Select to view content in your preferred language

How to assess the accuracy of the output from an Object Detection Deep Learning model?

92
1
Saturday
Labels (2)
ScarletMaguire
Occasional Contributor

I have trained an object classification model trained to classify polygon shapes into 4 different categories. I plan to use the model to classify around 100 objects. My inputs will be a raster layer for the input and a polygon feature class for the input features. Once this model has attempted to classify, how do I assess the accuracy?

ESRI documentation advises the use of the "Accuracy Assessment" tool, and that the input/reference dataset should be "A polygon feature class or a shapefile. The format of the feature class attribute table must match the training samples. To ensure this, you can create the reference dataset using the Training Samples Manager page to read and write the dataset". However, I cannot find a standalone "Accuracy Assessment" tool in 3.4 and I am not sure how to use the Training Samples Manager to create a reference dataset, as I thought this tool was only to create actual training samples.

Any advice or pointers would be really appreciated, I have hit a dead end here!

0 Kudos
1 Reply