Oreilly - Hands-On Problem Solving for Machine Learning
by Rudy Lai | Publisher: Packt Publishing | Release Date: March 2019 | ISBN: 9781789530087
Intuitive strategies to deal with messy data, weak models, and leaky machine-learning pipelinesAbout This VideoResolve challenges in supervised learning: misbehaving classifiers and wrong regressors.Practical solutions for building production-ready machine-learning pipelines that don't breakIntuition-driven practical tour through machine learning, packed with step-by-step instructions, working examples, and helpful advice.In DetailMachine learning is all the rage, and you have been tasked with creating models for your business. What looked simple on the surface quickly becomes a nightmare of messy data and non-performing models. What do you do?Hands-On Problem Solving for Machine Learning is packed with intuitive explanations of how machine learning works so that you can fix your models when they break. It presents a wide array of practical solutions for your machine learning pipeline, whether you are working with images, text, or numbers. You'll get a real feel for how to tackle challenges posed during regression and classification tasks.If you want to move past calling simple machine learning libraries, and start solving machine learning problems with real-world messy data, this course is for you!All the code and supporting files for this course are available on GitHub at - https://github.com/PacktPublishing/Machine-Learning-Problems-Solved-V-Downloading the example code for this course: You can download the example code files for all Packt video courses you have purchased from your account at http://www.PacktPub.com. If you purchased this course elsewhere, you can visit http://www.PacktPub.com/support and register to have the files e-mailed directly to you.
- Chapter 1 : Working with Machine Learning
- The Course Overview 00:02:09
- Goals and Variations in Machine Learning 00:10:05
- Installing WinPython and Using Jupyter Notebooks 00:08:39
- Exploring Your Data Using Pandas 00:09:30
- Chapter 2 : Data Wrangling
- Types of Messy Data and How to Clean Them 00:09:38
- Parsing Timestamps and Splitting Columns 00:09:15
- Loading Data from Excel, CSVs, and SQL 00:08:28
- Chapter 3 : Linear Regression — Predict Median Living Costs
- Understanding Linear Regression 00:08:45
- Implementing Linear Regression with Scikit-learn 00:07:47
- Troubleshooting Linear Regression 00:07:58
- Chapter 4 : Logistic Regression — Classify Flowers
- Exploring and Cleaning the Plants Dataset 00:09:49
- Understanding Logistic Regression 00:07:22
- Implementing Train-Test-Splits and Logistic Regression 00:08:25
- Chapter 5 : Predicting the Future
- Build a Robust Model with Cross Validation 00:08:59
- Create Complex Models with Scikit-learn Pipelines 00:09:20
- Find the Best Model with Hyperparameter Search 00:08:19
- Chapter 6 : Diagnosing Issues with Models
- Understanding Our Accuracy in Predicting Numbers 00:09:29
- Assessing Our Correctness in Predicting Labels 00:07:13
- Dealing with Overfitting Using Regularization 00:09:36