This repository provides a complete beginner-to-advanced journey into Machine Learning, covering both theoretical foundations and practical implementation using Python. Each section is modular and organized into individual notebooks that you can open directly in Google Colab for hands-on practice.
| No. | Notebook Name | Description |
|---|---|---|
| ✅ 01 | 01_ML_Libraries.ipynb |
Introduction to essential ML libraries: NumPy, Pandas, Matplotlib, Seaborn, Scikit-learn, TensorFlow. |
| ✅ 02 | 02_Data_Statistics.ipynb |
Exploratory Data Analysis (EDA), summary statistics, skewness, kurtosis, and distribution understanding. |
| ✅ 03 | 03_Data_Visualisation.ipynb |
Graphical EDA using Matplotlib and Seaborn – histograms, pair plots, heatmaps, and box plots. |
| ✅ 04 | 04_Data_Pre_Processing.ipynb |
Data cleaning, handling missing values, encoding, scaling, and transformations. |
| ✅ 05 | 05_Feature_Selection.ipynb |
Feature importance, SelectKBest, RFE, PCA, and Lasso for dimensionality reduction. |
| ✅ 06 | 06_Performance_Evaluation.ipynb |
Model validation using train-test-split, K-Fold, and cross-validation techniques. |
| ✅ 07 | 07_Performance_Metrics.ipynb |
Accuracy, precision, recall, F1-score, ROC curve, confusion matrix for classification tasks. |
| ✅ 08 | 08_Classification_Algorithms.ipynb |
Implementation of classifiers: Logistic Regression, SVM, KNN, Decision Trees, Naive Bayes, Random Forest. |
| ✅ 09 | 09_Regression_Algorithms.ipynb |
Implementation of regression models: Linear, Ridge, Lasso, ElasticNet, SVR, Decision Trees, and ensemble regressors. |
| ✅ 10 | 10_Image_Classification.ipynb |
Deep learning with CNNs using Keras: training MNIST digit classifier with accuracy visualization. |
| ✅ 11 | 11_Multiclass_Image_Classification.ipynb |
Multiclass classification using the Iris dataset: training and evaluating models with Scikit-learn classifiers. |
| ✅ 12 | 12_Comparing_ML_Algorithms.ipynb |
Systematic comparison of multiple ML models with tabular results, boxplots, and performance observations. |
| ✅ 13 | 13_Pipeline_Automation.ipynb |
Using Pipeline and FeatureUnion for combining preprocessing, feature selection, and modeling. |
| ✅ 14 | 14_Save_N_Load_ML_Models.ipynb |
Model persistence using Pickle: saving and loading trained models for reuse and deployment. |
To use these notebooks interactively without any setup, click below:
- Understand both theory and application of machine learning algorithms.
- Learn step-by-step how to preprocess data, build pipelines, and evaluate models.
- Gain real experience through hands-on projects including image classification.
- Build a strong foundation for advanced ML or deep learning courses.
Satwik Sai Prakash Sahoo
This project is licensed under the MIT License.