autodistill-yolonas's Introduction

Autodistill YOLO-NAS Module

This repository contains the code supporting the YOLO-NAS target model for use with Autodistill.

YOLO-NAS is an object detection model developed by Deci AI.

You can use autodistill to train a YOLO-NAS object detection model on a dataset of labelled images generated by the base models that autodistill supports.

Read the full Autodistill documentation.

Read the YOLO-NAS Autodistill documentation.

Installation

To use the YOLO-NAS target model, you will need to install the following dependency:

pip3 install autodistill-yolonas

Quickstart

from autodistill_yolonas import YOLONAS

target_model = YOLONAS()

# train a model
# specify the directory where your annotations (in YOLO format) are stored
target_model.train("./context_images_labeled", epochs=20)

# run inference on the new model
pred = target_model.predict("./context_images_labeled/train/images/dog-7.jpg", confidence=0.01)

License

The YOLO-NAS model is licensed under the YOLO-NAS License.

🏆 Contributing

We love your input! Please see the core Autodistill contributing guide to get started. Thank you 🙏 to all our contributors!

autodistill-yolonas's People

Contributors

Stargazers

Watchers

autodistill-yolonas's Issues

YoloNas training error

How to fix this error?

My code

from autodistill_yolonas import YOLONAS

# load the model
target_model = YOLONAS()

target_model.train("data/dataset", epochs=10)

Error:

root@5bbe96e58fd8:/workdir# python workdir/yolonas_train.py 
The console stream is logged into /root/sg_logs/console.log
[2023-07-19 05:46:53] INFO - crash_tips_setup.py - Crash tips is enabled. You can set your environment variable to CRASH_HANDLER=FALSE to disable it
Traceback (most recent call last):
  File "workdir/yolonas_train.py", line 2, in <module>
    from autodistill_yolonas import YOLONAS
  File "/opt/conda/lib/python3.7/site-packages/autodistill_yolonas/__init__.py", line 1, in <module>
    from autodistill_yolonas.yolonas_model import YOLONAS
  File "/opt/conda/lib/python3.7/site-packages/autodistill_yolonas/yolonas_model.py", line 6, in <module>
    from super_gradients.training import Trainer, models
  File "/opt/conda/lib/python3.7/site-packages/super_gradients/__init__.py", line 2, in <module>
    from super_gradients.training import losses, utils, datasets_utils, DataAugmentation, Trainer, KDTrainer, QATTrainer
  File "/opt/conda/lib/python3.7/site-packages/super_gradients/training/__init__.py", line 4, in <module>
    from super_gradients.training.sg_trainer import Trainer
  File "/opt/conda/lib/python3.7/site-packages/super_gradients/training/sg_trainer/__init__.py", line 3, in <module>
    from super_gradients.training.sg_trainer.sg_trainer import Trainer, MultiGPUMode, StrictLoad
  File "/opt/conda/lib/python3.7/site-packages/super_gradients/training/sg_trainer/sg_trainer.py", line 15, in <module>
    from piptools.scripts.sync import _get_installed_distributions
  File "/opt/conda/lib/python3.7/site-packages/piptools/scripts/sync.py", line 17, in <module>
    from .. import sync
  File "/opt/conda/lib/python3.7/site-packages/piptools/sync.py", line 11, in <module>
    from pip._internal.commands.freeze import DEV_PKGS
ImportError: cannot import name 'DEV_PKGS' from 'pip._internal.commands.freeze' (/opt/conda/lib/python3.7/site-packages/pip/_internal/commands/freeze.py)