RSPEC 能否处理 90% 通过的阈值测试？

发布于 2024-12-12 06:39:49 字数 262 浏览 0 评论 0原文

我有一个分类函数，它确定一段文本是否属于一个类别（即返回真/假）。我有一系列现实世界的例子，我不希望它 100% 正确分类。

def meal?(text)
  !text.match(/restaurant|meal|wine|.../i).nil?
end

RSPEC 能否指定一个测试，当我将这数百个示例提供给该函数时，只要正确识别 90%，就算通过？更好的是在报告中提供失败的 RSPEC 输出示例。如果可能的话如何编写这样的测试？

原文

I have a classification function, which determines if a paragraph of text belongs to a category (i.e. it returns true/false). I have an array of real world examples, and I do not expect it classify correctly 100% of the time.

def meal?(text)
  !text.match(/restaurant|meal|wine|.../i).nil?
end

Can RSPEC specify a test that, when I feed these hundreds of examples to that function, as long as 90% are identified correctly, then it counts as a pass? Even better would be to have RSPEC output examples that fails in the report. If it is possible how can one write such a test?

分享到QQ

分享到微博

如果你对这篇内容有疑问，欢迎到本站社区发帖提问参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。

发布评论

需要登录才能够评论，你可以免费注册一个本站的账号。

度的依靠╰つ 2024-12-19 06:39:49

当然。

total_correct = 0;
@examples.each do |e|
  match = text.match(/restaurant|meal|wine|.../i)
  match.nil? ? puts e : total_correct += 1
end
(total_correct.to_f / @examples.size).should >= 0.9

任何您可以比较或简化为布尔表达式的内容，RSpec 都可以处理……这几乎是软件中的所有内容。这只是弄清楚如何表达它的问题。

Sure.

total_correct = 0;
@examples.each do |e|
  match = text.match(/restaurant|meal|wine|.../i)
  match.nil? ? puts e : total_correct += 1
end
(total_correct.to_f / @examples.size).should >= 0.9

Anything you can compare, or reduce to a boolean expression, RSpec can handle...which is pretty much everything in software. It's just a matter of figuring out how to express it.

回复收藏 0 原文

~没有更多了~