💡 Machine Learning Zoomcamp 💡

This repo has been created while attending ML ZoomCamp organized by Alexey Grigorev) and team. This ZoomCamp is spread across 14 modules for upto 4 months.

Table of Contents

1. Open Ended Learning
2. Important Links & Organization
3. Week-1 learning
4. Week-2 learning
5. Week-3 learning
6. Week-4 learning
7. Week-5 learning
8. Week-6 learning
9. Week-7 learning
10. Week-midterm-project
11. Week-8 learning

📖 Targeted Open Learning

Going forward, I'm going to update this README.md on a weekly basis with an Overview for the week, Notes prepared during the week, Outcomes from the learning process, and, maybe, social Media posts that I write about the weekly outcome. Good Luck and Happy Learning!

🔗 Important Links & Organization:

• Files for ml-zoomcamp cohort 2022
• Syllabus and deadlines

🗃️ Organization of this repo

• data folder is collection of datasets used in different weeks during this zoomcamp
• images folder is simply a collection of snapshots created during preparing of weekly notes.
• ipynb folder contains all the notebooks created in different weeks of learning.
• models is collection of ML models developed during the course.
• notes folder is a set of readme files that contain notes from different weeks of the zoomcamp.
• src folder is collection of python scripts developed/used during different weeks of the course.

🗓️ Week-1

• Topic: Introduction to Machine Learning
• Notes: Week-1 Notes
• Outcomes: ML termilogy, rule vs ML, supervisedML, CRISP-DM, Model selection steps, Review of linear algebra and env setup
• Media: Week-1 Post

🗓️ Week-2

• Topic: Machine Learning for Regression
• Notes: Week-2 Notes
• Outcomes: Linear Regression, Car Price Prediction model, House Price Prediction model, Regularization, Shuffle data
• Media: Week-2 Post

🗓️ Week-3

• Topic: Machine Learning for Classification
• Notes: Week-3 Notes
• Outcomes: Binary Classification, Feature Importance, OneHot Encoding, Ouput of log reg - probability
• Media: Week-3 Post

🗓️ Week-4

• Topic: Evaluation Metrics for Classification
• Notes: Week-4 Notes
• Outcomes: Metrics for Binary Classification, Accuracy, Class Imbalance in Binary classification problems, Precision, Recall, ROC, ROC-AUC, K-Fold CV, F1-score - other metrics
• Media: Week-4 Post

🗓️ Week-5

• Topic: Deploying Machine Learning Models
• Notes: Week-5 Notes
• Outcomes: Save and Load ML models, Convert jupyter notebook into python script, Environment dependencies and management, Python level - pipenv, virtualenv, conda, etc., OS level - Docker
• Media: Week-5 Post

🗓️ Week-6

• Topic: Decision Trees and Ensemble Learning
• Notes: Week-6 Notes
• Outcomes: Decision Trees, Importannt parameters to control overfitting of DTs, Random Forest as combination of multiple DTs, Diversity of DTs for RF is important, Gradient Boosting is sequentially stiching DTs, Each DT model fixes errors from precious model, XGBoost - an efficient implementation of gradient boosting
• Media: Week-6 Post-1 Week-6 Post-2 Week-6 Post-3 Week-6 Post-4

🗓️ Week-7

• Topic: BentoML for Production
• Notes: Week-7 Notes
• Outcomes: Using BentoML for putting ML models into production
• Media: Week-7 Post

🗓️ Week-midterm

• Topic: Midterm Project
• Link to Project folder: Week-midterm-project
• Outcomes: End-End ML project
• Media: Week-midterm Post

🗓️ Week-8

• Topic: Neural Networks and Deep Learning
• Notes: Week-8 Notes
• Outcomes: Pre-Trained models, NN layers, learning rate, dropout rate, checkpoints etc.
• Media: Week-8 Post-1, Week-8 Post-2

🗓️ Week-9

• Topic: Serverless Deep Learning
• Notes: Week-9 Notes
• Outcomes: AWS Lambda, TensorFlow lite, Keras to TFLite, docker, API Gateway
• Media: Week-9 Post-1