pytorch/vision issues and pull requests

#3093 - VOCSegmentation transforms.ToTensor() not working

Issue - State: closed - Opened by sirtris almost 4 years ago - 7 comments
Labels: question, topic: semantic segmentation

#3058 - How to solve this error? RuntimeError: Could not run 'torchvision::nms' with arguments from the 'CUDA' backend

Issue - State: open - Opened by manmani3 almost 4 years ago - 30 comments
Labels: needs reproduction, module: ops

#2962 - Add pathlib.Path support to read_image

Issue - State: closed - Opened by Yura52 about 4 years ago - 6 comments
Labels: enhancement, module: io

#2950 - In transforms.Resize, tensor interpolate is not the same as PIL resize.

Issue - State: closed - Opened by hjinlee88 about 4 years ago - 34 comments
Labels: bug, module: transforms

#2929 - [Feature Request] Support same_on_batch option for transforms

Issue - State: open - Opened by xvjiarui about 4 years ago - 5 comments
Labels: module: transforms

#2921 - ModuleNotFoundError: No module named '_lzma'

Issue - State: closed - Opened by ZoranPandovski about 4 years ago - 18 comments
Labels: module: datasets

#2912 - Cannot pip install torchvision==0.8.0+cu110

Issue - State: closed - Opened by rsdmse about 4 years ago - 13 comments

#2868 - Make the resize anchor edge variable if size is scalar

Issue - State: closed - Opened by pmeier about 4 years ago - 11 comments
Labels: module: transforms, new feature

#2788 - Error with torchvision.io.read_image with models

Issue - State: closed - Opened by oke-aditya about 4 years ago - 5 comments
Labels: question, module: transforms

#2788 - Error with torchvision.io.read_image with models

Issue - State: closed - Opened by oke-aditya about 4 years ago - 5 comments
Labels: question, module: transforms

#2740 - ValueError: All bounding boxes should have positive height and width. Found invaid box [500.728515625, 533.3333129882812, 231.10546875, 255.2083282470703] for target at index 0.

Issue - State: closed - Opened by kashf99 about 4 years ago - 28 comments
Labels: question, topic: object detection

#2709 - Installing torchvision for ppc64le IBM architecture

Issue - State: open - Opened by brando90 about 4 years ago - 6 comments
Labels: topic: binaries

#2707 - Are new models planned to be added?

Issue - State: open - Opened by talcs about 4 years ago - 62 comments
Labels: needs discussion, module: models, new feature

#2691 - Reading PNG/JPG images into a torch::tensor and saving a torch::tensor to PNG/JPG in C++ without OpenCV

Issue - State: closed - Opened by ghost about 4 years ago - 8 comments
Labels: help wanted, module: io, module: c++ frontend

#2597 - [Legal question] Is it legal to use pre-trained models for commercial purposes?

Issue - State: open - Opened by ternaus about 4 years ago - 9 comments
Labels: module: models

#2593 - make ResNet and DenseNet more flexible for usage on smaller images

Pull Request - State: closed - Opened by stadlmax over 4 years ago - 8 comments

#2562 - Error(s) in loading state_dict in ..., unexpected keys

Issue - State: open - Opened by alex96295 over 4 years ago - 5 comments
Labels: module: models.quantization

#2447 - Label tracking meta-issue (edit me to get automatically CC'ed on issues! cc bot)

Issue - State: open - Opened by astaff over 4 years ago - 4 comments

#2402 - 3D NMS and RoiAlign for volumetric data

Issue - State: open - Opened by mibaumgartner over 4 years ago - 4 comments
Labels: needs discussion, module: ops

#2360 - subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1

Issue - State: open - Opened by puraminy over 4 years ago - 17 comments
Labels: awaiting response, needs reproduction, topic: build

#2308 - exported MASKRCNN ONNX model cannot run: Op (Slice) [ShapeInferenceError] Input axes has invalid data

Issue - State: closed - Opened by qizhen816 over 4 years ago - 7 comments
Labels: bug, help wanted, topic: object detection, module: onnx

#2291 - Package libjpeg and libpng within torchvision

Issue - State: closed - Opened by fmassa over 4 years ago - 7 comments
Labels: topic: build, topic: binaries, high priority

#2288 - Custom transform support

Pull Request - State: open - Opened by Sentient07 over 4 years ago - 1 comment
Labels: cla signed

#2262 - Unable to load CelebA dataset. File is not zip file error.

Issue - State: closed - Opened by ajayrfhp over 4 years ago - 25 comments
Labels: duplicate, help wanted, module: datasets

#2235 - NaN Loss for FasterRCNN on Multiclass Object Detection on Custom Dataset COCO

Issue - State: closed - Opened by sarmientoj24 over 4 years ago - 8 comments

#2134 - About RegisterOperators in C++

Issue - State: open - Opened by zsef123 over 4 years ago - 26 comments
Labels: help wanted, module: ops, module: c++ frontend, torchscript