Ivan comments

Repositories
Issues
Comments

Results 3 comments of


                                            Ivan

Instance segmentation at U7，onnx how to Infer？

If you check code at `seg/models.common.py` you can find that onnx return only first output from model. ```python elif self.onnx: # ONNX Runtime im = im.cpu().numpy() # torch to numpy...

Instance segmentation at U7，onnx how to Infer？

I change common file like yolov5 and it works` DetectMultiBackend init()` ```python elif onnx: # ONNX Runtime LOGGER.info(f'Loading {w} for ONNX Runtime inference...') cuda = torch.cuda.is_available() and device.type != 'cpu'...

I have problem when training: Function AddmmBackward returned an invalid gradient at index 1 - got [32, 25088] but expected shape compatible with [32, 32768]

I think you use 128x128 image instead of 112x112