YOLOX is an anchor-free version of YOLO with strong performance for object detection. YOLOX contains a Weights & Biases integration that enables you to turn on logging of training, validation and system metrics, as well as interactive validation predictions with just 1 command line argument.
Then just use the
--logger wandbcommand line argument to turn on logging with wandb. Optionally you can also pass all of the arguments that wandb.init would expect, just prepend
wandb-to the start of each argument
num_eval_imgescontrols the number of validation set images and predictions that will be logged to Weights & Biases Tables for model evaluation.
# login to wandb
# call your yolox training script with the `wandb` logger argument
python tools/train.py .... --logger wandb \
wandb-project <project-name> \
wandb-name <run-name> \
wandb-id <run-id> \
wandb-save_dir <save-dir> \
wandb-num_eval_imges <num-images> \