Hugging Face's logo

MRiabov
/
WireSegHR

Image Segmentation
Transformers
English
wire
drone
computer_vision
Model card Files
xet
 Community
1
WireSegHR

Commit History

  1. (debug) import yaml

    MRiabov
    Verified signature
    b7ea8a4

  2. Update model card

    MRiabov
    Verified signature
    e3db6f5

  3. Git lfs changed to accept png and jpg

    MRiabov
    4bdc4a4

  4. (minor) push paper files for easier LLM usage

    MRiabov
    1a0b138

  5. Add final pt to git

    MRiabov
    b4ec92c

  6. Safetensors model export

    MRiabov
    96ac66c

  7. Safetensors export and infer.

    MRiabov
    dae69c0

  8. Add images to README

    MRiabov
    befda65

  9. fix strip checkpoint saving

    MRiabov
    5571883

  10. README and cleanup

    MRiabov
    e05cc17

  11. Metrics computation in infer.py

    MRiabov
    d79c41b

  12. (minor util) strip checkpoint util

    MRiabov
    a0ad699

  13. (minor refactor) os.path to pathlib and tdqm to infer.py

    MRiabov
    8223f36

  1. Attempting export to onnx and tensorRT

    MRiabov
    2991754

  2. (test) test ResNet backbone

    MRiabov
    6e8ad0e

  3. Add export to onnx and tensorRT

    MRiabov
    c2fb738

  4. Add benchmark to infer.py

    MRiabov
    8943a48

  5. (debug) fix eval loop not selecting randomly

    MRiabov
    32c94fb

  6. Save test visuals during final eval

    MRiabov
    ba1bfde

  7. Add ResNet-50 as a backbone option

    MRiabov
    a2999cc

  8. (cleanup) removing timm from the codebase.

    MRiabov
    e3ab023

  9. Merge branch 'main' of https://github.com/MRiabov/wire-seg-hr-impl

    MRiabov
    8cd5198

  10. (debug) to torch operations in bfloat conversions.

    MRiabov
    854831b

  11. Add final checkpoint save and test set val

    MRiabov
    feec877

  12. (untested test) fixing some test errors

    MRiabov
    ae754fb

  13. Eval to infer.py

    MRiabov
    500cad9

  14. (optim) benchmark=True and higher eval batch size

    MRiabov
    d2db539

  15. Speeding up eval by reducing iters

    MRiabov
    8b1ceda

  16. Added precision training option

    MRiabov
    8743e1d

  17. (optim) prefetch and speed up eval.

    MRiabov
    efdb6a6

  18. (debug) 6 chan instead of 7. (warn: conflicting git changes)

    MRiabov
    a527623

  19. to mit_b2 as per paper

    MRiabov
    d46d294

  1. (debug) eval fixes

    MRiabov
    e160020

  2. (devops) automatic pull and preprocess of datasets

    MRiabov
    953508f

  3. (config) reducing training iters and eval, checkpoint intervals

    MRiabov
    e8ba7db

  4. (test) test dataset

    MRiabov
    1d66790

  5. Pull ttpla script

    MRiabov
    f2f9fec

  6. Running training

    MRiabov
    61275b8

  7. (format) format

    MRiabov
    6928e82

  8. Training loop and around the codebase

    MRiabov
    8e73ec9

  9. Dataset pull script

    MRiabov
    ebeb96c

  10. Project skeleton generated

    MRiabov
    8ea2eff

  11. Unpack `.tex` source.

    MRiabov
    5cab910

  1. Repo init (paper add)

    MRiabov
    b03a07c