How paritosh264/denovo works
This repository is a small educational machine-learning toolkit: it implements preprocessing, classification, and regression algorithms as independent Python classes. A typical user supplies data to a model’s fit method and obtains predictions with predict; preprocessing and train/test splitting are also available, but the sampled code does not wire them into a shared pipeline. The graph therefore shows the user-facing components without inventing calls between them. The README is empty, and no unsampled behavior is assumed.
Main entry points: train_test_split.py Train/test split · Normalizer.py Min-max normalizer · StandardScaler.py Standard scaler · KNN.py KNN classifier
Components 10
Preprocessing
- Train/test split
Preprocessing/train_test_split.py - Min-max normalizer
Preprocessing/Normalizer.py - Standard scaler
Preprocessing/StandardScaler.py
Classification
- KNN classifier
Classifiers/KNN.py - Logistic regression
Classifiers/Logistic_regression.py - NaiveBias stub
Classifiers/Naive_bias.py
Regression
- Linear regression
Regression/Lr.py - Polynomial regression
Regression/Polynomial_lr.py
Other components
- Data scientist
- Scratch models
scratch_ml.py
Connections 9
- Data scientist → Train/test split: uses
Preprocessing/train_test_split.py - Data scientist → Min-max normalizer: uses
Preprocessing/Normalizer.py - Data scientist → Standard scaler: uses
Preprocessing/StandardScaler.py - Data scientist → KNN classifier: fits and predicts
Classifiers/KNN.py - Data scientist → Logistic regression: fits and predicts
Classifiers/Logistic_regression.py - Data scientist → NaiveBias stub: instantiates
Classifiers/Naive_bias.py - Data scientist → Linear regression: fits and predicts
Regression/Lr.py - Data scientist → Polynomial regression: fits and predicts
Regression/Polynomial_lr.py - Data scientist → Scratch models: uses
scratch_ml.py