tensorboard#
Source code: sensai/util/tensorboard.py
- class TensorboardData(events: tensorboard.backend.event_processing.event_accumulator.EventAccumulator)[source]#
Bases:
object- get_series(tag: str, smoothing_factor: float = 0.0) pandas.Series[source]#
Gets the (smoothed) pandas Series for a specific tensorboard tag.
- Parameters:
tag – the tensorboard tag
smoothing_factor – the smoothing factor between 0 and 1 which determines the relative importance of past values. 0: no smoothing 1: maximum smoothing (all values will be equal to the first value)
- Returns:
the pandas series with the step as the index
- get_tags() List[str][source]#
Get list of available scalar tags in the events.
- Returns:
list of tag names
- get_data_frame(tags: Optional[List[str]] = None, smoothing_factor: float = 0.0) pandas.DataFrame[source]#
Gets multiple series as a DataFrame.
- Parameters:
tags – the list of tensorboard tags to consider; if None, use all
smoothing_factor – smoothing factor to apply to all series
- Returns:
DataFrame with steps as index and tags as columns
- class TensorboardSeriesComparison(tb_reference: TensorboardData, tb_current: TensorboardData, tag: str, index_start: int, index_end: int)[source]#
Bases:
object- mean_relative_difference()[source]#
Computes the difference between the current series and the reference series, relative to the reference, e.g. if the current series is on average 105% of the reference series (5% relative difference), then the value will be 0.05. Since we divide by the absolute value of the reference, this also works for negative cases, i.e. if the reference series value is -0.10 and the current series value is -0.08, then the relative difference is 0.2 (20%).
- Returns:
the mean relative difference