You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I added a test script under src/image_classification. This script is pretty straightforward, as it loads the saved model and makes a prediction on a single image.
If you want to test multiple images, I would recommend using a data generator. (Documentation)
@Horea94 Thanks so much for this update. I run this file and its working fine. I'm getting result in terminal mode, how to save resultant image? I mean how to get predicted image with detected box around the object?
The model in this repo is a classifier. That is, it provides only a label prediction based on an input image.
For bounding box prediction, you would need a detector like Faster R-CNN, YOLO or SSD.
Hi @Horea94 Thanks for this wonderful work. Training is ok but how I can test any image? Seems that there is no code for testing.
The text was updated successfully, but these errors were encountered: