sockeye/training.py [848:898]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        count = batches
        if self.last_count > count:
            self.init = False
        self.last_count = count
        self.samples += samples
        self.tokens += tokens

        if self.init:
            if count % self.frequency == 0:
                toc = (time.time() - self.tic)
                update_interval = batches / max(1, updates)
                updates_per_sec = self.frequency / update_interval / toc
                samples_per_sec = self.samples / toc
                tokens_per_sec = self.tokens / toc
                self.samples = 0
                self.tokens = 0

                if metrics is not None:
                    metric_values = []  # type: List[Tuple[str, float]]
                    for metric in metrics:
                        metric_values.append((metric.short_name, metric.get()))
                        if self.auto_reset:
                            metric.reset()
                    logger.info(self.msg + '%s=%f ' * len(metric_values),
                                epoch, count, samples_per_sec, tokens_per_sec, updates_per_sec, *sum(metric_values, ()))

                else:
                    logger.info(self.msg, epoch, count, samples_per_sec, tokens_per_sec, updates_per_sec)

                self.tic = time.time()
        else:
            self.init = True
            self.tic = time.time()


def safe_custom_metrics_logger(logging_function: Callable,
                               metrics: Dict,
                               global_step: int = None):
    """
    A thin wrapper for calling a custom metrics logging function, if supplied. As it uses an external function,
    it should never throw an exception. If there is no logging_function supplied, the function does nothing.
    :param logging_function: The function supplied by a caller of sockeye.train
    :param metrics: A non-empty dict of (nonempty str, float/int/bool) pairs.
    :param global_step: Optional argument, which can be used e.g. by Tensorboard.
    """
    if logging_function is None:
        return
    try:
        logging_function(metrics, global_step)
    except Exception as e:
        logging.warning("Didn't use custom metrics logger, exception '{}' occurred".format(str(e)))
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -



sockeye/training_pt.py [827:877]:
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
        count = batches
        if self.last_count > count:
            self.init = False
        self.last_count = count
        self.samples += samples
        self.tokens += tokens

        if self.init:
            if count % self.frequency == 0:
                toc = (time.time() - self.tic)
                update_interval = batches / max(1, updates)
                updates_per_sec = self.frequency / update_interval / toc
                samples_per_sec = self.samples / toc
                tokens_per_sec = self.tokens / toc
                self.samples = 0
                self.tokens = 0

                if metrics is not None:
                    metric_values = []  # type: List[Tuple[str, float]]
                    for metric in metrics:
                        metric_values.append((metric.short_name, metric.get()))
                        if self.auto_reset:
                            metric.reset()
                    logger.info(self.msg + '%s=%f ' * len(metric_values),
                                epoch, count, samples_per_sec, tokens_per_sec, updates_per_sec, *sum(metric_values, ()))

                else:
                    logger.info(self.msg, epoch, count, samples_per_sec, tokens_per_sec, updates_per_sec)

                self.tic = time.time()
        else:
            self.init = True
            self.tic = time.time()


def safe_custom_metrics_logger(logging_function: Callable,
                               metrics: Dict,
                               global_step: int = None):
    """
    A thin wrapper for calling a custom metrics logging function, if supplied. As it uses an external function,
    it should never throw an exception. If there is no logging_function supplied, the function does nothing.
    :param logging_function: The function supplied by a caller of sockeye.train
    :param metrics: A non-empty dict of (nonempty str, float/int/bool) pairs.
    :param global_step: Optional argument, which can be used e.g. by Tensorboard.
    """
    if logging_function is None:
        return
    try:
        logging_function(metrics, global_step)
    except Exception as e:
        logging.warning("Didn't use custom metrics logger, exception '{}' occurred".format(str(e)))
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -