未加星标

Neural Regression Using PyTorch

字体大小 | |
[开发(python) 所属分类 开发(python) | 发布者 店小二05 | 时间 2018 | 作者 红领巾 ] 0人收藏点击收藏

For the past several months I’ve been poking around the PyTorch neural network library. One of the fundamental problems for any NN library is numeric regression. I attacked one of my standard regression problems, the Boston Housing dataset.

The goal is to predict the median house price for one of 506 small towns near Boston. The data comes from a 1978 research paper that looked at air pollution. House prices were incredibly low back in the 1970s. There are 13 predictor variables: crime rate in the town, large lot percentage, industry information, Charles River information, pollution, average number rooms per house, house age information, distance to Boston, accessibility to highways, tax rate, pupil-teacher ratio, proportion of Black residents, and percentage of low-status residents.


Neural Regression Using PyTorch

After a couple of hours of work, I got a demo up and running. But it wasn’t easy. PyTorch has a lot of quirks and the only reason I as able to get my regression model working relatively quickly was that I’d seen most of the strange error messages before, and I’ve spent many, many hours tracking PyTorch errors down.

Even though my demo is working, there are still a few parts of my code that I don’t fully understand, so I’ve got more work to do. Eventually I’ll satisfy myself and write up a complete explanation. The good news is that once you’ve got a robust regression problem solved using PyTorch (or Keras or CNTK or TensorFlow) you can use your solution as a template for most other regression problems.


Neural Regression Using PyTorch

According to Wikipedia, the nicest area (low crime, high house prices, high education) of Boston is the Back Bay and the worst area (high crime, low house prices, low education) is Mattapan. They’re only about four miles away from each other geographically. Strange. I’ve only been to Boston a few times and it’s always been in the downtown area for work so I haven’t formed an impression of the city.

本文开发(python)相关术语:python基础教程 python多线程 web开发工程师 软件开发工程师 软件开发流程

代码区博客精选文章
分页:12
转载请注明
本文标题:Neural Regression Using PyTorch
本站链接:https://www.codesec.net/view/611491.html


1.凡CodeSecTeam转载的文章,均出自其它媒体或其他官网介绍,目的在于传递更多的信息,并不代表本站赞同其观点和其真实性负责;
2.转载的文章仅代表原创作者观点,与本站无关。其原创性以及文中陈述文字和内容未经本站证实,本站对该文以及其中全部或者部分内容、文字的真实性、完整性、及时性,不作出任何保证或承若;
3.如本站转载稿涉及版权等问题,请作者及时联系本站,我们会及时处理。
登录后可拥有收藏文章、关注作者等权限...
技术大类 技术大类 | 开发(python) | 评论(0) | 阅读(55)