SanyHe
commented
2 weeks ago Regression - XGBoost - Manual Hyperparameter Selection - Sany Test
Test Result: Successful
Process Step:
Template -> [Section Name: Option]
- Built-in Training Data Option: 1
- Output Data Identifier Column Selection: 1
- World Map Projection: 1
- Distribution in World Map: 3
- Continue World Map: 2
- Data Selection: [2, 5]
- Missing Values Process: 1
- Strategy for Missing Values: 2
- Imputation Method Option: 3
- Feature Engineering: 1
- Feature Engineering: ABC
- Feature Engineering: b * c - d
- Continue Feature Engineering: 2
- Mode Selection: 1
- Data Segmentation - X Set and Y Set: [2, 4]
- Data Segmentation - X Set and Y Set: 5
- Feature Scaling on X Set: 1
- Feature Scaling on X Set: 3
- Feature Selection on X set: 1
- Feature Selection on X set: 1
- Feature Selection on X set: 2
- Data Split - Train Set and Test Set: 0.3
- Model Selection: 9
- Automated Machine Learning: 2
- XGBoost - Hyper-parameters Specification
- N Estimators: 100
- Learning Rate: 0.5
- Max Depth: 4
- Subsample: 0.4
- Colsample Bytree: 1
- Alpha: 0.8
- Lambda: 0.5
Verification Step:
-
Check If the position of ID matches the correct row in Geochemistrypi+ID
- Data under the path .../artifacts/data
- Application Data Feature-Engineering Selected.xlsx ✅
- Application Data Feature-Engineering.xlsx ❌
- Application Data Original.xlsx ✅
- Data Original.xlsx ✅
- Data Selected Dropped-Imputed Feature-Engineering.xlsx ✅
- Data Selected Dropped-Imputed.xlsx ✅
- Data Selected.xlsx ✅
- X Test.xlsx ✅
- X Train.xlsx ✅
- X Without Scaling.xlsx ✅
- Y.xlsx ✅
- Y Train.xlsx ✅
- Y Test.xlsx ✅
- Data under the path .../artifacts/image/map
- Map Projection - Al2O3.xlsx ✅
- Data under the path .../artifacts/image/model_output
- Permutation Importance - Y Test.xlsx ✅
- Predicted vs. Actual Diagram - XGBoost.xlsx ✅
- Residuals Diagram - XGBoost.xlsx ✅
- Data under the path .../artifacts/image/statistic
- Correlation Plot.xlsx ✅
- Distribution Histogram.xlsx ✅
-
check if the rows in Geochemistrypi+ID match to the corresponding one in Geochemistrypi
- Data under the path .../artifacts/data
- Application Data Feature-Engineering Selected.xlsx ✅
- Application Data Feature-Engineering.xlsx ✅
- Application Data Original.xlsx ✅
- Application Data Predicted.xlsx ✅
- Data Original.xlsx ✅
- Data Selected Dropped-Imputed Feature-Engineering.xlsx ✅
- Data Selected Dropped-Imputed.xlsx ✅
- Data Selected.xlsx ✅
- X After Feature Selection.xlsx ✅
- X Test.xlsx ✅
- Y Train.xlsx ✅
- X With Scaling.xlsx ✅
- X Without Scaling.xlsx ✅
- Y Test Predict.xlsx ✅
- Y Test.xlsx ✅
- Y Train Predict.xlsx ✅
- Y Train.xlsx ✅
- Y.xlsx ✅
- Data under the path .../artifacts/image/map
- Map Projection - Al2O3.xlsx ✅
- Data under the path .../artifacts/image/model_output
- Feature Importance - XGBoost.xlsx ✅
- Permutation Importance - X Test.xlsx ✅
- Permutation Importance - Y Test.xlsx ✅
- Predicted vs. Actual Diagram - XGBoost.xlsx ❌
- Residuals Diagram - XGBoost.xlsx ❌
- Data under the path .../artifacts/image/statistic
- Correlation Plot.xlsx ✅
- Distribution Histogram After Log.xlsx ✅
- Distribution Histogram.xlsx ✅
- Probability Plot.xlsx ❌
- Data under the path .../artifacts
- Transform Pipeline Configuration.txt ✅
- Data under the path .../metrics
- Cross Validation - XGBoost.txt ✅
- Model Score - XGBoost.txt ✅
- Data under the path .../parameters
- Hyper Parameters - XGBoost.txt ✅
PotatoXi
The following diagrams is used for markdown.