Home

Darknet (neural network framework)

Basic terms:

• cfg-files - are structures of neural networks: https://github.com/AlexeyAB/darknet/tree/master/cfg
• weights-files - are weights for correspond cfg-file, can be downloaded from: https://pjreddie.com/darknet/

There are several tasks that are implemented in the Darknet framework:

• Yolo detection: https://pjreddie.com/darknet/yolo/
• ImageNet classification: https://pjreddie.com/darknet/imagenet/
• Text Generation: https://pjreddie.com/darknet/rnns-in-darknet/
• GO playing: https://pjreddie.com/darknet/darkgo-go-in-darknet/
• GAN Nightmare: https://pjreddie.com/darknet/nightmare/

Meaning of configuration parameters in the cfg-files

1. CFG-Parameters in the [net] section: https://github.com/AlexeyAB/darknet/wiki/CFG-Parameters-in-the-%5Bnet%5D-section

2. CFG-Parameters in the different layers: https://github.com/AlexeyAB/darknet/wiki/CFG-Parameters-in-the-different-layers

3. Train Classifier on ImageNet (ILSVRC2012): https://github.com/AlexeyAB/darknet/wiki/Train-Classifier-on-ImageNet-(ILSVRC2012)

4. Converting Yolo v3 models to TensorFlow and OpenVINO(IR) models: https://github.com/AlexeyAB/darknet/wiki/Converting-Yolo-v3-models-to-TensorFlow-and-OpenVINO(IR)-models