I am getting this error, can anyone please suggest the solution?
DLL 2022-09-01 23:16:27.714134 - PARAMETER dataset path : ./coco epochs : 65 batch size : 32 eval batch size : 32 no cuda : False seed : None checkpoint path : ./models/epoch_*.pt mode : evaluation eval on epochs : [21, 31, 37, 42, 48, 53, 59, 64] lr decay epochs : [43, 54] learning rate : 0.0026 momentum : 0.9 weight decay : 0.0005 lr warmup : None backbone : resnet50 backbone path : None num workers : 4 AMP : False precision : fp32
Using seed = 2860
loading annotations into memory...
Done (t=0.40s)
creating index...
Traceback (most recent call last):
File "./main.py", line 286, in
train(train_loop_func, logger, args)
File "./main.py", line 148, in train
train_loader = get_train_loader(args, args.seed - 2**31)
File "/data4/home/mehuls/Project/DeepLearningExamples/PyTorch/Detection/SSD/ssd/data.py", line 42, in get_train_loader
train_loader = DALICOCOIterator(train_pipe, 118287 / args.N_gpu)
File "/data4/home/mehuls/Project/DeepLearningExamples/PyTorch/Detection/SSD/ssd/coco_pipeline.py", line 192, in init
self._first_batch = self.next()
File "/data4/home/mehuls/Project/DeepLearningExamples/PyTorch/Detection/SSD/ssd/coco_pipeline.py", line 302, in next
return self.next();
File "/data4/home/mehuls/Project/DeepLearningExamples/PyTorch/Detection/SSD/ssd/coco_pipeline.py", line 251, in next
images_torch_type = to_torch_type[np.dtype(images[0].dtype)]
TypeError: Cannot interpret '<DALIDataType.FLOAT: 9>' as a data type
NVIDIA
