Apex recommended for faster mixed precision training: https://github.com/NVIDIA/apex
Namespace(adam=False, batch_size=16, bucket='', cache_images=False, cfg='/content/darknet/cfg/yolov4.cfg', data='/content/objcoco.data', device='', epochs=300, evolve=False, freeze_layers=False, img_size=[320, 640], multi_scale=False, name='', nosave=False, notest=False, rect=False, resume=False, single_cls=False, weights='/content/darknet/weights/yolov4.weights')
Using CUDA device0 _CudaDeviceProperties(name='Tesla T4', total_memory=15079MB)

Start Tensorboard with "tensorboard --logdir=runs", view at http://localhost:6006/
2020-12-17 15:56:24.207482: I tensorflow/stream_executor/platform/default/dso_loader.cc:48] Successfully opened dynamic library libcudart.so.10.1
Traceback (most recent call last):
  File "/content/archived_yolov3/yolov3/train.py", line 431, in <module>
    train(hyp)  # train normally
  File "/content/archived_yolov3/yolov3/train.py", line 91, in train
    model = Darknet(cfg).to(device)
  File "/content/archived_yolov3/yolov3/models.py", line 231, in __init__
    self.module_defs = parse_model_cfg(cfg)
  File "/content/archived_yolov3/yolov3/utils/parse_config.py", line 50, in parse_model_cfg
    assert not any(u), "Unsupported fields %s in %s. See https://github.com/ultralytics/yolov3/issues/631" % (u, path)
AssertionError: Unsupported fields ['max_delta'] in /content/darknet/cfg/yolov4.cfg. See https://github.com/ultralytics/yolov3/issues/631