I am trying to train the efficientnet model on the coco dataset and it is running through the training process well. However, when it comes to the evaluation phase, I am getting a valueerror in the "compute_on_dataset" method, specifically from line 45. It is trying to unpack the batch list into images, targets, and image_ids, but it only finds 2 elements in the list, the first being a 300x300 image and second being "None". Could I get some help in understanding where the targets and image_ids are supposed to be coming from?
Hello,
I am trying to train the efficientnet model on the coco dataset and it is running through the training process well. However, when it comes to the evaluation phase, I am getting a valueerror in the "compute_on_dataset" method, specifically from line 45. It is trying to unpack the batch list into images, targets, and image_ids, but it only finds 2 elements in the list, the first being a 300x300 image and second being "None". Could I get some help in understanding where the targets and image_ids are supposed to be coming from?
BN