As marked by @dyuliu
We observed one limitation in current prediction-based pipelines (e.g.,
lstm_dynamic_thresholdandarima).High anomaly scores at the early indices often result in false-positive predictions. This error is likely the byproduct
of using the exponential weighted moving average function to smooth the anomaly scores. The function requires at least
the same number of observations as the size of the smoothing window before it can produce stable anomaly scores.
To overcome this limitation, add an optional masking argument. If set to True, it will replace the beginning portion of the error scores with the minimum error value.
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4