cancel
Showing results for 
Search instead for 
Did you mean: 

How to train multi-class image classification model?

How to train multi-class image classification model?

I've uploaded the dataset to the platform and trained 7 models with the auto modeling tool. But when I check the trained models, most of them seem to be outlier detection models. The training set had 4 classes and all of them were separated into different folders. How do I build a multi-classification model?

Labels (1)
1 Solution

Accepted Solutions

@shofam What you've described as putting the images into the different folders specifies them as separate classes and that is how one would build multi-class models.

When you evaluate the models using the Understand tab I presume you are able to identify the correct image classes from the models you've created, presuming you did not select the Anomaly detection option.

 

To get a further understanding of the Visual AI capabilities you may attend the instructor-led course listed below:

https://university.datarobot.com/visual-ai-virtual-instructor-led

 

View solution in original post

2 Replies

@shofam What you've described as putting the images into the different folders specifies them as separate classes and that is how one would build multi-class models.

When you evaluate the models using the Understand tab I presume you are able to identify the correct image classes from the models you've created, presuming you did not select the Anomaly detection option.

 

To get a further understanding of the Visual AI capabilities you may attend the instructor-led course listed below:

https://university.datarobot.com/visual-ai-virtual-instructor-led

 

Oh my bad! I've set up the project as an outlier detection project by mistake. Thanks for the help though 😄

0 Kudos