100 Days of Machine Learning

My journey through machine learning — from data fundamentals to deep learning.

Progress

46 / 100 days complete

Structure

├── phase-1-data-foundations/     # Days 1-15
├── phase-2-classical-ml/         # Days 16-40
└── phase-3-deep-learning/        # Days 41-60 (in progress)

Log

Phase 1: Data Foundations (Days 1-15)

Day	Topic	Dataset
1	NumPy & Pandas Fundamentals	Synthetic
2	Missing Data & Imputation	Healthcare
3	Distribution Analysis	Retail Sales
4	Outlier Detection	Credit Card Transactions
5	Feature Scaling	Mixed Numerical
6	Encoding Categoricals	Employee Data
7	Feature Engineering	E-commerce
8	EDA Workflow	Airbnb Listings
9	Correlation Analysis	Boston 311 Requests
10	Dimensionality Reduction (PCA)	Wine Quality
11	Data Cleaning Pipeline	Messy Sales Data
12	Time Series Basics	Stock Prices
13	Geospatial Data	NYC Taxi
14	Data Integration	Multi-source Merge
15	Capstone: Rain Prediction	Australian Weather

Phase 2: Classical ML (Days 16-40)

Day	Topic	Dataset
16	Linear Regression	California Housing
17	Polynomial & Regularization	California Housing
18	Logistic Regression	Bank Marketing
19	Decision Trees	Bank Marketing
20	Random Forest	Bank Marketing
21	Gradient Boosting (XGBoost)	Bank Marketing
22	Support Vector Machines	Bank Marketing
23	Cross-Validation Deep Dive	Various
24	K-Nearest Neighbors	Pima Diabetes
25	Naive Bayes	SMS Spam
26	K-Means Clustering	Mall Customers
27	DBSCAN	Mall Customers
28	Hierarchical Clustering	Mall Customers
29	Anomaly Detection	Credit Card Fraud
30	Gaussian Mixture Models	Synthetic
31	Model Showdown	Heart Disease
32	Hyperparameter Tuning	Heart Disease
33	Stacking & Voting Ensembles	Heart Disease
34	Feature Engineering Advanced	Heart Disease
35	Imbalanced Learning	Credit Card Fraud
36	SHAP Interpretability	German Credit
37	Pipelines Masterclass	Melbourne Housing
38-40	Capstone: Loan Default Prediction	Lending Club

Phase 3: Deep Learning (Days 41-60)

Day	Topic	Framework	Dataset
41	Neural Network Fundamentals	PyTorch	Moons, Dry Bean
42	Optimizers & Regularization	PyTorch	HTRU2 Pulsar Stars
43	CNNs	PyTorch	Intel Image
44	Transfer Learning	TensorFlow	Intel Image
45	CNN Architectures	PyTorch	Intel Image
46	Data Augmentation	TensorFlow	Intel Image
47	Object Detection	PyTorch	—
...	...	...	...

Capstone Projects

Phase 1: Rain Prediction (Day 15)

Binary classification predicting next-day rain using Australian weather data.

Phase 2: Loan Default Prediction (Days 38-40)

End-to-end ML project: EDA, feature engineering, model tuning, SHAP interpretability, production pipeline.

• Standalone repo

Tech Stack

• Languages: Python
• ML: scikit-learn, XGBoost, LightGBM
• Deep Learning: PyTorch, TensorFlow/Keras
• Data: pandas, NumPy
• Visualization: matplotlib, seaborn
• Interpretability: SHAP

Notebooks

All notebooks are designed for Google Colab with GPU support.