#2020-07-18 12:53:47
If you like manual customization and lots of technical hyper parameters, SVMs might be the algorithm for you. They come with built-in Kernel implementation and have been quite popular in recent years.
They are great for stepping-up your technical data and analysis knowledge and promise a great range of classification and regression applications though sometimes, at the cost of computation resources.
#SUPERVISED
#CLASSIFICATION
#REGRESSION
Support Vector Machine
- Data Size: Small
- Speed: Fair
- Ease of Use: Hard
- Normalization: No
- Predictor: Numeric or Categorical
- Primary Problem: Binary
- Mixed-type: No
- Missing Data Handling: No
- Popularity: 75%
support vector machines tutorials in python
1- Simple Implementation
This Machine Learning implementation demonstrates a simple and straightforward Support Vector Machine implementation for a general audience. Sometimes it's helpful to start with taking an easy, practical and top-down step.
2- Step by Step
A step-by-step explanation of the simple support vector machine implementation that's previously introduced.
3- Optimization
Optimization is generally a big part of Machine Learning however, with Support Vector Machines it's even more so. There are many settings and fine-tuning involved that must be handled with care when working with SVMs.
4- Pros & Cons
SVM pros & cons are crucial to know because there are plenty to name on the both side of the list. This list will help you understand advantages and disadvantages that come with Support Vector Machines, help you make strategic Data Science decision and compliment your Machine Learning knowledge.
5- History
Read about the history of Support Vector Machine algorithms. It's a modern story that makes your respect the innovation process that's usually involved in such developments.
