This is a simple converter which converts:
- Darknet weights (
.weights) to TensorFlow weights (.ckpt) - Darknet model (
.cfg) to TensorFlow graph (.pb,.meta)
For a full list of options:
python3 main.py -h
Provide optional argument --training to generate training graph (uses batch norm in training mode).
python3 main.py \
--cfg 'data/yolov2.cfg' \
--weights 'data/yolov2.weights' \
--output 'data/' \
--prefix 'yolov2/' \
--gpu 0
python3 main.py \
--cfg 'data/yolov2-tiny.cfg' \
--weights 'data/yolov2-tiny.weights' \
--output 'data/' \
--prefix 'yolov2-tiny/' \
--gpu 0
python3 main.py \
--cfg 'data/yolov3.cfg' \
--weights 'data/yolov3.weights' \
--output 'data/' \
--prefix 'yolov3/' \
--gpu 0
python3 main.py \
--cfg 'data/yolov3-tiny.cfg' \
--weights 'data/yolov3-tiny.weights' \
--output 'data/' \
--prefix 'yolov3-tiny/' \
--gpu 0
python3 main.py \
--cfg 'data/darknet19.cfg' \
--weights 'data/darknet19.weights' \
--output 'data/' \
--prefix 'darknet19/' \
--gpu 0
python3 main.py \
--cfg 'data/darknet19_448.cfg' \
--weights 'data/darknet19_448.weights' \
--output 'data/' \
--prefix 'darknet19_448/' \
--gpu 0
- More layer types