[python][examples] Added example of custom feval with sklearn wrapper (#1503)

* added example about custom feval with sklearn wrapper

* fixed pylint

* added new example in contents

* unified run example commands for Windows and Linux

* added note about comments in examples conf files

examples/README.md

@@ -3,6 +3,8 @@ Examples
 
 You can learn how to use LightGBM by these examples.
 
+Comments in configuration files might be outdated. Actual information about parameters always can be found [here](https://github.com/Microsoft/LightGBM/blob/master/docs/Parameters.rst).
+
 Machine Learning Challenge Winning Solutions
 ============================================
 

examples/binary_classification/README.md

@@ -5,19 +5,13 @@ Here is an example for LightGBM to run binary classification task.
 
 ***You should copy executable file to this folder first.***
 
-Trainin
--------
+Training
+--------
 
-For Windows, by running following command in this folder:
+Run the following command in this folder:
 
 ```
-lightgbm.exe config=train.conf
-```
-
-For Linux, by running following command in this folder:
-
-```
-./lightgbm config=train.conf
+"./lightgbm" config=train.conf
 ```
 
 Prediction
@@ -25,14 +19,8 @@ Prediction
 
 You should finish training first.
 
-For Windows, by running following command in this folder:
-
-```
-lightgbm.exe config=predict.conf
-```
-
-For Linux, by running following command in this folder:
+Run the following command in this folder:
 
 ```
-./lightgbm config=predict.conf
+"./lightgbm" config=predict.conf
 ```

examples/lambdarank/README.md

@@ -8,16 +8,10 @@ Here is an example for LightGBM to run lambdarank task.
 Training
 --------
 
-For Windows, by running following command in this folder:
+Run the following command in this folder:
 
 ```
-lightgbm.exe config=train.conf
-```
-
-For Linux, by running following command in this folder:
-
-```
-./lightgbm config=train.conf
+"./lightgbm" config=train.conf
 ```
 
 Prediction
@@ -25,14 +19,8 @@ Prediction
 
 You should finish training first.
 
-For Windows, by running following command in this folder:
-
-```
-lightgbm.exe config=predict.conf
-```
-
-For Linux, by running following command in this folder:
+Run the following command in this folder:
 
 ```
-./lightgbm config=predict.conf
+"./lightgbm" config=predict.conf
 ```

examples/multiclass_classification/README.md

@@ -8,16 +8,10 @@ Here is an example for LightGBM to run multiclass classification task.
 Training
 --------
 
-For Windows, by running following command in this folder:
+Run the following command in this folder:
 
 ```
-lightgbm.exe config=train.conf
-```
-
-For Linux, by running following command in this folder:
-
-```
-./lightgbm config=train.conf
+"./lightgbm" config=train.conf
 ```
 
 Prediction
@@ -25,14 +19,8 @@ Prediction
 
 You should finish training first.
 
-For Windows, by running following command in this folder:
-
-```
-lightgbm.exe config=predict.conf
-```
-
-For Linux, by running following command in this folder:
+Run the following command in this folder:
 
 ```
-./lightgbm config=predict.conf
+"./lightgbm" config=predict.conf
 ```

examples/parallel_learning/README.md

@@ -3,21 +3,19 @@ Parallel Learning Example
 
 Here is an example for LightGBM to perform parallel learning for 2 machines.
 
-1. Edit mlist.txt, write the ip of these 2 machines that you want to run application on.
+1. Edit [mlist.txt](./mlist.txt): write the ip of these 2 machines that you want to run application on.
 
-  ```
-  machine1_ip 12400
-  machine2_ip 12400
-  ```
+   ```
+   machine1_ip 12400
+   machine2_ip 12400
+   ```
 
 2. Copy this folder and executable file to these 2 machines that you want to run application on.
 
 3. Run command in this folder on both 2 machines:
 
-   For Windows: ```lightgbm.exe config=train.conf```
+   ```"./lightgbm" config=train.conf```
 
-   For Linux: ```./lightgbm config=train.conf```
-
-This parallel learning example is based on socket. LightGBM also support parallel learning based on mpi.
+This parallel learning example is based on socket. LightGBM also supports parallel learning based on mpi.
 
 For more details about the usage of parallel learning, please refer to [this](https://github.com/Microsoft/LightGBM/blob/master/docs/Parallel-Learning-Guide.rst).

examples/python-guide/README.md

@@ -29,6 +29,7 @@ Examples include:
     - Create data for learning with sklearn interface 
     - Basic train and predict with sklearn interface
     - Feature importances with sklearn interface
+    - Self-defined eval metric with sklearn interface
     - Find best parameters for the model with sklearn's GridSearchCV
 - [advanced_example.py](https://github.com/Microsoft/LightGBM/blob/master/examples/python-guide/advanced_example.py)
     - Set feature names

examples/python-guide/advanced_example.py

@@ -144,7 +144,7 @@ def loglikelood(preds, train_data):
 
 
 # self-defined eval metric
-# f(preds: array, train_data: Dataset) -> name: string, value: array, is_higher_better: bool
+# f(preds: array, train_data: Dataset) -> name: string, eval_result: float, is_higher_better: bool
 # binary error
 def binary_error(preds, train_data):
     labels = train_data.get_label()

examples/python-guide/sklearn_example.py

 # coding: utf-8
 # pylint: disable = invalid-name, C0111
-import lightgbm as lgb
+import numpy as np
 import pandas as pd
+import lightgbm as lgb
+
 from sklearn.metrics import mean_squared_error
 from sklearn.model_selection import GridSearchCV
 
@@ -35,6 +37,27 @@ print('The rmse of prediction is:', mean_squared_error(y_test, y_pred) ** 0.5)
 # feature importances
 print('Feature importances:', list(gbm.feature_importances_))
 
+
+# self-defined eval metric
+# f(y_true: array, y_pred: array) -> name: string, eval_result: float, is_higher_better: bool
+# Root Mean Squared Logarithmic Error (RMSLE)
+def rmsle(y_true, y_pred):
+    return 'RMSLE', np.sqrt(np.mean(np.power(np.log1p(y_pred) - np.log1p(y_true), 2))), False
+
+
+print('Start training with custom eval function...')
+# train
+gbm.fit(X_train, y_train,
+        eval_set=[(X_test, y_test)],
+        eval_metric=rmsle,
+        early_stopping_rounds=5)
+
+print('Start predicting...')
+# predict
+y_pred = gbm.predict(X_test, num_iteration=gbm.best_iteration_)
+# eval
+print('The rmsle of prediction is:', rmsle(y_test, y_pred)[1])
+
 # other scikit-learn modules
 estimator = lgb.LGBMRegressor(num_leaves=31)
 

examples/regression/README.md

@@ -8,16 +8,10 @@ Here is an example for LightGBM to run regression task.
 Training
 --------
 
-For Windows, by running following command in this folder:
+Run the following command in this folder:
 
 ```
-lightgbm.exe config=train.conf
-```
-
-For Linux, by running following command in this folder:
-
-```
-./lightgbm config=train.conf
+"./lightgbm" config=train.conf
 ```
 
 Prediction
@@ -25,14 +19,8 @@ Prediction
 
 You should finish training first.
 
-For Windows, by running following command in this folder:
-
-```
-lightgbm.exe config=predict.conf
-```
-
-For Linux, by running following command in this folder:
+Run the following command in this folder:
 
 ```
-./lightgbm config=predict.conf
+"./lightgbm" config=predict.conf
 ```