RuntimeError: range of [0, nan] is not finite

[RonaldYuren]

I made a few modification for yolov7-tiny ( Use a new kind of activation function )，
But there comes a new problem just like below.

######## below are the error message:
Traceback (most recent call last):
File "/home/si2/YuRen/yolov7/yolov7_qat/scripts/qat.py", line 319, in
cmd_quantize(
File "/home/si2/YuRen/yolov7/yolov7_qat/scripts/qat.py", line 161, in cmd_quantize
quantize.calibrate_model(model, train_dataloader, device)
File "/home/si2/YuRen/yolov7/yolov7_qat/scripts/quantize.py", line 292, in calibrate_model
collect_stats(model, dataloader, device, num_batch=num_batch)
File "/home/si2/YuRen/yolov7/yolov7_qat/scripts/quantize.py", line 273, in collect_stats
output = model(imgs)
^^^^^^^^^^^
File "/home/si2/anaconda3/envs/YuRen1/lib/python3.11/site-packages/torch/nn/modules/module.py", lin e 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/si2/anaconda3/envs/YuRen1/lib/python3.11/site-packages/torch/nn/modules/module.py", lin e 1527, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/si2/YuRen/yolov7/models/yolo.py", line 606, in forward
return self.forward_once(x, profile) # single-scale inference, train
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/si2/YuRen/yolov7/models/yolo.py", line 632, in forward_once
x = m(x) # run
^^^^
File "/home/si2/anaconda3/envs/YuRen1/lib/python3.11/site-packages/torch/nn/modules/module.py", lin e 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/si2/anaconda3/envs/YuRen1/lib/python3.11/site-packages/torch/nn/modules/module.py", lin e 1527, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/si2/YuRen/yolov7/models/common.py", line 129, in fuseforward
return self.act(self.conv(x))
^^^^^^^^^^^^
File "/home/si2/anaconda3/envs/YuRen1/lib/python3.11/site-packages/torch/nn/modules/module.py", lin e 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/si2/anaconda3/envs/YuRen1/lib/python3.11/site-packages/torch/nn/modules/module.py", lin e 1527, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/si2/anaconda3/envs/YuRen1/lib/python3.11/site-packages/pytorch_quantization/nn/modules/ quant_conv.py", line 125, in forward
quant_input, quant_weight = self._quant(input)
^^^^^^^^^^^^^^^^^^
File "/home/si2/anaconda3/envs/YuRen1/lib/python3.11/site-packages/pytorch_quantization/nn/modules/ quant_conv.py", line 85, in _quant
quant_input = self._input_quantizer(input)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/si2/anaconda3/envs/YuRen1/lib/python3.11/site-packages/torch/nn/modules/module.py", lin e 1518, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/si2/anaconda3/envs/YuRen1/lib/python3.11/site-packages/torch/nn/modules/module.py", lin e 1527, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/si2/anaconda3/envs/YuRen1/lib/python3.11/site-packages/pytorch_quantization/nn/modules/ tensor_quantizer.py", line 338, in forward
self._calibrator.collect(inputs)
File "/home/si2/anaconda3/envs/YuRen1/lib/python3.11/site-packages/pytorch_quantization/calib/histo gram.py", line 109, in collect
self._calib_hist = torch.histc(x, bins=self._num_bins, min=0, max=x_max)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: range of [0, nan] is not finite

########

How do I solve this question about the "Nan" value ?