How keisya-putri/eksperimen_sml_keisya works
This repository contains a small dataset-preprocessing workflow. Its sampled script cleans a supplied DataFrame, prepares numeric and categorical features, splits the data, fits and applies a transformer, saves a column header and serialized transformer, and returns the split arrays. The tree also contains a survey CSV and an experiment notebook, but their relationship to the script is not established by the available excerpts.
Main entry points: automate_keisya.py Select Features · osmi_features_header.csv Feature Header · preprocessor_osmi.joblib Saved Transformer
Components 10
Other components
- Input DataFrame
- Clean Dataset
preprocessing/automate_keisya.py - Normalize Gender
preprocessing/automate_keisya.py - Select Features
preprocessing/automate_keisya.py - Train Test Split
preprocessing/automate_keisya.py - Fit Transformer
preprocessing/automate_keisya.py - Feature Header
preprocessing/dataset_preprocessing/osmi_features_header.csv - Saved Transformer
preprocessing/dataset_preprocessing/preprocessor_osmi.joblib - Split Arrays
- Experiment Notebook
preprocessing/Eksperimen_keisya.ipynb
Connections 8
- Input DataFrame → Clean Dataset: supplies data
preprocessing/automate_keisya.py - Clean Dataset → Normalize Gender: normalizes gender
preprocessing/automate_keisya.py - Normalize Gender → Select Features: selects features
preprocessing/automate_keisya.py - Select Features → Train Test Split: splits data
preprocessing/automate_keisya.py - Train Test Split → Fit Transformer: fits and transforms
preprocessing/automate_keisya.py - Select Features → Feature Header: writes columns
preprocessing/automate_keisya.py - Fit Transformer → Saved Transformer: saves transformer
preprocessing/automate_keisya.py - Fit Transformer → Split Arrays: returns splits
preprocessing/automate_keisya.py