lhayana / datascience2machinelearning

Data Science & Machine Learning Course

• Lesson 1

  • Motivation
  • Provocations: hw, data, internet, AI, cloud computing
  • Tendencies
  • Platforms
  • References

• Lesson 2

  • Conda essentials
  • Platforms (jupyter, jupyterlab, colab)
  • Python Crash Course

• Lesson 3

  • Modules, Iterations, List Comprehesion
  • String and date operations
  • Introducting to Object-Oriented Programming (OOP)

• Lesson 4

  • Introduction to Numpy
  • Introduction to Pandas

• Lesson 5

  • Data Cleaning Basic

• Lesson 6

  • Exploratory Data Analysis I
  • Matplotlib
  • Line, Bar and Scatter Plots

• Lesson 7

  • Exploratory Data Analysis II
  • Histogram and Box Plots
  • Wrapper from Pandas to Matplotlib

• Lesson 8

  • Exploratory Data Analysis III
  • Case study: gender gap
  • Aesthetics
  • Colors, Lines width
  • Annotations

• Lesson 9

  • Exploratory Data Analysis IV
  • Case study: titanic
  • Visualizing missing values
  • Aggregate data using pivot table
  • Storytelling from Seaborn

• Lesson 10

  • Exploratory Data Analysis V
  • Visualizing geographical data
  • Working with basemap
  • Customizing the plot
  • Folium
  • Maps with markers
  • Maker clusters -Heatmap

• Lesson 11

  • Exploratory Data Analysis VI
  • Case Study #1 - Jonh Snow Map
  • Case Study #2 - Open Data Natal-RN

• Lesson 12

  • Exploratory Data Analysis VII
  • Case study: IBGE
  • Geojson
  • Importing files
  • Creating maps
  • Choropleths maps

• Lesson 13

  • Case study: NYC open data (education)
  • Data cleaning walkthrough
  • Combining data
  • Groupby
  • Merge (inner, outer, right, left)

• Lesson 14

  • Sampling
    • Population and sampling
    • Sampling error
    • Simple random sampling (SRS)
    • Stratified sampling
    • Clustering sampling
  • Variables in statistics
    • Quantitative and qualitative variables
    • Scale of measurements (nominal, ordinal, interval, ratio)

• Lesson 15

  • Frequency Distributions
    • Sorting frequency distribution tables
    • Percentiles and percentiles ranks
    • Information loss
  • Visualizing Distributions
    • Bar, Pie, Histograms plots
    • Skewed distributions
    • Symmetrical Distributions
  • Comparing Frequency Distribution

• Lesson 16

  • A brief history of AI
  • Key definitions
  • Types of Machine Learning
  • Machine Learning Workflow
  • Main challenges
  • End-to-end ML project

• Lesson 17

  • Univariate KNN
    • Euclidean distance for univariate
    • Function to make predictions
    • Error metrics
  • Multivariate KNN
    • Normalize columns
    • Euclidean distance for multivariate
  • Hyperparameter optimization
  • Cross-Validation

• Lesson 18

  • Linear Regression (one variable)
  • Cost function
  • Gradient descent
  • Refresher on linear algebra concepts
  • Linear Regresion (multiple variables)

• Lesson 19

  • Classification
  • Binary Classification
  • Decision Boundary
  • Cost Function
  • Multiclass Classification
  • Regularization
  • Hands on Scikit

• Lesson 20

  • Clustering Basic
  • K-Means
  • Case study: senators votes, nba

• Lesson 21

  • Introduction to Decision Tree
  • Converting categorical variables
  • Splitting Data
  • Decision Trees as flows of data
  • Entropy
  • Information gain
  • Applying Decision Trees
  • Overfitting problem

• Lesson 22

  • Ensembles (Random Forest)
  • Combining predictions
  • Why Ensembling works
  • Introduction variation with bagging and random features
  • Reducing overfitting using Random Forest
  • Case study: US Census, predicting bike rentals

• Lesson 23

  • Getting Started with Kaggle
  • Feature Preparation, Selection and Engineering
  • Model Selection and Tuning
  • Creating a Kaggle Workflow

• Lesson 24

  • Deep Learning Fundamentals I
  • Representing neural network
  • Nonlinear activation functions
  • Hidden Layers
  • Case study: build a handwritten digit classified

• Lessson 25

  • Deep Learning Fundamentals II
  • Mathematical building blocks of neural networks
  • Getting started with neural networks
  • Classifying movie reviews: a binary classification example
  • Classifying newswires: a multiclass classification problem
  • Predicting houses price: a regression problem

• Lesson 26

  • Deep Learning Fundamentals III
  • Formal evaluation procedures for machine learning models
  • Preparing data for deep learning
  • Feature engineering
  • Tackling overfitting -The universal workflow for approaching machine learning problems
  • Case study: titanic