrice"] = train.groupby(['Product_Category_1'])['Purchase'].transform('max')
pc1_max_dict = train.groupby(['Product_Category_1'])['Purchase'].max().to_dict()
test['Product_Cat1_MaxPrice'] = test['Product_Category_1'].apply(lambda x:pc1_max_dict.get(x,0))
train["Product_Cat1_MeanPrice"] = train.groupby(['Product_Category_1'])['Purchase'].transform('mean')
pc1_mean_dict = train.groupby(['Product_Category_1'])['Purchase'].mean().to_dict()
test['Product_Cat1_MeanPrice'] = test['Product_Category_1'].apply(lambda x:pc1_mean_dict.get(x,0))
train["Age_Count"] = train.groupby(['Age'])['Age'].transform('count')
age_count_dict = train.groupby(['Age']).size().to_dict()
View fe9.py
train["User_ID_MinPrice"] = train.groupby(['User_ID'])['Purchase'].transform('min')
userID_min_dict = train.groupby(['User_ID'])['Purchase'].min().to_dict()
test['User_ID_MinPrice'] = test['User_ID'].apply(lambda x:userID_min_dict.get(x,0))
train["User_ID_MaxPrice"] = train.groupby(['User_ID'])['Purchase'].transform('max')
userID_max_dict = train.groupby(['User_ID'])['Purchase'].max().to_dict()
test['User_ID_MaxPrice'] = test['User_ID'].apply(lambda x:userID_max_dict.get(x,0))
train["Product_ID_MinPrice"] = train.groupby(['Product_ID'])['Purchase'].transform('min')
productID_min_dict = train.groupby(['Product_ID'])['Purchase'].min().to_dict()
View fe8.py
train["User_ID_MeanPrice"] = train.groupby(['User_ID'])['Purchase'].transform('mean')
userID_mean_dict = train.groupby(['User_ID'])['Purchase'].mean().to_dict()
test['User_ID_MeanPrice'] = test['User_ID'].apply(lambda x:userID_mean_dict.get(x,0))
train["Product_ID_MeanPrice"] = train.groupby(['Product_ID'])['Purchase'].transform('mean')
productID_mean_dict = train.groupby(['Product_ID'])['Purchase'].mean().to_dict()
test['Product_ID_MeanPrice'] = test['Product_ID'].apply(lambda x:productID_mean_dict.get(x,0))
View fe7.py
dtr = DecisionTreeRegressor()
dtr.fit(X_train,Y_train)
y_pred = dtr.predict(X_test)
y_pred_dt=dtr.predict(test)
submission['Purchase'] = y_pred_dt
submission.to_csv('dtr_model3.csv',index=False)
mse = mean_squared_error(Y_test, y_pred)
print("RMSE Error:", np.sqrt(mse))
r2 = r2_score(Y_test, y_pred)
print("R2 Score:", r2)
View fe6.py
train.fillna(-999, inplace=True)
test.fillna(-999, inplace=True)
View fe5.py
gender_dict = {'F':0, 'M':1}
age_dict = {'0-17':0, '18-25':1, '26-35':2, '36-45':3, '46-50':4, '51-55':5, '55+':6}
city_dict = {'A':0, 'B':1, 'C':2}
stay_dict = {'0':0, '1':1, '2':2, '3':3, '4+':4}
train["Gender"] = train["Gender"].apply(lambda x: gender_dict[x])
test["Gender"] = test["Gender"].apply(lambda x: gender_dict[x])
train["Age"] = train["Age"].apply(lambda x: age_dict[x])
test["Age"] = test["Age"].apply(lambda x: age_dict[x])
View fe4.py
submission = pd.DataFrame()
submission['User_ID'] = test['User_ID']
submission['Product_ID'] = test['Product_ID']