pytorch/vision

Issues

• Difficulty building on macOS

  #8421 opened a month ago by joshestein
  3

• Build fails: error: unknown type name 'j_decompress_ptr'

  #8397 opened a month ago by yurivict
  3

• -

  #8424 opened a month ago by benchd
  0

• Video Reader's get_metadata function fails on videos with sound

  #8420 opened a month ago by NSalberg
  0

• Can't use gaussian_blur if sigma is a tensor on gpu

  #8401 opened a month ago by Xact-sniper
  2

• Mypy job is broken

  #8377 opened a month ago by NicolasHug
  0

• Mask r-cnn training runs infinitely without output or error

  #8409 opened a month ago by MontassarTn
  1

• detection AnchorGenerator Source code issues

  #8410 opened a month ago by yz4322gly
  1

• Differences in CPU vs CUDA resize for uint8 images

  #8399 opened a month ago by gau-nernst
  2

• Compiling resize_image: function interpolate not_implemented

  #8389 opened 2 months ago by Tomsen1410
  1

• Run all torchvision models in one script.

  #8394 opened a month ago by leizhenyuan
  1

• Choose either 'long' or 'short' options for the resize anchor edge if the size variable is scalar

  #8358 opened 2 months ago by sidijju
  3

• AttributeError: module 'torchvision.transforms' has no attribute 'v2'

  #8390 opened a month ago by ImahnShekhzadeh
  1

• Enable Video models for other tasks

  #8400 opened a month ago by bhack
  0

• Improved functionality for Oxford IIIT Pet data loader

  #8364 opened 2 months ago by matlabninja
  2

• support compact/encoded RLE masks in COCO dataset wrapper

  #8351 opened 3 months ago by rizavelioglu
  5

• Release 2.3 create release notes for vision, audio and text

  #8323 opened 2 months ago by atalman
  1

• Image scaling is performed incorrectly (in all detection models!)

  #8370 opened 2 months ago by robwijnhoven
  5

• Regarding IMAGENET1K_V1 and IMAGENET1K_V2 weights

  #8382 opened 2 months ago by asusdisciple
  0

• RuntimeError/AssertionError when finetuning fasterrcnn_resnet50_fpn on visdrone dataset

  #8376 opened 2 months ago by aimrgni
  3

• AVX512 support machine cannot resize uint8 image with BILINEAR interpolation as it is

  #8374 opened 2 months ago by vinnamkim
  0

• IndexError: index 168 is out of bounds for dimension 0 with size 168 in keypointrcnn_loss

  #8371 opened 2 months ago by ErikDerGute
  1

• Release 2.3 branch cut and apply release only changes

  #8322 opened 2 months ago by atalman
  0

• Nightly build flaky pytorch/vision / conda-py3_11-cpu builds

  #8372 opened 2 months ago by atalman
  1

• Any operation on loaded image segfaults since 0.17.1 on Mac

  #8367 opened 2 months ago by twoertwein
  2

• position, colour, and background colour of text labels in draw_bounding_boxes

  #8317 opened 3 months ago by carandraug
  1

• Cherry-picks for 0.18

  #8336 opened 3 months ago by NicolasHug
  4

• When VideoClip processes video, the final voice output time is not equal to the video time because of the different sampling of video and voice.

  #8362 opened 2 months ago by lxygoodjob
  1

• Unable to obtain results of ResNet50 v1

  #8363 opened 2 months ago by netw0rkf10w
  1

• About uint16 support

  #8359 opened 2 months ago by NicolasHug
  0

• Allow size to equal max_size for resizes on longer edge

  #8357 opened 2 months ago by sidijju
  0

• PyTorch standard Coco dataset (datasets.CocoDetection) not compatible with Faster R-CNN object detection model

  #8353 opened 3 months ago by ranjaniocl
  6

• Update typing input annotations to `convert_bounding_box_format`.

  #8354 opened 2 months ago by GdoongMathew
  2

• No module named 'torchvision.transforms.v2'

  #8349 opened 3 months ago by noivan0
  4

• Extracting custom EXIF data from PNG files

  #8348 opened 3 months ago by frank690
  2

• Main branch build failing with gcc 11.4.0

  #8339 opened 3 months ago by sanchitintel
  2

• Building on archlinux fails with include error

  #8307 opened 3 months ago by binarynoise
  8

• MD5 checksum of download file does not match the one on record

  #8338 opened 3 months ago by andreasfloros
  1

• RuntimeError: The size of tensor a (14) must match the size of tensor b (6) at non-singleton dimension 0

  #8326 opened 3 months ago by ranjaniocl
  4

• Model instantiation without loading from disk

  #8327 opened 3 months ago by AndreasKaratzas
  2

• Add Jpeg Compression augmentation

  #8290 opened 3 months ago by gau-nernst
  5

• Cleaner SanitizeBoundingBoxes transform

  #8294 opened 3 months ago by MarcSzafraniec
  6

• Add 3.12 CI jobs

  #8293 opened 3 months ago by NicolasHug
  3

• Allow empty class folders in ImageFolder

  #8297 opened 3 months ago by NicolasHug
  0

• aarch64 build for AWS Linux - Failed to load image Python extension

  #8305 opened 3 months ago by elkay
  6

• inference gives arbitrary output from trained model on `np.ones`

  #8304 opened 3 months ago by innat
  1

• Support reading WebP and HEIC image formats natively

  #8301 opened 3 months ago by gau-nernst
  0

• All CI job are failing

  #8292 opened 3 months ago by NicolasHug
  0

• Densenet pretrained model - “unexpected EOF, expected 899672 more bytes. The file might be corrupted.“)

  #8296 opened 3 months ago by JillCakmak
  2

• "ValueError: Could not find the operator torchvision::nms" upon importing torchvision

  #8289 opened 3 months ago by shinobiultra
  4