Tetaddjik6hgj7otuejd
3 DAY COURSE

Fast Track to Machine Learning with Louis Dorard

Topics covered at MACHINE-LEARNING-01-03
View Schedule & Book More dates available

Next up:

Are you ready to demystify Machine Learning and learn how to put it to work on real problems? Kick start your skills in this hands on workshop and learn the secrets of this powerful method for data analysis!

Join Louis Dorard for this Machine Learning course and gain the skills to integrate ML into your applications, using the most modern and widely used techniques in the industry. In this workshop, we'll demystify Machine Learning, you'll gain an understanding of its possibilities and limitations, and how to put it to work on real problems.

As Vinod Khosla said, “in the next 20 years, ML will have more impact than mobile has.” In this course, we'll demystify Machine (and Deep) Learning, you'll gain an understanding of its possibilities and limitations, and how to put it to work on real problems.

You'll learn to prepare data, to create ML models, to evaluate them in your domain of application, to optimize them, and to deploy them. We'll adopt a top-down, results-first and experimentation-driven approach, where we'll focus on practical techniques applied to concrete examples.

If you're interested in leveraging the power of machine learning to improve your applications, then make sure to book now!

Learn how to:

  • Understand the possibilities and limitations of Machine Learning, and know how to formulate your own ML problem
  • Inspect, wrangle and prepare data for ML
  • Understand the main ideas behind modern learning techniques based on decision trees (random forests and gradient boosting) and on neural networks
  • Build predictive models from data and interpret them
  • Analyse models' errors and performance, optimise them, and deploy to production with APIs

About the Author

Day 1: ML basics

Introduction to Machine Learning

  • [Slides] Key ML concepts and terminology
  • [Slides] Possibilities and example use cases
  • [Slides + Exercise] Formalizing your own ML problem

Model creation

  • [Slides] Learning techniques: Nearest Neighbors, Linear Models, Decision Trees
  • [Hands-on] Introduction to Jupyter notebooks
  • [Hands-on + Exercise] Creating and interpreting Decision Trees with BigML (ML-as-a-Service tool) and scikit-learn (open source ML library)

Evaluation

  • [Slides] Performance criteria for ML models, evaluation procedure, and error interpretation
  • [Slides + Exercise] Aggregate metrics for regression: MAE, MSE, R-squared, MAPE
  • [Hands-on + Exercise] Evaluating models with BigML and scikit-learn
  • [Slides + Hands-on] Aggregate metrics for classification: accuracy, confusion and cost matrices, precision, recall, F-measure

Operationalization

  • [Slides] Functioning of REST APIs and importance in the context of real-world ML
  • [Hands-on] Demo of cloud APIs: BigML and Indico
  • [Hands-on] Deploying your own Python models as APIs with Microsoft Azure ML and/or Flask (open source library)
  • [Slides] Critical overview of open source and cloud ML products and deployment solutions

Day 2: Improving your usage of ML

Better models: ensembles and cross-validation

  • [Slides] Improving predictions with Ensembles; application to Decisions Trees: Random Forests
  • [Exercise] Experimenting with multiple evaluations of different learning techniques
  • [Slides + Hands-on] Evaluating models with cross-validation

Better classifiers: probabilities and thresholding (aka "soft-classification")

  • [Slides] Building class probabilities for Linear Models (Logistic Regression) and Decision Tree-based models
  • [Slides] Performance metrics: log-loss and AUC
  • [Hands-on + Exercise] Comparing models efficiently, tuning soft-classifiers' thresholds and finding trade-offs between competing performance metrics

Data preparation

  • [Slides] Limitations of ML
  • [Slides + Exercise] Feature engineering
  • [Hands-on + Exercise] Data inspection techniques
  • [Hands-on + Exercise] Data wrangling with Dataiku's Data Science Studio and/or Pandas (open source library)

ML on text — Natural Language Processing

  • [Slides + Hands-on] Text pre-processing tips; feature extraction (bag of words and n-grams) and feature selection; application with NLTK (open source library) and scikit-learn
  • [Exercise] Creating models from numerical, categorical and textual features; evaluating them; submitting to Kaggle

Day 3: Going further with ML

Better models: boosting and parameter tuning

  • [Hands-on] Model creation pipelines in scikit-learn and grid search
  • [Slides + Hands-on] Comparing randomized and "smart" parameter search with Hyperopt (open source library)
  • [Slides + Hands-on] Improving predictions with Gradient Boosting and Xgboost
  • [Exercise] Combining all techniques to create better models

Neural Networks and Deep Learning

  • [Slides] Perceptron algorithm (single- + multi-layered)
  • [Hands-on] Neural Network training and predictions with open source libraries TensorFlow, Keras and scikit-learn
  • [Slides + Hands-on] Introduction to Transfer Learning; application to images with Indico's feature extraction API for images

Other ML problems

  • [Slides + Hands-on] Unsupervised learning: clustering (k-means) and anomaly detection (isolation forests); visualizations on BigML
  • [Slides + Hands-on] Time series prediction: lag features and reduction to regression
  • [Slides] Recommender systems: reduction to classification and collaborative filtering

Getting ready to integrate ML into your applications / projects (if time allows)

  • [Slides] Specifying ML systems with the Machine Learning Canvas
  • [Slides] Example canvases
  • [Exercise] Application to your own ideas

Conclusions

  • Key take-aways
  • Current trends in applied ML
  • Resources to go further and customized suggestions

Audience

Prerequisites

A general understanding of undergraduate level maths such as algebra and statistics will be beneficial for this course.

Prepare

Please consult Codeacademy's Learn Python and Robert Johansson's Introduction to Python programming (in particular the following sections: Python program files, Modules, Assignment, Fundamental types, Control Flow et Functions) to learn or revise Python's basics.

Bring your own hardware

To benefit most from this Machine Learning course, please bring your own laptop, so you can develop with your own tools and languages, rather than something you are not familiar with.

If you are not able to bring in your own laptop, please contact the Skills Matter team on +44 207 1839040 or email info@skillsmatter.com

Please submit all laptop requests a minimum of 48 hours prior to the course as laptops are subject to availability.

Overview

Are you ready to demystify Machine Learning and learn how to put it to work on real problems? Kick start your skills in this hands on workshop and learn the secrets of this powerful method for data analysis!

Join Louis Dorard for this Machine Learning course and gain the skills to integrate ML into your applications, using the most modern and widely used techniques in the industry. In this workshop, we'll demystify Machine Learning, you'll gain an understanding of its possibilities and limitations, and how to put it to work on real problems.

As Vinod Khosla said, “in the next 20 years, ML will have more impact than mobile has.” In this course, we'll demystify Machine (and Deep) Learning, you'll gain an understanding of its possibilities and limitations, and how to put it to work on real problems.

You'll learn to prepare data, to create ML models, to evaluate them in your domain of application, to optimize them, and to deploy them. We'll adopt a top-down, results-first and experimentation-driven approach, where we'll focus on practical techniques applied to concrete examples.

If you're interested in leveraging the power of machine learning to improve your applications, then make sure to book now!

Learn how to:

  • Understand the possibilities and limitations of Machine Learning, and know how to formulate your own ML problem
  • Inspect, wrangle and prepare data for ML
  • Understand the main ideas behind modern learning techniques based on decision trees (random forests and gradient boosting) and on neural networks
  • Build predictive models from data and interpret them
  • Analyse models' errors and performance, optimise them, and deploy to production with APIs

About the Author

Program

Day 1: ML basics

Introduction to Machine Learning

  • [Slides] Key ML concepts and terminology
  • [Slides] Possibilities and example use cases
  • [Slides + Exercise] Formalizing your own ML problem

Model creation

  • [Slides] Learning techniques: Nearest Neighbors, Linear Models, Decision Trees
  • [Hands-on] Introduction to Jupyter notebooks
  • [Hands-on + Exercise] Creating and interpreting Decision Trees with BigML (ML-as-a-Service tool) and scikit-learn (open source ML library)

Evaluation

  • [Slides] Performance criteria for ML models, evaluation procedure, and error interpretation
  • [Slides + Exercise] Aggregate metrics for regression: MAE, MSE, R-squared, MAPE
  • [Hands-on + Exercise] Evaluating models with BigML and scikit-learn
  • [Slides + Hands-on] Aggregate metrics for classification: accuracy, confusion and cost matrices, precision, recall, F-measure

Operationalization

  • [Slides] Functioning of REST APIs and importance in the context of real-world ML
  • [Hands-on] Demo of cloud APIs: BigML and Indico
  • [Hands-on] Deploying your own Python models as APIs with Microsoft Azure ML and/or Flask (open source library)
  • [Slides] Critical overview of open source and cloud ML products and deployment solutions

Day 2: Improving your usage of ML

Better models: ensembles and cross-validation

  • [Slides] Improving predictions with Ensembles; application to Decisions Trees: Random Forests
  • [Exercise] Experimenting with multiple evaluations of different learning techniques
  • [Slides + Hands-on] Evaluating models with cross-validation

Better classifiers: probabilities and thresholding (aka "soft-classification")

  • [Slides] Building class probabilities for Linear Models (Logistic Regression) and Decision Tree-based models
  • [Slides] Performance metrics: log-loss and AUC
  • [Hands-on + Exercise] Comparing models efficiently, tuning soft-classifiers' thresholds and finding trade-offs between competing performance metrics

Data preparation

  • [Slides] Limitations of ML
  • [Slides + Exercise] Feature engineering
  • [Hands-on + Exercise] Data inspection techniques
  • [Hands-on + Exercise] Data wrangling with Dataiku's Data Science Studio and/or Pandas (open source library)

ML on text — Natural Language Processing

  • [Slides + Hands-on] Text pre-processing tips; feature extraction (bag of words and n-grams) and feature selection; application with NLTK (open source library) and scikit-learn
  • [Exercise] Creating models from numerical, categorical and textual features; evaluating them; submitting to Kaggle

Day 3: Going further with ML

Better models: boosting and parameter tuning

  • [Hands-on] Model creation pipelines in scikit-learn and grid search
  • [Slides + Hands-on] Comparing randomized and "smart" parameter search with Hyperopt (open source library)
  • [Slides + Hands-on] Improving predictions with Gradient Boosting and Xgboost
  • [Exercise] Combining all techniques to create better models

Neural Networks and Deep Learning

  • [Slides] Perceptron algorithm (single- + multi-layered)
  • [Hands-on] Neural Network training and predictions with open source libraries TensorFlow, Keras and scikit-learn
  • [Slides + Hands-on] Introduction to Transfer Learning; application to images with Indico's feature extraction API for images

Other ML problems

  • [Slides + Hands-on] Unsupervised learning: clustering (k-means) and anomaly detection (isolation forests); visualizations on BigML
  • [Slides + Hands-on] Time series prediction: lag features and reduction to regression
  • [Slides] Recommender systems: reduction to classification and collaborative filtering

Getting ready to integrate ML into your applications / projects (if time allows)

  • [Slides] Specifying ML systems with the Machine Learning Canvas
  • [Slides] Example canvases
  • [Exercise] Application to your own ideas

Conclusions

  • Key take-aways
  • Current trends in applied ML
  • Resources to go further and customized suggestions
Audience

Audience

Prerequisites

A general understanding of undergraduate level maths such as algebra and statistics will be beneficial for this course.

Prepare

Please consult Codeacademy's Learn Python and Robert Johansson's Introduction to Python programming (in particular the following sections: Python program files, Modules, Assignment, Fundamental types, Control Flow et Functions) to learn or revise Python's basics.

Bring your own hardware

To benefit most from this Machine Learning course, please bring your own laptop, so you can develop with your own tools and languages, rather than something you are not familiar with.

If you are not able to bring in your own laptop, please contact the Skills Matter team on +44 207 1839040 or email info@skillsmatter.com

Please submit all laptop requests a minimum of 48 hours prior to the course as laptops are subject to availability.