你是否好奇数据可以告诉你什么？你是否想在关于机器学习促进商业的核心方式上有深层次的理解？你是否想能同专家们讨论关于回归，分类，深度学习以及推荐系统的一切？在这门课上，你将会通过一系列实际案例学习来获取实践经历。在这门课结束的时候，

Loading...

来自 华盛顿大学 的课程

机器学习基础：案例研究

6933 评分

你是否好奇数据可以告诉你什么？你是否想在关于机器学习促进商业的核心方式上有深层次的理解？你是否想能同专家们讨论关于回归，分类，深度学习以及推荐系统的一切？在这门课上，你将会通过一系列实际案例学习来获取实践经历。在这门课结束的时候，

从本节课中

Regression: Predicting House Prices

This week you will build your first intelligent application that makes predictions from data.<p>We will explore this idea within the context of our first case study, predicting house prices, where you will create models that predict a continuous value (price) from input features (square footage, number of bedrooms and bathrooms,...). <p>This is just one of the many places where regression can be applied.Other applications range from predicting health outcomes in medicine, stock prices in finance, and power usage in high-performance computing, to analyzing which regulators are important for gene expression.</p>You will also examine how to analyze the performance of your predictive model and implement regression in practice using an iPython notebook.

- Carlos GuestrinAmazon Professor of Machine Learning

Computer Science and Engineering - Emily FoxAmazon Professor of Machine Learning

Statistics

[MUSIC]

And so let's actually use the model.

So, next we've now trained a liner regression model, let's evaluate it.

So, what we're gonna do next is

#Evaluate the simple model.

So how to evaluate it?

We're gonna look at the test data.

So, remember we had the split test data.

Let's understand test data a little bit.

So, for example, let's print.

For the test data, for

the price column, what's the average price?

What's the mean price?

So this just computes the average price and the average price for

the test data, for this data from Seattle is $543,000.

That's how average house costs, it's pretty expensive actually.

Now, we've built a square foot model, and so

what we want to do is evaluate it on this test data.

So we're going to take the sqft_model

that we built and we're gonna call what's called evaluate function.

Which can take a test data set and print out or

return some statistics of how well that fit is doing.

So let's do that.

So, I'm actually going to type, print in the beginning,

because it formats it a little bit nicer.

And you'll see that the maximal error over all test houses was 4.1 million.

So there was one house that was an outlier, it was really badly predicted.

And average error, so the RMSE, so the root means squared error,

we talked about this.

Emily talked about this with us during the lectures,

is $255,000, so that's the RMSE.

So we built this simple model, we tested, has pretty high RMSE,

but let's look at some predictions it tries to make from the data.

[MUSIC]