dav-ell

Thanks very much for your work on this @FidgetySo. Does this PR work yet? How much more is left to go?

Running into this same issue.

Any progress on this? If exporting quantized pytorch models to ONNX is not supported, is there a preferred route? i.e. TensorRT?

Upsample causing issues for me too, documenting here: https://github.com/NVIDIA/Torch-TensorRT/issues/961

> I've successfully convert it to TensorRT version for inference. Kindly check it here: https://github.com/kongyanye/EfficientDet-TensorRT Amazing work! Thanks very much for sharing. I noticed, though, that you didn't include `fold_constants=True`...

@zylo117 are you planning on making further commits to this repo, or should someone fork this and apply updates elsewhere?

I'm also looking for a solution to this problem in this repo. Small objects are completely ignored for me, so far.

Attempting to make a repro script but stuck on a different issue now... ```python import torch_tensorrt torch_tensorrt.logging.set_reportable_log_level(torch_tensorrt.logging.Level.Graph) print(torch_tensorrt.__version__) import torch import torch.nn as nn from torch.nn import functional as F...

Apologies, I forgot `model.eval()`. After doing that, it just hangs...

Correction, it has a segmentation fault. See attached logs. Code is in temp.py.txt. [logs_graph.txt](https://github.com/NVIDIA/Torch-TensorRT/files/8420282/logs_graph.txt) [logs_debug.txt](https://github.com/NVIDIA/Torch-TensorRT/files/8420283/logs_debug.txt) [temp.py.txt](https://github.com/NVIDIA/Torch-TensorRT/files/8420284/temp.py.txt)