How aaryan-sharmaa/100daysofml works
This repository is a hands-on machine-learning curriculum, not a single deployable application. Its principal user journey follows the README roadmap from data handling through model training and evaluation to advanced methods. The tree supports the named lecture directories; most notebook contents were not sampled, so the graph represents documented learning stages rather than inferred calls between notebooks.
Main entry points: lecture15_csv_practice Data handling · lecture19_data_understanding_practice Data understanding · lecture24_feature_scaling_standardization Feature preparation · lecture50_51_simple_linear_regression Regression
Components 15
Data preparation
- Data handling
lecture15_csv_practice - Data understanding
lecture19_data_understanding_practice - Feature preparation
lecture24_feature_scaling_standardization
Supervised learning
- Regression
lecture50_51_simple_linear_regression - Classification
lecture70_75_logistic_regression - Model evaluation
lecture76_79_classification_metrics
Model methods
- Dimensionality reduction
lecture47_49_principal_component_analysis - KNN
lecture91_KNN - SVM
lecture92_96_SVM - Decision trees
lecture97_100_decision_trees - Ensemble learning
lecture102_104_voting_ensemble
Advanced topics
- Clustering
lecture128_130_kmeans - Imbalanced data
lecture133_imbalanced_data_smote - Hyperparameter optimization
lecture134_optuna
Other components
- Learner
Connections 14
- Learner → Data handling: studies
- Data handling → Data understanding: progresses to
- Data understanding → Feature preparation: progresses to
- Feature preparation → Regression: progresses to
- Regression → Classification: progresses to
- Classification → Model evaluation: progresses to
- Model evaluation → Dimensionality reduction: progresses to
- Dimensionality reduction → KNN: progresses to
- KNN → SVM: progresses to
- SVM → Decision trees: progresses to
- Decision trees → Ensemble learning: progresses to
- Ensemble learning → Clustering: progresses to
- Clustering → Imbalanced data: progresses to
- Imbalanced data → Hyperparameter optimization: progresses to