About this Analysis
The purpose of this regression analysis is to predict the performance of CPU of different computer machines based on certain parameters. The data set is obtained from UCI Machine Learning Laboratory.
Description about the data is provided below. Detailed source code can be found on my github,
Abstract: Relative CPU Performance Data, described in terms of its cycle time, memory size, etc
Data Set Information:
The estimated relative performance values were estimated by the authors using a linear regression method. See their article (pp 308-313) for more details on how the relative performance values were set.
Load the data
Exploratory Analysis and cleaning
1)Ordinary Least Squares Regression
2)Stepwise Linear Regression
3)Principal Component Regression
4)Partial Least Squares Regression
B)Penalized Linear regression
2)Least Absolute Shrinkage and Selection Operator(LASSO)regression
1)Multivariate Adaptive Regression Splines(MARS)
2)Support Vector Machine
D)Decision Trees for Regression
1)Classification and Regression Trees (CART)
2)Conditional Decision Trees
7)Gradient Boosted Machine
What you will learn in this course?
• Linear Algebra (Course 1)-Transformation Matrix, Introduction to Linear Algebra and to Mathematics for Machine Learning, Matrices in Linear Algebra: Objects that operate on Vectors, Matrices make linear mappings, Eigenvalues and Eigenvectors: Application to Data Problems.
• Multivariate Calculus (Course 2)- Linear Regression, Vector Calculus, Gradient Descent, Introduction of calculus, Multivariate Calculus, Multivariate chain rule and its applications, Taylor series and linearization, Intro to optimization, Regression.
• Principal Component Analysis(PCA) (Course 3)- Python Programming, Projection Matrix, Mathematical Optimization, Statistics of Datasets, Inner Products, Orthogonal Projections, Principal Component Analysis.
About this course
These courses aim to get you up and to speed in the underlying mathematics, building an intuitive understanding, and relating it to Machine Learning and Data Science. At the end of all three courses, you will gain the prerequisite mathematical knowledge to continue your journey and take more advanced courses in machine learning. It has three parts as follows-
Linear Algebra-You’ll learn how linear algebra is related to vectors and matrices, what they are and how they work, also includes the solving methods of difficult problem using eigenvalues and eigenvectors.In the end, you will learn an intuitive understanding of vectors and matrices which help in linear algebra problems and application of these concepts to machine learning. Moreover, you’ll be able to write code blocks and encounter Jupyter notebooks in Python.
Multivariate Calculus- You’ll learn the formulation of a slope before converting intothe formal definition of the gradient of a function,how to build up a set of tools for making calculus easier and faster, uses of calculus to build approximations to functions and application in linear regression models. In the end, you’ll learn an intuitive understanding of calculus and see how calculus used in neural networks.
Principal Component Analysis -This intermediate-level course requiresmore programming (i.e. Basic Python) than the other two courses (discussed above) and Numpy knowledge. Itintroduces the mathematical foundations to derive (PCA), you’ll learn fundamental dimensionality reduction technique, usage of the mathematics from the first two courses to compress high-dimensional data, basic statistics of data sets,uses of inner products, creation of orthogonal projections of data onto lower-dimensional subspaces. In the end, you'll be familiar with important mathematical concepts and implementation of PCA.
Click the link or image below to access the course contents: