Quantcast
Channel: Machine Learning
Viewing all articles
Browse latest Browse all 62611

How to classify 'everything else'?

$
0
0

Hypothetical scenario.

I have 10 types of dogs and then a whole host of 'noisy samples' which can be literally anything: cats, humans, tables, phones, buildings, etc etc.

I couldn't find any resources online for how to categorize 'everything else' so I'm wondering if anyone has experience doing so?

Right now I'm leaning towards doing a classification by 2 and then further classifying the dogs by 10....

submitted by Should_I_say_this
[link][6 comments]

Viewing all articles
Browse latest Browse all 62611

Trending Articles