video object detection for half precision throwing error

[spaul13]

while running the object detection using video_demo_half.py script (16-bit half precision floats), I am getting the following error

Traceback (most recent call last):
File "video_demo_half.py", line 112, in
model(get_test_input(inp_dim, CUDA), CUDA)
File "C:\Users\paul90\AppData\Local\Continuum\anaconda3\envs\yolov3\lib\site-packages\torch\nn\modules\module.py", line 477, in call
result = self.forward(*input, **kwargs)
File "C:\yolo\DroNet_PyTorch\darknet.py", line 319, in forward
x = self.module_listi
File "C:\Users\paul90\AppData\Local\Continuum\anaconda3\envs\yolov3\lib\site-packages\torch\nn\modules\module.py", line 477, in call
result = self.forward(*input, **kwargs)
File "C:\Users\paul90\AppData\Local\Continuum\anaconda3\envs\yolov3\lib\site-packages\torch\nn\modules\container.py", line 91, in forward
input = module(input)
File "C:\Users\paul90\AppData\Local\Continuum\anaconda3\envs\yolov3\lib\site-packages\torch\nn\modules\module.py", line 477, in call
result = self.forward(*input, **kwargs)
File "C:\Users\paul90\AppData\Local\Continuum\anaconda3\envs\yolov3\lib\site-packages\torch\nn\modules\conv.py", line 301, in forward
self.padding, self.dilation, self.groups)
RuntimeError: Input type (torch.cuda.FloatTensor) and weight type (torch.cuda.HalfTensor) should be the same

[gplast]

Hello @spaul13 , i didn't add support for video_demo_half.py, i will try to change the code in order to work and update my repo. Just give me a few days.

Thank you.