My implementation of YOLO v1
You should downloads pretrained wheights of Yolo architecture to reduce the amount of time required for the convergence and thus ease your burden on the training.
Then chose your dataset and put it a folder named dataset
.
python3 train.py workdir
workdir
is .
by default.
You can use your model or just load a trained model.
If you want to load another trained model put the weights in checkpoints
directory and name it best
.