Try to implement YOLOv3 with TensorFlow
- transform darknet weight into npz: save_npz.py
- darknet 52 conv layer: darknet53_trainable.py
- yolo layer(loss & predict): yolo_layer.py
1. for now, only train detector layer(50 epoch, batch size: 8)
2. froze feature extractor('darknet53.conv.74')
3. TODO: finetune end2end
traindata: VOC2007 trainval & VOC2012 trainval
1. TensorFlow 1.*
2. easydict
3. PIL
4. numpy
5. matplotlib