An implementation of the Random Cut Forest data structure for sketching streaming data, with support for anomaly detection, density estimation, imputation, and more.
The encoding of state has been updated between 3.0-rc1 and 3.0-rc2. Using a model created via 3.0-rc1 in 3.0-rc2 code will create an incorrect state.
This is going to be fixed but we recommend not using older models (for reasons of efficiency, both storage as well as compute). Fixes will only apply to most recent branch in main.
The encoding of state has been updated between 3.0-rc1 and 3.0-rc2. Using a model created via 3.0-rc1 in 3.0-rc2 code will create an incorrect state.
This is going to be fixed but we recommend not using older models (for reasons of efficiency, both storage as well as compute). Fixes will only apply to most recent branch in main.