Expand Up @@ -185,7 +185,9 @@ def train_network( train_network method here. These arguments are passed to the downstream method. Some of the parameters that can be passed are * ``device`` (the CUDA device to use for training) ` * ``batch_size`` (the batch size to use while training). * ``batch_size`` (the batch size to use while training) * ``snapshot_path`` (the pose model snapshot to resume training from) * ``detector_path`` (the detector model snapshot to resume training from)
When training a top-down model, these parameters are also available for the detector, with the parameters ``detector_batch_size``, ``detector_epochs`` and Expand Down