Closed xf739645524 closed 5 years ago
Okey , I trained for a week and the crossentropy loss finally came down.
Hi, how about the results when you train 20W IDs? Better than less IDs?
Hi, how about the results when you train 20W IDs? Better than less IDs?
This loss is very difficult to drop
This loss is very difficult to drop
I have trained on my own dataset, The 8000ID training produced the best results,If you add more ID, the accuracy starts to go down.Have you solved this problem yet?
If I understand correctly,change num_classes in network.py to 200,000. Maybe I should write it in argparse.