@@ -69,6 +69,14 @@ This tagger shares the same characteristics as `20230111` and differs in three i
Note the last difference was not meant to improve performance. The `20230129` was trained this way to study its performance on t5, and was kept as is after it showed better properties (less temporal leakage, fewer hunches and rolls except on stimulus onset) on t5 data.
### 12-behavior classification task
*12-behavior* refers to *run_strong*, *cast_strong*, *back_strong*, *hunch_strong*, *roll_strong*, *stop_strong*, *run_weak*, *cast_weak*, *back_weak*, *hunch_weak*, *roll_weak* and *stop_weak*.
#### `20230311-0` and `20230311`
To achieve higher similarity in tag predictions with JBM's tagger output, the `20230129` pretrained encoder was used to train the `20230311-0` tagger on the 12-behavior classification. The `20230311` tagger is essentially the same tagger, with the predicted tags remapped onto the 7-behavior classification.
## Usage
For installation, see [TaggingBackends' README](https://gitlab.pasteur.fr/nyx/TaggingBackends/-/tree/dev#recommended-installation).
