Machine Learning Bootcamp

Python (2 days)

1. Python for Datascience

Probability and Statistic Skills (6 days)

1. Calculus
2. Linear Algebra
3. Probability
4. Descriptive Statistics
5. Random Variables
6. Hypothesis Testing

Computer Science (1 day)

1. Optimizing Algorithms

Collect and store data (4 days)

1. Data Science project structure.
   2.1. Static Files (csv, json, yml).
   2.2. Web Scraping tools and techniques.
2. How to connect to a SQL database using Python.

Project part I: Involves creating your own dataset from web scraping and storing it on SQL DB.

Data Management (3 days)

1. Exploratory data analysis
2. Data Cleaning
3. Feature Engineering: creating new features from existing features.
4. How to deal with outliers
5. How to deal with missing data
6. Label encoding and Normalization Techniques

Project Part II: Data cleaning

Modeling (6 days)

1. Supervised and Unsupervised Learning
2. Cross Validation (Overfitting vs underfitting)
3. Introduction to the scikit-learn library.
4. Metrics: Measuring your results
5. Model parameters

Machine Learning Algorithms (6 days)

1. Regression Algorithms
2. Classification Algorithms
3. Clustering
4. Hypertuning your machine learning algorithm
5. Time Series Forecasting and Recommender Systems
6. Introduction to Deep Learning.

Storytelling, Communicating your results (2 days)

1. Business intelligence tools
2. Using the ideal graph to show insights
3. Communicating statistics in a simple way

Data Science as Software (1 day)

1. How to create a machine learning web app with Heroku.

Week 12: Real life cases

1. Look for a statistics case that can be solved in paper (finish the solution) and scale it to coding big amounts of data ---> can be the rats problem but reshaped.
2. Look for an image classification good case-----> can be a case applied to medicine (xrays, etc)
3. Look for an NLP, neural networks or recommender system(clustering) good case-----> Look for the new Spotify case in Kaggle competitions.