Using the Darknet Framework with Azure ML

Note: The purpose of this repo is to serve as a sample only and should not be considered production quality. It may have future rapid iterations and additions of other algorithms/approaches for use with Azure ML. Thank you for your patience.

The algorithm used here is Darknet Tiny YOLOv4.

Prerequisites

1. Python 3.6+ installed locally
   • Recommend using a virtual environment or conda environment for this project to keep everything contained
2. Azure ML Workspace
3. Labeled data (YOLO format)
   • Convert to YOLO format as needed

Instructions

1. Clone this repository.
2. Calculate anchor boxes.
   • You may find an anchor box calculator script here.
3. Create .azureml folder in the root of this repository and download config.json from Azure ML Workspace resource in the Azure Portal to this folder.
4. Install the required packages.

pip install -r requirements_local.txt

5. Run Jupyter and then navigate to the given URL in a browser.

jupyter notebook

6. Open Train.ipynb and follow along (runs may be monitored from Azure ML Studio - also found at https://ml.azure.com).

   • Change the num_classes and anchors in the Train script writing cell.
   • Update the hyperparameter sweep values for your scenario to experiment.
   • Update the epochs for training to experiment.

7. Download the Darknet .weights and config, .cfg file from Azure ML Workspace run outputs folder.

8. Convert to ONNX

• Use this script to convert Darknet weights to ONNX (note, this script is in a different repo). e.g.:

python yolo_to_onnx.py --model yolov4-tiny-custom_final

• Use the helper_scripts/demo_onnx.py to predict with your model on a single image. - e.g.:

python helper_scripts/onnx_export_demo/demo_onnx.py --model yolov4-tiny-custom_final.onnx --image mytestimage.jpg --labels obj.names --thresh 0.7`

Todo

• Automatcially calcuate anchor boxes
• Support full sized YOLOv4
• Add hyperparameters and class number as params in train.ipynb Jupyter notebook
• Show converting to ONNX