TQDMProgressBar

class torchtnt.framework.callbacks.TQDMProgressBar(refresh_rate: int = 1)

A callback for progress bar visualization in training, evaluation, and prediction. It is initialized only on rank 0 in distributed environments.

Parameters:refresh_rate – Determines at which rate (in number of steps) the progress bars get updated.

on_eval_epoch_end(state: State, unit: EvalUnit[TEvalData]) → None

Hook called after an eval epoch ends.

on_eval_epoch_start(state: State, unit: EvalUnit[TEvalData]) → None

Hook called before a new eval epoch starts.

on_eval_step_end(state: State, unit: EvalUnit[TEvalData]) → None

Hook called after an eval step ends.

on_predict_epoch_end(state: State, unit: PredictUnit[TPredictData]) → None

Hook called after a predict epoch ends.

on_predict_epoch_start(state: State, unit: PredictUnit[TPredictData]) → None

Hook called before a new predict epoch starts.

on_predict_step_end(state: State, unit: PredictUnit[TPredictData]) → None

Hook called after a predict step ends.

on_train_epoch_end(state: State, unit: TrainUnit[TTrainData]) → None

Hook called after a train epoch ends.

on_train_epoch_start(state: State, unit: TrainUnit[TTrainData]) → None

Hook called before a new train epoch starts.

on_train_step_end(state: State, unit: TrainUnit[TTrainData]) → None

Hook called after a train step ends.