Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Add TensorFlow and TFLite export #1127

Merged
merged 71 commits into from
Aug 17, 2021
Merged

Add TensorFlow and TFLite export #1127

merged 71 commits into from
Aug 17, 2021

Commits on Nov 28, 2020

  1. Add models/tf.py for TensorFlow and TFLite export

    @zldrobit @glenn-jocher
    zldrobit authored and glenn-jocher committed Nov 28, 2020
    Configuration menu
    Copy the full SHA
    5b6528c View commit details
    Browse the repository at this point in the history

  2. Set auto=False for int8 calibration

    @zldrobit @glenn-jocher
    zldrobit authored and glenn-jocher committed Nov 28, 2020
    Configuration menu
    Copy the full SHA
    a30dad4 View commit details
    Browse the repository at this point in the history

  3. Update requirements.txt for TensorFlow and TFLite export

    @zldrobit @glenn-jocher
    zldrobit authored and glenn-jocher committed Nov 28, 2020
    Configuration menu
    Copy the full SHA
    f0cb6e2 View commit details
    Browse the repository at this point in the history

  4. Read anchors directly from PyTorch weights

    @zldrobit @glenn-jocher
    zldrobit authored and glenn-jocher committed Nov 28, 2020
    Configuration menu
    Copy the full SHA
    ce73d3d View commit details
    Browse the repository at this point in the history

  5. Add --tf-nms to append NMS in TensorFlow SavedModel and GraphDef export

    @zldrobit @glenn-jocher
    zldrobit authored and glenn-jocher committed Nov 28, 2020
    Configuration menu
    Copy the full SHA
    d101f7e View commit details
    Browse the repository at this point in the history

  6. Remove check_anchor_order, check_file, set_logging from import

    @zldrobit @glenn-jocher
    zldrobit authored and glenn-jocher committed Nov 28, 2020
    Configuration menu
    Copy the full SHA
    55fb2eb View commit details
    Browse the repository at this point in the history

  7. Reformat code and optimize imports

    @glenn-jocher
    glenn-jocher committed Nov 28, 2020
    Configuration menu
    Copy the full SHA
    3dd69d9 View commit details
    Browse the repository at this point in the history

  8. Autodownload model and check cfg

    @glenn-jocher
    glenn-jocher committed Nov 28, 2020
    Configuration menu
    Copy the full SHA
    e8a9ad2 View commit details
    Browse the repository at this point in the history

  9. update --source path, img-size to 320, single output

    @glenn-jocher
    glenn-jocher committed Nov 28, 2020
    Configuration menu
    Copy the full SHA
    efbb853 View commit details
    Browse the repository at this point in the history

  10. Adjust representative_dataset

    @glenn-jocher
    glenn-jocher committed Nov 28, 2020
    Configuration menu
    Copy the full SHA
    aed53ce View commit details
    Browse the repository at this point in the history

Commits on Nov 30, 2020

  1. Put representative dataset in tfl_int8 block

    @zldrobit
    zldrobit committed Nov 30, 2020
    Configuration menu
    Copy the full SHA
    ccb2336 View commit details
    Browse the repository at this point in the history

Commits on Dec 2, 2020

  1. detect.py TF inference

    @glenn-jocher
    glenn-jocher committed Dec 2, 2020
    Configuration menu
    Copy the full SHA
    9f893c8 View commit details
    Browse the repository at this point in the history

  2. Merge remote-tracking branch 'origin/tf-only-export' into tf-only-export

    @glenn-jocher
    glenn-jocher committed Dec 2, 2020
    Configuration menu
    Copy the full SHA
    d9fad06 View commit details
    Browse the repository at this point in the history

  3. weights to string

    @glenn-jocher
    glenn-jocher committed Dec 2, 2020
    Configuration menu
    Copy the full SHA
    49a9e05 View commit details
    Browse the repository at this point in the history

  4. weights to string

    @glenn-jocher
    glenn-jocher committed Dec 2, 2020
    Configuration menu
    Copy the full SHA
    1867bb4 View commit details
    Browse the repository at this point in the history

Commits on Dec 4, 2020

  1. cleanup tf.py

    @glenn-jocher
    glenn-jocher committed Dec 4, 2020
    Configuration menu
    Copy the full SHA
    4eed608 View commit details
    Browse the repository at this point in the history

Commits on Dec 22, 2020

  1. Add --dynamic-batch-size

    @zldrobit
    zldrobit committed Dec 22, 2020
    Configuration menu
    Copy the full SHA
    8ba2ca9 View commit details
    Browse the repository at this point in the history

  2. Add xywh normalization to reduce calibration error

    @zldrobit
    zldrobit committed Dec 22, 2020
    Configuration menu
    Copy the full SHA
    4d9104a View commit details
    Browse the repository at this point in the history

  3. Merge branch 'master' into tf-only-export

    @glenn-jocher
    glenn-jocher committed Dec 22, 2020
    Configuration menu
    Copy the full SHA
    ae9bce8 View commit details
    Browse the repository at this point in the history

Commits on Dec 23, 2020

  1. Update requirements.txt 

    TensorFlow 2.3.1 -> 2.4.0 to avoid int8 quantization error
    @zldrobit
    zldrobit committed Dec 23, 2020
    Configuration menu
    Copy the full SHA
    cabb802 View commit details
    Browse the repository at this point in the history

Commits on Dec 24, 2020

  1. Fix imports 

    Move C3 from models.experimental to models.common
    @zldrobit
    zldrobit committed Dec 24, 2020
    Configuration menu
    Copy the full SHA
    a5967f8 View commit details
    Browse the repository at this point in the history

Commits on Jan 8, 2021

  1. Add models/tf.py for TensorFlow and TFLite export

    @zldrobit @glenn-jocher
    zldrobit authored and glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    dbc7f71 View commit details
    Browse the repository at this point in the history

  2. Set auto=False for int8 calibration

    @zldrobit @glenn-jocher
    zldrobit authored and glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    565e620 View commit details
    Browse the repository at this point in the history

  3. Update requirements.txt for TensorFlow and TFLite export

    @zldrobit @glenn-jocher
    zldrobit authored and glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    fc26561 View commit details
    Browse the repository at this point in the history

  4. Read anchors directly from PyTorch weights

    @zldrobit @glenn-jocher
    zldrobit authored and glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    05cc389 View commit details
    Browse the repository at this point in the history

  5. Add --tf-nms to append NMS in TensorFlow SavedModel and GraphDef export

    @zldrobit @glenn-jocher
    zldrobit authored and glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    817fcf8 View commit details
    Browse the repository at this point in the history

  6. Remove check_anchor_order, check_file, set_logging from import

    @zldrobit @glenn-jocher
    zldrobit authored and glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    9a4ce5e View commit details
    Browse the repository at this point in the history

  7. Reformat code and optimize imports

    @glenn-jocher
    glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    719479e View commit details
    Browse the repository at this point in the history

  8. Autodownload model and check cfg

    @glenn-jocher
    glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    243fa7f View commit details
    Browse the repository at this point in the history

  9. update --source path, img-size to 320, single output

    @glenn-jocher
    glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    3b6cf12 View commit details
    Browse the repository at this point in the history

  10. Adjust representative_dataset

    @glenn-jocher
    glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    5a5f949 View commit details
    Browse the repository at this point in the history

  11. detect.py TF inference

    @glenn-jocher
    glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    00d50fd View commit details
    Browse the repository at this point in the history

  12. Put representative dataset in tfl_int8 block

    @zldrobit @glenn-jocher
    zldrobit authored and glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    061907b View commit details
    Browse the repository at this point in the history

  13. weights to string

    @glenn-jocher
    glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    ca4550b View commit details
    Browse the repository at this point in the history

  14. weights to string

    @glenn-jocher
    glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    5e04c5c View commit details
    Browse the repository at this point in the history

  15. cleanup tf.py

    @glenn-jocher
    glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    9121a87 View commit details
    Browse the repository at this point in the history

  16. Add --dynamic-batch-size

    @zldrobit @glenn-jocher
    zldrobit authored and glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    e9bc606 View commit details
    Browse the repository at this point in the history

  17. Add xywh normalization to reduce calibration error

    @zldrobit @glenn-jocher
    zldrobit authored and glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    dacd8af View commit details
    Browse the repository at this point in the history

  18. Update requirements.txt 

    TensorFlow 2.3.1 -> 2.4.0 to avoid int8 quantization error
    @zldrobit @glenn-jocher
    zldrobit authored and glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    b492af9 View commit details
    Browse the repository at this point in the history

  19. Fix imports 

    Move C3 from models.experimental to models.common
    @zldrobit @glenn-jocher
    zldrobit authored and glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    fbf5a45 View commit details
    Browse the repository at this point in the history

  20. Merge branch 'tf-only-export' of https://github.com/zldrobit/yolov5 i… 

    …nto tf-only-export
    @glenn-jocher
    glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    c4cfbd9 View commit details
    Browse the repository at this point in the history

  21. implement C3() and SiLU()

    @glenn-jocher
    glenn-jocher committed Jan 8, 2021
    Configuration menu
    Copy the full SHA
    c761637 View commit details
    Browse the repository at this point in the history

Commits on Feb 2, 2021

  1. Fix reshape dim to support dynamic batching

    @zldrobit
    zldrobit committed Feb 2, 2021
    Configuration menu
    Copy the full SHA
    36ed3cd View commit details
    Browse the repository at this point in the history

Commits on Feb 4, 2021

  1. Merge branch 'master' into tf-only-export

    @glenn-jocher
    glenn-jocher committed Feb 4, 2021
    Configuration menu
    Copy the full SHA
    aad9e24 View commit details
    Browse the repository at this point in the history

Commits on Feb 6, 2021

  1. Merge branch 'master' into tf-only-export

    @glenn-jocher
    glenn-jocher committed Feb 6, 2021
    Configuration menu
    Copy the full SHA
    8ec975e View commit details
    Browse the repository at this point in the history

Commits on Mar 12, 2021

  1. Add epsilon argument in tf_BN, which is different between TF and PT

    @zldrobit
    zldrobit committed Mar 12, 2021
    Configuration menu
    Copy the full SHA
    b0fa5a3 View commit details
    Browse the repository at this point in the history

  2. Set stride to None if not using PyTorch, and do not warmup without Py… 

    …Torch
    @zldrobit
    zldrobit committed Mar 12, 2021
    Configuration menu
    Copy the full SHA
    710bf56 View commit details
    Browse the repository at this point in the history

Commits on Mar 23, 2021

  1. Add list support in check_img_size()

    @zldrobit
    zldrobit committed Mar 23, 2021
    Configuration menu
    Copy the full SHA
    c45ceef View commit details
    Browse the repository at this point in the history

  2. Add list input support in detect.py

    @zldrobit
    zldrobit committed Mar 23, 2021
    Configuration menu
    Copy the full SHA
    0d39b24 View commit details
    Browse the repository at this point in the history

Commits on Mar 25, 2021

  1. merge ultralytics:master

    @glenn-jocher
    glenn-jocher committed Mar 25, 2021
    Configuration menu
    Copy the full SHA
    47da942 View commit details
    Browse the repository at this point in the history

Commits on Mar 28, 2021

  1. merge ultralytics:master

    @glenn-jocher
    glenn-jocher committed Mar 28, 2021
    Configuration menu
    Copy the full SHA
    8cb7032 View commit details
    Browse the repository at this point in the history

  2. sys.path.append('./') to run from yolov5/

    @glenn-jocher
    glenn-jocher committed Mar 28, 2021
    Configuration menu
    Copy the full SHA
    4e1485b View commit details
    Browse the repository at this point in the history

Commits on Apr 3, 2021

  1. Add int8 quantization support for TensorFlow 2.5

    @zldrobit
    zldrobit committed Apr 3, 2021
    Configuration menu
    Copy the full SHA
    e4e6d6f View commit details
    Browse the repository at this point in the history

Commits on May 6, 2021

  1. Add get_coco128.sh

    @zldrobit
    zldrobit committed May 6, 2021
    Configuration menu
    Copy the full SHA
    aafe224 View commit details
    Browse the repository at this point in the history

  2. Remove --no-tfl-detect in models/tf.py (Use tf-android-tfl-detect bra… 

    …nch for EdgeTPU)
    @zldrobit
    zldrobit committed May 6, 2021
    Configuration menu
    Copy the full SHA
    d3be281 View commit details
    Browse the repository at this point in the history

Commits on Jun 2, 2021

  1. Merge branch 'develop' into tf-only-export

    @glenn-jocher
    glenn-jocher committed Jun 2, 2021
    Configuration menu
    Copy the full SHA
    9ca5d7a View commit details
    Browse the repository at this point in the history

  2. Update requirements.txt

    @glenn-jocher
    glenn-jocher committed Jun 2, 2021
    Configuration menu
    Copy the full SHA
    215c865 View commit details
    Browse the repository at this point in the history

Commits on Jun 7, 2021

  1. Replace torch.load() with attempt_load()

    @zldrobit
    zldrobit committed Jun 7, 2021
    Configuration menu
    Copy the full SHA
    a2867da View commit details
    Browse the repository at this point in the history

  2. Update requirements.txt

    @zldrobit
    zldrobit committed Jun 7, 2021
    Configuration menu
    Copy the full SHA
    86768d1 View commit details
    Browse the repository at this point in the history

Commits on Jun 11, 2021

  1. Add --tf-raw-resize to set half_pixel_centers=False

    @zldrobit
    zldrobit committed Jun 11, 2021
    Configuration menu
    Copy the full SHA
    972c6a2 View commit details
    Browse the repository at this point in the history

Commits on Jun 18, 2021

  1. Add --agnostic-nms for TF class-agnostic NMS

    @zldrobit
    zldrobit committed Jun 18, 2021
    Configuration menu
    Copy the full SHA
    eed4980 View commit details
    Browse the repository at this point in the history

Commits on Aug 16, 2021

  1. Merge master

    @glenn-jocher
    glenn-jocher committed Aug 16, 2021
    Configuration menu
    Copy the full SHA
    34b7d67 View commit details
    Browse the repository at this point in the history

  2. Cleanup after merge

    @glenn-jocher
    glenn-jocher committed Aug 16, 2021
    Configuration menu
    Copy the full SHA
    10eebf2 View commit details
    Browse the repository at this point in the history

  3. Cleanup2 after merge

    @glenn-jocher
    glenn-jocher committed Aug 16, 2021
    Configuration menu
    Copy the full SHA
    e3aa755 View commit details
    Browse the repository at this point in the history

  4. Cleanup3 after merge

    @glenn-jocher
    glenn-jocher committed Aug 16, 2021
    Configuration menu
    Copy the full SHA
    27fc39b View commit details
    Browse the repository at this point in the history

  5. Add tf.py docstring with credit and usage

    @glenn-jocher
    glenn-jocher committed Aug 16, 2021
    Configuration menu
    Copy the full SHA
    48a6bf8 View commit details
    Browse the repository at this point in the history

Commits on Aug 17, 2021

  1. pb saved_model and tflite use only one model in detect.py

    @zldrobit
    zldrobit committed Aug 17, 2021
    Configuration menu
    Copy the full SHA
    4e12c70 View commit details
    Browse the repository at this point in the history

  2. Add use cases in docstring of tf.py

    @zldrobit
    zldrobit committed Aug 17, 2021
    Configuration menu
    Copy the full SHA
    ea0274f View commit details
    Browse the repository at this point in the history

  3. Remove redundant stride definition

    @glenn-jocher
    glenn-jocher committed Aug 17, 2021
    Configuration menu
    Copy the full SHA
    2f63c6d View commit details
    Browse the repository at this point in the history

  4. Remove keras direct import

    @glenn-jocher
    glenn-jocher committed Aug 17, 2021
    Configuration menu
    Copy the full SHA
    c3f46c3 View commit details
    Browse the repository at this point in the history

  5. Fix check_requirements(('tensorflow>=2.4.1',))

    @glenn-jocher
    glenn-jocher committed Aug 17, 2021
    Configuration menu
    Copy the full SHA
    3133381 View commit details
    Browse the repository at this point in the history