FLYolo training

[dalqattan]

Hi,
I am trying to text the FLyolo using quick start example.

I run train.py but I got the error below

Starting training for 5 epochs...

Epoch   gpu_mem       box       obj       cls    labels  img_size
 0%|                                                                                                                                                                                                                                                                                                | 0/12 [00:08<?, ?it/s]
Traceback (most recent call last):
 File "C:\temp\CarlaFLCAV\FLYolo\yolov5\train.py", line 628, in <module>
   main(opt)
 File "C:\temp\CarlaFLCAV\FLYolo\yolov5\train.py", line 525, in main
   train(opt.hyp, opt, device, callbacks)
 File "C:\temp\CarlaFLCAV\FLYolo\yolov5\train.py", line 322, in train
   loss, loss_items = compute_loss(pred, targets.to(device))  # loss scaled by batch_size
 File "C:\temp\CarlaFLCAV\FLYolo\yolov5\utils\loss.py", line 120, in __call__
   tcls, tbox, indices, anchors = self.build_targets(p, targets)  # targets
 File "C:\temp\CarlaFLCAV\FLYolo\yolov5\utils\loss.py", line 217, in build_targets
   indices.append((b, a, gj.clamp_(0, gain[3] - 1), gi.clamp_(0, gain[2] - 1)))  # image, anchor, grid indices
RuntimeError: result type Float can't be cast to the desired output type __int64

It would be appreciated if you can help.

Thanks
Duaa

[bearswang]

It seems that the problem occured due to the torch version. You may change "gain = torch.ones(7, device=targets.device)" into "gain = torch.ones(7, device=targets.device).long()" in file loss.py under folder "yolov5/utils" (C:\temp\CarlaFLCAV\FLYolo\yolov5\utils\loss.py​)