mirror of
https://github.com/macaodha/batdetect2.git
synced 2026-04-04 15:20:19 +02:00
55 lines
1.3 KiB
Markdown
55 lines
1.3 KiB
Markdown
# How to configure ROI mapping
|
|
|
|
Use this guide to control how annotation geometry is encoded into training
|
|
targets and decoded back into boxes.
|
|
|
|
## 1) Set the default ROI mapper
|
|
|
|
The default mapper is `anchor_bbox`.
|
|
|
|
```yaml
|
|
roi:
|
|
name: anchor_bbox
|
|
anchor: bottom-left
|
|
time_scale: 1000.0
|
|
frequency_scale: 0.001163
|
|
```
|
|
|
|
## 2) Choose an anchor strategy
|
|
|
|
Typical options include `bottom-left` and `center`.
|
|
|
|
- `bottom-left` is the current default.
|
|
- `center` can be easier to reason about in some workflows.
|
|
|
|
## 3) Set scale factors intentionally
|
|
|
|
- `time_scale` controls width scaling.
|
|
- `frequency_scale` controls height scaling.
|
|
|
|
Use values that are consistent with your model setup and keep them fixed when
|
|
comparing experiments.
|
|
|
|
## 4) (Optional) override ROI mapping for specific classes
|
|
|
|
You can set class-level `roi` in `classification_targets` when needed.
|
|
|
|
```yaml
|
|
classification_targets:
|
|
- name: species_x
|
|
tags:
|
|
- key: class
|
|
value: Species X
|
|
roi:
|
|
name: anchor_bbox
|
|
anchor: center
|
|
time_scale: 1000.0
|
|
frequency_scale: 0.001163
|
|
```
|
|
|
|
## Related pages
|
|
|
|
- Target definitions: {doc}`configure-target-definitions`
|
|
- Class definitions: {doc}`define-target-classes`
|
|
- Target encoding overview: {doc}`../explanation/target-encoding-and-decoding`
|