PRTS
PRTS copied to clipboard
CompML
Unofficial Python implementation of "Precision and Recall for Time Series".
Hello, thank you for the implementation. However, as the title indicates, when I want to assess the performance of a prediction (that says that the full time-serie is normal) compared...
It would be extremely useful to have the numenta scoring algorithm for comparison purposes.
## Summary This is the sub-issue of #41 ## Goal Describe the definition by which this issue will be closed. ## Todo - [ ] task1 to do - [...
## Summary Add unit tests for TimeSeriesRecall class. ## Goal Make it possible to test all the functions of TimeSeriesRecall class reasonably. ## Todo - [ ] unit test of...
## Summary Compute precision-recall pairs for different probability thresholds. ## Goal Compute precision-recall pairs for different probability thresholds. ## Todo - [ ] add class PrecisionRecallCurve - [ ] add...
## Summary Currently, udf_gamma is always 1. However, it should be specified by users. Then, I think it is better to modify the argument ```cardinality``` as follows: ```python assert (cardinality...
## Summary Implement ```TimeSeriesPrecisionRecall``` class. ## Goal Make it enable us to calculate precision and recall simultaneously. ## Todo - [ ] ```TimeSeriesPrecisionRecall``` class - [ ] ```ts_precision_recall``` function -...
## Summary Write docstring in each function in BaseTimeSeriesMetrics. Format: ``` """Description Args: args1: type Description args2: type Description Returns: type: Description """ ``` ## Goal Write docstrings of BaseTImeSeriesMetrics...
