Update structure

docs/source/explanation/evaluation-concepts-and-matching.md

@@ -2,7 +2,8 @@
 
 Evaluation is not just "run predictions and compute one number".
 
-The reported metric depends on the evaluation task, the matching rule, and the treatment of clip boundaries and generic labels.
+The reported metric depends on the evaluation task, the matching rule, and the
+treatment of clip boundaries and generic labels.
 
 ## Task families answer different questions
 
@@ -18,7 +19,8 @@ Choose the task that matches the scientific or engineering question.
 
 ## Matching matters
 
-For sound-event-style tasks, predictions and annotations are matched using an affinity function.
+For sound-event-style tasks, predictions and annotations are matched using an
+affinity function.
 
 Important controls include:
 
@@ -27,22 +29,28 @@ Important controls include:
 - `strict_match`,
 - `ignore_start_end`.
 
-Small changes here can change the reported metric without changing the underlying predictions.
+Small changes here can change the reported metric without changing the
+underlying predictions.
 
 ## Boundary handling matters
 
-The evaluation base task can exclude events near clip boundaries through `ignore_start_end`.
+The evaluation base task can exclude events near clip boundaries through
+`ignore_start_end`.
 
 This is useful when clip boundaries make matches ambiguous.
 
 ## Generic labels can matter in classification
 
-Classification tasks can include or exclude generic targets depending on configuration.
+Classification tasks can include or exclude generic targets depending on
+configuration.
 
 That affects what counts as a valid class-level comparison.
 
 ## Related pages
 
-- Evaluate on a test set: {doc}`../tutorials/evaluate-on-a-test-set`
-- Evaluation config reference: {doc}`../reference/evaluation-config`
-- Model output and validation: {doc}`model-output-and-validation`
+- Evaluate on a test set:
+  {doc}`../tutorials/evaluate-on-a-test-set`
+- Evaluation config reference:
+  {doc}`../reference/configs/evaluation/evaluation-config`
+- Model output and validation:
+  {doc}`model-output-and-validation`

docs/source/explanation/extracted-features-and-embeddings.md

@@ -38,6 +38,6 @@ These are related ideas, but not necessarily one-to-one replacements.
 ## Related pages
 
 - Inspect detection features in Python:
-  {doc}`../how_to/inspect-detection-features-in-python`
+  {doc}`../how_to/analysis/inspect-detection-features-in-python`
 - Legacy migration guide:
   {doc}`../legacy/migration-guide`

docs/source/explanation/interpreting-formatted-outputs.md

@@ -2,21 +2,25 @@
 
 BatDetect2 can write predictions in several output formats.
 
-Those formats are different views of the same underlying detections, not different model behaviors.
+Those formats are different views of the same underlying detections, not
+different model behaviors.
 
 ## Separate the underlying detection from the serialized file
 
-Internally, the current stack works with clip-level detections containing geometry, detection score, class scores, and features.
+Internally, the current stack works with clip-level detections containing
+geometry, detection score, class scores, and features.
 
 Output formatters then serialize those detections in different ways.
 
 ## Raw outputs are richest
 
-The `raw` format preserves the broadest structured view of detections and is a good default when you want to inspect or reload predictions later.
+The `raw` format preserves the broadest structured view of detections and is a
+good default when you want to inspect or reload predictions later.
 
 ## Tabular outputs are for analysis convenience
 
-The `parquet` format is convenient for data analysis workflows, but the tabular representation is only one projection of the underlying detection object.
+The `parquet` format is convenient for data analysis workflows, but the tabular
+representation is only one projection of the underlying detection object.
 
 ## Legacy-shaped outputs are mainly for compatibility
 
@@ -26,11 +30,14 @@ Use it when you need compatibility with older downstream tools or workflows.
 
 ## The meaning does not come from the file extension
 
-Do not assume that a `.json`, `.parquet`, or `.nc` file changes what the model predicted.
+Do not assume that a `.json`, `.parquet`, or `.nc` file changes what the model
+predicted.
 
 It changes how the prediction is packaged and how much detail is retained.
 
 ## Related pages
 
-- Output formats reference: {doc}`../reference/output-formats`
-- Outputs config reference: {doc}`../reference/outputs-config`
+- Output formats reference:
+  {doc}`../reference/configs/outputs/output-formats`
+- Outputs config reference:
+  {doc}`../reference/configs/outputs/outputs-config`

docs/source/explanation/model-output-and-validation.md

@@ -1,8 +1,8 @@
 # Model output and validation
 
-BatDetect2 outputs model predictions, not ground truth. The same configuration
-can behave differently across recording conditions, species compositions, and
-acoustic environments.
+BatDetect2 outputs model predictions, not ground truth.
+The same configuration can behave differently across recording conditions,
+species compositions, and acoustic environments.
 
 ## Why threshold choice matters
 
@@ -10,8 +10,9 @@ acoustic environments.
   positives.
 - Higher thresholds reduce false positives but can miss faint calls.
 
-No threshold is universally correct. The right setting depends on your survey
-objectives and tolerance for false positives versus missed detections.
+No threshold is universally correct.
+The right setting depends on your survey objectives and tolerance for false
+positives versus missed detections.
 
 ## Why local validation is required
 
@@ -26,4 +27,4 @@ Recommended validation checks:
 3. Repeat checks across sites, seasons, and recorder setups.
 
 For practical threshold workflows, see
-{doc}`../how_to/tune-detection-threshold`.
+{doc}`../how_to/inference/tune-detection-threshold`.

docs/source/explanation/postprocessing-and-thresholds.md

@@ -1,8 +1,9 @@
 # Postprocessing and thresholds
 
 After the detector runs on a spectrogram, the model output is still a set of
-dense prediction tensors. Postprocessing turns that into a final list of call
-detections with positions, sizes, and class scores.
+dense prediction tensors.
+Postprocessing turns that into a final list of call detections with positions,
+sizes, and class scores.
 
 ## What postprocessing does
 
@@ -27,9 +28,10 @@ You can tune this behavior per run without retraining the model.
 
 ## Two common threshold controls
 
-- `detection_threshold`: minimum score required to keep a detection.
-- `classification_threshold`: minimum class score used when assigning class
-  labels.
+- `detection_threshold`:
+  minimum score required to keep a detection.
+- `classification_threshold`:
+  minimum class score used when assigning class labels.
 
 Both settings shape the final output and should be validated on reviewed local
 data.
@@ -39,5 +41,7 @@ data.
 Tune thresholds on a representative subset first, then lock settings for the
 full analysis run.
 
-- How-to: {doc}`../how_to/tune-detection-threshold`
-- CLI reference: {doc}`../reference/cli/predict`
+- How-to:
+  {doc}`../how_to/inference/tune-detection-threshold`
+- CLI reference:
+  {doc}`../reference/cli/predict`

docs/source/explanation/preprocessing-consistency.md

@@ -6,8 +6,9 @@ performance.
 ## Why consistency matters
 
 The detector is trained on spectrograms produced by a specific preprocessing
-pipeline. If inference uses different settings, the model can see a shifted
-input distribution and performance may drop.
+pipeline.
+If inference uses different settings, the model can see a shifted input
+distribution and performance may drop.
 
 Typical mismatch sources:
 
@@ -30,7 +31,8 @@ re-validate on reviewed local data.
 ## Related pages
 
 - Configure audio preprocessing:
-  {doc}`../how_to/configure-audio-preprocessing`
+  {doc}`../how_to/data/configure-audio-preprocessing`
 - Configure spectrogram preprocessing:
-  {doc}`../how_to/configure-spectrogram-preprocessing`
-- Preprocessing config reference: {doc}`../reference/preprocessing-config`
+  {doc}`../how_to/data/configure-spectrogram-preprocessing`
+- Preprocessing config reference:
+  {doc}`../reference/configs/data/preprocessing-config`

docs/source/explanation/target-encoding-and-decoding.md

@@ -11,8 +11,8 @@ At training time, the target system:
 2. assigns a classification label (or none for non-specific class matches),
 3. maps event geometry into position and size targets.
 
-This behaviour is configured through `TargetConfig`,
-`TargetClassConfig`, and ROI mapper settings.
+This behaviour is configured through `TargetConfig`, `TargetClassConfig`, and
+ROI mapper settings.
 
 ## Decoding path (model outputs -> tags and geometry)
 
@@ -24,7 +24,8 @@ annotations.
 
 ## Why this matters
 
-Target definitions are not just metadata. They directly shape:
+Target definitions are not just metadata.
+They directly shape:
 
 - what events are treated as positive examples,
 - which class names the model learns,
@@ -34,7 +35,11 @@ Small changes here can alter both training outcomes and prediction semantics.
 
 ## Related pages
 
-- Configure detection target logic: {doc}`../how_to/configure-target-definitions`
-- Configure class mapping: {doc}`../how_to/define-target-classes`
-- Configure ROI mapping: {doc}`../how_to/configure-roi-mapping`
-- Target config reference: {doc}`../reference/targets-config-workflow`
+- Configure detection target logic:
+  {doc}`../how_to/data/configure-target-definitions`
+- Configure class mapping:
+  {doc}`../how_to/data/define-target-classes`
+- Configure ROI mapping:
+  {doc}`../how_to/data/configure-roi-mapping`
+- Target config reference:
+  {doc}`../reference/configs/data/targets-config-workflow`

docs/source/faq.md

@@ -5,50 +5,64 @@
 ### Do I need Python knowledge to use batdetect2?
 
 Not much.
-If you only want to run the model on your own recordings, you can use the CLI and follow the steps in {doc}`getting_started`.
+If you only want to run the model on your own recordings, you can use the CLI
+and follow the steps in {doc}`getting_started`.
 
-Some command-line familiarity helps, but you do not need to write Python code for standard inference workflows.
+Some command-line familiarity helps, but you do not need to write Python code
+for standard inference workflows.
 
 ### Are there plans for an R version?
 
 Not currently.
-Output files are plain formats (for example CSV/JSON), so you can read and analyze them in R or other environments.
+Output files are plain formats (for example CSV/JSON), so you can read and
+analyze them in R or other environments.
 
 ### I cannot get installation working. What should I do?
 
 First, re-check {doc}`getting_started` and confirm your environment is active.
-If it still fails, open an issue with your OS, install method, and full error output: [GitHub Issues](https://github.com/macaodha/batdetect2/issues).
+If it still fails, open an issue with your OS, install method, and full error
+output:
+[GitHub Issues](https://github.com/macaodha/batdetect2/issues).
 
 ## Model behavior and performance
 
 ### The model does not perform well on my data
 
 This usually means your data distribution differs from training data.
-The best next step is to validate on reviewed local data and then fine-tune/train on your own annotations if needed.
+The best next step is to validate on reviewed local data and then
+fine-tune/train on your own annotations if needed.
 
 ### The model confuses insects/noise with bats
 
-This can happen, especially when recording conditions differ from training conditions.
+This can happen, especially when recording conditions differ from training
+conditions.
 Threshold tuning and training with local annotations can improve results.
 
-See {doc}`how_to/tune-detection-threshold`.
+See {doc}`how_to/inference/tune-detection-threshold`.
 
 ### The model struggles with feeding buzzes or social calls
 
 This is a known limitation of available training data in some settings.
-If you have high-quality annotated examples, they are valuable for improving models.
+If you have high-quality annotated examples, they are valuable for improving
+models.
 
 ### Calls in the same sequence are predicted as different species
 
-Currently we do not do any sophisticated post processing on the results output by the model.
+Currently we do not do any sophisticated post processing on the results output
+by the model.
 We return a probability associated with each species for each call.
-You can use these predictions to clean up the noisy predictions for sequences of calls.
+You can use these predictions to clean up the noisy predictions for sequences of
+calls.
 
 ### Can I trust model outputs for biodiversity conclusions?
 
-The models developed and shared as part of this repository should be used with caution.
-While they have been evaluated on held out audio data, great care should be taken when using the model outputs for any form of biodiversity assessment.
-Your data may differ, and as a result it is very strongly recommended that you validate the model first using data with known species to ensure that the outputs can be trusted.
+The models developed and shared as part of this repository should be used with
+caution.
+While they have been evaluated on held out audio data, great care should be
+taken when using the model outputs for any form of biodiversity assessment.
+Your data may differ, and as a result it is very strongly recommended that you
+validate the model first using data with known species to ensure that the
+outputs can be trusted.
 
 ### The pipeline is slow
 
@@ -65,7 +79,8 @@ You can train/fine-tune with your own annotated data and species labels.
 ### Does this work on frequency-division or zero-crossing recordings?
 
 Not directly.
-The workflow assumes audio can be converted to spectrograms from the raw waveform.
+The workflow assumes audio can be converted to spectrograms from the raw
+waveform.
 
 ### Can this be used for non-bat bioacoustics (for example insects or birds)?
 

docs/source/how_to/analysis-in-python.md

@@ -0,0 +1,11 @@
+# Analysis in Python
+
+Use this section when you want to inspect model outputs more closely from
+Python, beyond the default saved predictions.
+
+```{toctree}
+:maxdepth: 1
+
+analysis/inspect-class-scores-in-python
+analysis/inspect-detection-features-in-python
+```

docs/source/how_to/analysis/inspect-detection-features-in-python.md

@@ -1,6 +1,7 @@
 # How to inspect detection features in Python
 
-Use this guide when you want the per-detection feature vectors exposed by the current API.
+Use this guide when you want the per-detection feature vectors exposed by the
+current API.
 
 ## Get the feature vector for one detection
 
@@ -34,16 +35,21 @@ They can be useful for:
 
 They do not replace validation.
 
-They also do not automatically have a one-to-one interpretation as ecological variables.
+They also do not automatically have a one-to-one interpretation as ecological
+variables.
 
 ## Save predictions with features included
 
-If you need features on disk, use an output format that supports them, such as `raw` or `parquet`, and keep feature inclusion enabled.
+If you need features on disk, use an output format that supports them, such as
+`raw` or `parquet`, and keep feature inclusion enabled.
 
-See {doc}`save-predictions-in-different-output-formats`.
+See {doc}`../inference/save-predictions-in-different-output-formats`.
 
 ## Related pages
 
-- Understanding features and embeddings: {doc}`../explanation/extracted-features-and-embeddings`
-- Output formats reference: {doc}`../reference/output-formats`
-- API reference: {doc}`../reference/api`
+- Understanding features and embeddings:
+  {doc}`../../explanation/extracted-features-and-embeddings`
+- Output formats reference:
+  {doc}`../../reference/configs/outputs/output-formats`
+- API reference:
+  {doc}`../../reference/api`

docs/source/how_to/data-and-targets.md

@@ -0,0 +1,16 @@
+# Data and Targets
+
+Use this section when you are preparing datasets, preprocessing audio, or
+defining the targets used for training and evaluation.
+
+```{toctree}
+:maxdepth: 1
+
+data/configure-aoef-dataset
+data/import-legacy-batdetect2-annotations
+data/configure-audio-preprocessing
+data/configure-spectrogram-preprocessing
+data/configure-target-definitions
+data/define-target-classes
+data/configure-roi-mapping
+```

docs/source/how_to/data/configure-aoef-dataset.md

@@ -5,7 +5,8 @@ including exports from Whombat.
 
 ## 1) Add an AOEF source entry
 
-In your dataset config, add a source with `format: aoef`.
+In your dataset config, add a source with `format:
+aoef`.
 
 ```yaml
 sources:
@@ -49,5 +50,7 @@ batdetect2 data summary path/to/dataset.yaml
 
 ## 4) Continue to training or evaluation
 
-- For training: {doc}`../tutorials/train-a-custom-model`
-- For field-level reference: {doc}`../reference/data-sources`
+- For training:
+  {doc}`../../tutorials/train-a-custom-model`
+- For field-level reference:
+  {doc}`../../reference/configs/data/data-sources`

docs/source/how_to/data/configure-audio-preprocessing.md

@@ -63,4 +63,4 @@ before full-batch runs.
 - Spectrogram settings:
   {doc}`configure-spectrogram-preprocessing`
 - Preprocessing config reference:
-  {doc}`../reference/preprocessing-config`
+  {doc}`../../reference/configs/data/preprocessing-config`

docs/source/how_to/data/configure-roi-mapping.md

@@ -52,6 +52,9 @@ roi:
 
 ## Related pages
 
-- Target definitions: {doc}`configure-target-definitions`
-- Class definitions: {doc}`define-target-classes`
-- Target encoding overview: {doc}`../explanation/target-encoding-and-decoding`
+- Target definitions:
+  {doc}`configure-target-definitions`
+- Class definitions:
+  {doc}`define-target-classes`
+- Target encoding overview:
+  {doc}`../../explanation/target-encoding-and-decoding`

docs/source/how_to/data/configure-spectrogram-preprocessing.md

@@ -55,5 +55,7 @@ Large mismatches can degrade model performance.
 
 ## Related pages
 
-- Why consistency matters: {doc}`../explanation/preprocessing-consistency`
-- Preprocessing config reference: {doc}`../reference/preprocessing-config`
+- Why consistency matters:
+  {doc}`../../explanation/preprocessing-consistency`
+- Preprocessing config reference:
+  {doc}`../../reference/configs/data/preprocessing-config`

docs/source/how_to/data/configure-target-definitions.md

@@ -53,6 +53,9 @@ selection logic keeps the events you expect.
 
 ## Related pages
 
-- Class mapping: {doc}`define-target-classes`
-- ROI mapping: {doc}`configure-roi-mapping`
-- Targets reference: {doc}`../reference/targets-config-workflow`
+- Class mapping:
+  {doc}`define-target-classes`
+- ROI mapping:
+  {doc}`configure-roi-mapping`
+- Targets reference:
+  {doc}`../../reference/configs/data/targets-config-workflow`

docs/source/how_to/data/define-target-classes.md

@@ -1,7 +1,6 @@
 # How to define target classes
 
-Use this guide to map annotations to classification labels used during
-training.
+Use this guide to map annotations to classification labels used during training.
 
 ## 1) Add classification target entries
 
@@ -54,6 +53,9 @@ classification_targets:
 
 ## Related pages
 
-- Detection-target filtering: {doc}`configure-target-definitions`
-- ROI mapping: {doc}`configure-roi-mapping`
-- Targets config reference: {doc}`../reference/targets-config-workflow`
+- Detection-target filtering:
+  {doc}`configure-target-definitions`
+- ROI mapping:
+  {doc}`configure-roi-mapping`
+- Targets config reference:
+  {doc}`../../reference/configs/data/targets-config-workflow`

docs/source/how_to/data/import-legacy-batdetect2-annotations.md

@@ -4,12 +4,15 @@ Use this guide if your annotations are in older batdetect2 JSON formats.
 
 Two legacy formats are supported:
 
-- `batdetect2`: one annotation JSON file per recording
-- `batdetect2_file`: one merged JSON file for many recordings
+- `batdetect2`:
+  one annotation JSON file per recording
+- `batdetect2_file`:
+  one merged JSON file for many recordings
 
 ## 1) Choose the correct source format
 
-Directory-based annotations (`format: batdetect2`):
+Directory-based annotations (`format:
+batdetect2`):
 
 ```yaml
 sources:
@@ -19,7 +22,8 @@ sources:
     annotations_dir: /path/to/annotation_json_dir
 ```
 
-Merged annotation file (`format: batdetect2_file`):
+Merged annotation file (`format:
+batdetect2_file`):
 
 ```yaml
 sources:
@@ -61,6 +65,9 @@ batdetect2 data convert path/to/dataset.yaml --output path/to/output.json
 
 ## 4) Continue with current workflows
 
-- Run predictions: {doc}`run-batch-predictions`
-- Train on imported data: {doc}`../tutorials/train-a-custom-model`
-- Field-level reference: {doc}`../reference/data-sources`
+- Run predictions:
+  {doc}`../inference/run-batch-predictions`
+- Train on imported data:
+  {doc}`../../tutorials/train-a-custom-model`
+- Field-level reference:
+  {doc}`../../reference/configs/data/data-sources`

docs/source/how_to/evaluation.md

@@ -0,0 +1,11 @@
+# Evaluation
+
+Use this section when you want to choose evaluation tasks or understand the
+artifacts produced by an evaluation run.
+
+```{toctree}
+:maxdepth: 1
+
+evaluation/choose-and-configure-evaluation-tasks
+evaluation/interpret-evaluation-outputs
+```

docs/source/how_to/index.md

@@ -6,25 +6,15 @@ tutorial.
 Use this section when you already know the basic workflow and want help with one
 specific task.
 
+The guides are grouped by topic so you can jump straight to the part of the
+workflow you are working on.
+
 ```{toctree}
 :maxdepth: 1
 
-choose-a-model
-choose-an-inference-input-mode
-run-batch-predictions
-tune-inference-clipping
-tune-detection-threshold
-inspect-class-scores-in-python
-inspect-detection-features-in-python
-save-predictions-in-different-output-formats
-fine-tune-from-a-checkpoint
-choose-and-configure-evaluation-tasks
-interpret-evaluation-outputs
-configure-aoef-dataset
-import-legacy-batdetect2-annotations
-configure-audio-preprocessing
-configure-spectrogram-preprocessing
-configure-target-definitions
-define-target-classes
-configure-roi-mapping
+inference-and-outputs
+analysis-in-python
+training-and-fine-tuning
+evaluation
+data-and-targets
 ```

docs/source/how_to/inference-and-outputs.md

@@ -0,0 +1,15 @@
+# Inference and Outputs
+
+Use this section when you want to run predictions, choose how inference runs, or
+control what gets written to disk.
+
+```{toctree}
+:maxdepth: 1
+
+inference/choose-a-model
+inference/choose-an-inference-input-mode
+inference/run-batch-predictions
+inference/tune-inference-clipping
+inference/tune-detection-threshold
+inference/save-predictions-in-different-output-formats
+```

docs/source/how_to/inference/choose-a-model.md

@@ -101,12 +101,12 @@ api = BatDetect2API.from_checkpoint(
 ## Related pages
 
 - Run inference on a folder:
-  {doc}`../tutorials/run-inference-on-folder`
+  {doc}`../../tutorials/run-inference-on-folder`
 - `BatDetect2API` reference:
-  {doc}`../reference/api`
+  {doc}`../../reference/api`
 - Process command reference:
-  {doc}`../reference/cli/predict`
+  {doc}`../../reference/cli/predict`
 - Train a custom model:
-  {doc}`../tutorials/train-a-custom-model`
+  {doc}`../../tutorials/train-a-custom-model`
 - Fine-tune from a checkpoint:
-  {doc}`fine-tune-from-a-checkpoint`
+  {doc}`../training/fine-tune-from-a-checkpoint`

docs/source/how_to/inference/choose-an-inference-input-mode.md

@@ -68,4 +68,4 @@ recording paths before inference.
 - Tune inference clipping:
   {doc}`tune-inference-clipping`
 - Process command reference:
-  {doc}`../reference/cli/predict`
+  {doc}`../../reference/cli/predict`

docs/source/how_to/inference/tune-inference-clipping.md

@@ -66,8 +66,8 @@ Check a reviewed subset before applying clipping changes to a full project.
 ## Related pages
 
 - Inference config reference:
-  {doc}`../reference/inference-config`
+  {doc}`../../reference/configs/inference/inference-config`
 - Run batch predictions:
   {doc}`run-batch-predictions`
 - Understanding the pipeline:
-  {doc}`../explanation/pipeline-overview`
+  {doc}`../../explanation/pipeline-overview`

docs/source/how_to/training-and-fine-tuning.md

@@ -0,0 +1,10 @@
+# Training and Fine-Tuning
+
+Use this section when you already have a training workflow and want help with a
+specific fine-tuning step.
+
+```{toctree}
+:maxdepth: 1
+
+training/fine-tune-from-a-checkpoint
+```

docs/source/index.md

@@ -25,8 +25,8 @@ For more detail on the underlying approach, see the pre-print:
 [Towards a General Approach for Bat Echolocation Detection and Classification](https://www.biorxiv.org/content/10.1101/2022.12.14.520490v1)
 
 ```{warning}
-Treat outputs as model predictions, not ground truth.
-Always validate on reviewed local data before using results for ecological inference.
+Model outputs are predictions, not ground truth, and may be incorrect in subtle or significant ways.
+We strongly encourage validating results on reviewed local data before using them for downstream analyses.
 ```
 
 ## What can I do with it?

docs/source/reference/api.md

@@ -34,6 +34,6 @@ Defined in `batdetect2.api_v2`.
 - Python tutorial:
   {doc}`../tutorials/integrate-with-a-python-pipeline`
 - Outputs config reference:
-  {doc}`outputs-config`
+  {doc}`configs/outputs/outputs-config`
 - Output formats reference:
-  {doc}`output-formats`
+  {doc}`configs/outputs/output-formats`

docs/source/reference/cli/index.md

@@ -33,8 +33,8 @@ Prefer `batdetect2 process directory` for new workflows.
 ## Related pages
 
 - {doc}`../../tutorials/run-inference-on-folder`
-- {doc}`../../how_to/run-batch-predictions`
-- {doc}`../../how_to/tune-detection-threshold`
+- {doc}`../../how_to/inference/run-batch-predictions`
+- {doc}`../../how_to/inference/tune-detection-threshold`
 - {doc}`../configs`
 
 ```{toctree}

docs/source/reference/configs.rst

@@ -3,16 +3,23 @@ Config reference
 
 BatDetect2 uses separate config objects for different workflow surfaces.
 
-Use the dedicated reference pages for each config family:
+Use this section when you need exact config fields for training, inference,
+evaluation, outputs, preprocessing, postprocessing, or target definitions.
 
-- model config
-- training config
-- logging config
-- inference config
-- evaluation config
-- outputs config
-- preprocessing config
-- postprocess config
-- targets config workflow
+.. toctree::
+   :maxdepth: 1
+
+   configs/training/model-config
+   configs/training/training-config
+   configs/training/logging-config
+   configs/inference/inference-config
+   configs/evaluation/evaluation-config
+   configs/outputs/outputs-config
+   configs/outputs/output-formats
+   configs/outputs/output-transforms
+   configs/data/data-sources
+   configs/data/preprocessing-config
+   configs/data/postprocess-config
+   configs/data/targets-config-workflow
 
 Example config files live under `example_data/configs/`.

docs/source/reference/configs/data/data-sources.md

@@ -24,16 +24,19 @@ Optional fields:
 - `description`
 - `filter`
 
-`filter` is only used when `annotations_path` points to an
-`AnnotationProject`.
+`filter` is only used when `annotations_path` points to an `AnnotationProject`.
 
 AOEF filter options:
 
-- `only_completed` (default: `true`)
-- `only_verified` (default: `false`)
-- `exclude_issues` (default: `true`)
+- `only_completed` (default:
+  `true`)
+- `only_verified` (default:
+  `false`)
+- `exclude_issues` (default:
+  `true`)
 
-Use `filter: null` to disable project filtering.
+Use `filter:
+null` to disable project filtering.
 
 ## Legacy per-file (`format: batdetect2`)
 
@@ -65,12 +68,15 @@ Optional fields:
 
 Legacy filter options:
 
-- `only_annotated` (default: `true`)
-- `exclude_issues` (default: `true`)
+- `only_annotated` (default:
+  `true`)
+- `exclude_issues` (default:
+  `true`)
 
-Use `filter: null` to disable filtering.
+Use `filter:
+null` to disable filtering.
 
 ## Related guides
 
-- {doc}`../how_to/configure-aoef-dataset`
-- {doc}`../how_to/import-legacy-batdetect2-annotations`
+- {doc}`../../../how_to/data/configure-aoef-dataset`
+- {doc}`../../../how_to/data/import-legacy-batdetect2-annotations`

docs/source/reference/configs/data/postprocess-config.md

@@ -18,14 +18,20 @@ Defined in `batdetect2.postprocess.config`.
 
 ## Defaults
 
-- `detection_threshold`: `0.01`
-- `classification_threshold`: `0.1`
-- `top_k_per_sec`: `100`
+- `detection_threshold`:
+  `0.01`
+- `classification_threshold`:
+  `0.1`
+- `top_k_per_sec`:
+  `100`
 
 `nms_kernel_size` defaults to the library constant used by the NMS module.
 
 ## Related pages
 
-- Threshold behaviour: {doc}`../explanation/postprocessing-and-thresholds`
-- Threshold tuning workflow: {doc}`../how_to/tune-detection-threshold`
-- CLI predict options: {doc}`cli/predict`
+- Threshold behaviour:
+  {doc}`../../../explanation/postprocessing-and-thresholds`
+- Threshold tuning workflow:
+  {doc}`../../../how_to/inference/tune-detection-threshold`
+- CLI predict options:
+  {doc}`../../cli/predict`

docs/source/reference/configs/data/preprocessing-config.md

@@ -8,9 +8,12 @@ Defined in `batdetect2.audio.loader`.
 
 Fields:
 
-- `samplerate` (int): target audio sample rate in Hz.
-- `resample.enabled` (bool): whether to resample loaded audio.
-- `resample.method` (`poly` or `fourier`): resampling method.
+- `samplerate` (int):
+  target audio sample rate in Hz.
+- `resample.enabled` (bool):
+  whether to resample loaded audio.
+- `resample.method` (`poly` or `fourier`):
+  resampling method.
 
 ## Model preprocessing config (`PreprocessingConfig`)
 
@@ -18,11 +21,16 @@ Defined in `batdetect2.preprocess.config`.
 
 Top-level fields:
 
-- `audio_transforms`: ordered waveform transforms.
-- `stft`: STFT parameters.
-- `frequencies`: spectrogram frequency range.
-- `spectrogram_transforms`: ordered spectrogram transforms.
-- `size`: final resize settings.
+- `audio_transforms`:
+  ordered waveform transforms.
+- `stft`:
+  STFT parameters.
+- `frequencies`:
+  spectrogram frequency range.
+- `spectrogram_transforms`:
+  ordered spectrogram transforms.
+- `size`:
+  final resize settings.
 
 ### `audio_transforms` built-ins
 
@@ -44,7 +52,8 @@ Top-level fields:
 ### `spectrogram_transforms` built-ins
 
 - `pcen`
-- `scale_amplitude` (`scale: db|power`)
+- `scale_amplitude` (`scale:
+  db|power`)
 - `spectral_mean_subtraction`
 - `peak_normalize`
 
@@ -55,7 +64,9 @@ Top-level fields:
 
 ## Related pages
 
-- Audio preprocessing how-to: {doc}`../how_to/configure-audio-preprocessing`
+- Audio preprocessing how-to:
+  {doc}`../../../how_to/data/configure-audio-preprocessing`
 - Spectrogram preprocessing how-to:
-  {doc}`../how_to/configure-spectrogram-preprocessing`
-- Why consistency matters: {doc}`../explanation/preprocessing-consistency`
+  {doc}`../../../how_to/data/configure-spectrogram-preprocessing`
+- Why consistency matters:
+  {doc}`../../../explanation/preprocessing-consistency`

docs/source/reference/configs/data/targets-config-workflow.md

@@ -0,0 +1,78 @@
+# Targets config workflow reference
+
+This page summarizes the target-definition configuration used by batdetect2.
+
+## `TargetConfig`
+
+Defined in `batdetect2.targets.config`.
+
+Fields:
+
+- `detection_target`:
+  one `TargetClassConfig` defining detection eligibility.
+- `classification_targets`:
+  list of `TargetClassConfig` entries for class encoding/decoding.
+- `roi`:
+  ROI mapping config with `default` mapper and optional per-class `overrides`.
+
+## `TargetClassConfig`
+
+Defined in `batdetect2.targets.classes`.
+
+Fields:
+
+- `name`:
+  class label name.
+- `tags`:
+  tag list used for matching (shortcut for `match_if`).
+- `match_if`:
+  explicit condition config (`match_if` is accepted as alias).
+- `assign_tags`:
+  tags used when decoding this class.
+
+`tags` and `match_if` are mutually exclusive.
+
+## Supported condition config types
+
+Built from `batdetect2.data.conditions`.
+
+- `has_tag`
+- `has_all_tags`
+- `has_any_tag`
+- `duration`
+- `frequency`
+- `all_of`
+- `any_of`
+- `not`
+
+## ROI mapper config
+
+`roi.default` and each `roi.overrides.<class_name>` entry support built-in
+mappers including:
+
+- `anchor_bbox`
+- `peak_energy_bbox`
+
+Key `anchor_bbox` fields:
+
+- `anchor`
+- `time_scale`
+- `frequency_scale`
+
+Top-level ROI mapping shape:
+
+- `default`:
+  fallback mapper used for all classes.
+- `overrides`:
+  optional mapping from class name to mapper config.
+
+## Related pages
+
+- Detection target setup:
+  {doc}`../../../how_to/data/configure-target-definitions`
+- Class setup:
+  {doc}`../../../how_to/data/define-target-classes`
+- ROI setup:
+  {doc}`../../../how_to/data/configure-roi-mapping`
+- Concept overview:
+  {doc}`../../../explanation/target-encoding-and-decoding`

docs/source/reference/configs/evaluation/evaluation-config.md

@@ -1,6 +1,7 @@
 # Evaluation config reference
 
-`EvaluationConfig` defines which evaluation tasks run and which plots they generate.
+`EvaluationConfig` defines which evaluation tasks run and which plots they
+generate.
 
 Defined in `batdetect2.evaluate.config`.
 
@@ -41,6 +42,9 @@ The default evaluation config starts with:
 
 ## Related pages
 
-- Choose and configure evaluation tasks: {doc}`../how_to/choose-and-configure-evaluation-tasks`
-- Evaluation concepts: {doc}`../explanation/evaluation-concepts-and-matching`
-- Evaluate CLI reference: {doc}`cli/evaluate`
+- Choose and configure evaluation tasks:
+  {doc}`../../../how_to/evaluation/choose-and-configure-evaluation-tasks`
+- Evaluation concepts:
+  {doc}`../../../explanation/evaluation-concepts-and-matching`
+- Evaluate CLI reference:
+  {doc}`../../cli/evaluate`

docs/source/reference/configs/inference/inference-config.md

@@ -1,6 +1,7 @@
 # Inference config reference
 
-`InferenceConfig` controls how files are clipped and batched during prediction-time workflows.
+`InferenceConfig` controls how files are clipped and batched during
+prediction-time workflows.
 
 Defined in `batdetect2.inference.config`.
 
@@ -37,5 +38,7 @@ Override `InferenceConfig` when:
 
 ## Related pages
 
-- Tune inference clipping: {doc}`../how_to/tune-inference-clipping`
-- Predict CLI reference: {doc}`cli/predict`
+- Tune inference clipping:
+  {doc}`../../../how_to/inference/tune-inference-clipping`
+- Predict CLI reference:
+  {doc}`../../cli/predict`

docs/source/reference/configs/outputs/output-formats.md

@@ -70,6 +70,6 @@ It can also write legacy `_cnn_features.csv` sidecars when
 - Outputs config:
   {doc}`outputs-config`
 - Save predictions in different output formats:
-  {doc}`../how_to/save-predictions-in-different-output-formats`
+  {doc}`../../../how_to/inference/save-predictions-in-different-output-formats`
 - Understanding formatted outputs:
-  {doc}`../explanation/interpreting-formatted-outputs`
+  {doc}`../../../explanation/interpreting-formatted-outputs`

docs/source/reference/configs/outputs/output-transforms.md

@@ -33,5 +33,7 @@ Built-in examples include:
 
 ## Related pages
 
-- Outputs config: {doc}`outputs-config`
-- Understanding outputs: {doc}`../explanation/interpreting-formatted-outputs`
+- Outputs config:
+  {doc}`outputs-config`
+- Understanding outputs:
+  {doc}`../../../explanation/interpreting-formatted-outputs`

docs/source/reference/configs/outputs/outputs-config.md

@@ -38,4 +38,4 @@ If no outputs config is provided, the CLI still uses its command defaults.
 - Output transforms:
   {doc}`output-transforms`
 - Save predictions in different output formats:
-  {doc}`../how_to/save-predictions-in-different-output-formats`
+  {doc}`../../../how_to/inference/save-predictions-in-different-output-formats`

docs/source/reference/configs/training/model-config.md

@@ -30,8 +30,8 @@ Example files live under `example_data/configs/`, including
 ## Related pages
 
 - Preprocessing config:
-  {doc}`preprocessing-config`
+  {doc}`../data/preprocessing-config`
 - Postprocess config:
-  {doc}`postprocess-config`
+  {doc}`../data/postprocess-config`
 - Train command reference:
-  {doc}`cli/train`
+  {doc}`../../cli/train`

docs/source/reference/configs/training/training-config.md

@@ -43,8 +43,8 @@ Example files live under `example_data/configs/`, including
 ## Related pages
 
 - Evaluation config:
-  {doc}`evaluation-config`
+  {doc}`../evaluation/evaluation-config`
 - Train command reference:
-  {doc}`cli/train`
+  {doc}`../../cli/train`
 - Fine-tune from a checkpoint:
-  {doc}`../how_to/fine-tune-from-a-checkpoint`
+  {doc}`../../../how_to/training/fine-tune-from-a-checkpoint`

docs/source/reference/index.md

@@ -11,18 +11,6 @@ details, or Python API entries.
 cli/index
 api
 detections
-model-config
-training-config
-logging-config
-inference-config
-evaluation-config
-outputs-config
-output-formats
-output-transforms
-data-sources
-preprocessing-config
-postprocess-config
-targets-config-workflow
 configs
 targets
 ```

docs/source/reference/targets-config-workflow.md

@@ -1,67 +0,0 @@
-# Targets config workflow reference
-
-This page summarizes the target-definition configuration used by batdetect2.
-
-## `TargetConfig`
-
-Defined in `batdetect2.targets.config`.
-
-Fields:
-
-- `detection_target`: one `TargetClassConfig` defining detection eligibility.
-- `classification_targets`: list of `TargetClassConfig` entries for class
-  encoding/decoding.
-- `roi`: ROI mapping config with `default` mapper and optional per-class
-  `overrides`.
-
-## `TargetClassConfig`
-
-Defined in `batdetect2.targets.classes`.
-
-Fields:
-
-- `name`: class label name.
-- `tags`: tag list used for matching (shortcut for `match_if`).
-- `match_if`: explicit condition config (`match_if` is accepted as alias).
-- `assign_tags`: tags used when decoding this class.
-
-`tags` and `match_if` are mutually exclusive.
-
-## Supported condition config types
-
-Built from `batdetect2.data.conditions`.
-
-- `has_tag`
-- `has_all_tags`
-- `has_any_tag`
-- `duration`
-- `frequency`
-- `all_of`
-- `any_of`
-- `not`
-
-## ROI mapper config
-
-`roi.default` and each `roi.overrides.<class_name>` entry support built-in
-mappers including:
-
-- `anchor_bbox`
-- `peak_energy_bbox`
-
-Key `anchor_bbox` fields:
-
-- `anchor`
-- `time_scale`
-- `frequency_scale`
-
-Top-level ROI mapping shape:
-
-- `default`: fallback mapper used for all classes.
-- `overrides`: optional mapping from class name to mapper config.
-
-## Related pages
-
-- Detection target setup: {doc}`../how_to/configure-target-definitions`
-- Class setup: {doc}`../how_to/define-target-classes`
-- ROI setup: {doc}`../how_to/configure-roi-mapping`
-- Concept overview: {doc}`../explanation/target-encoding-and-decoding`

docs/source/tutorials/evaluate-on-a-test-set.md

@@ -41,7 +41,8 @@ For an example, see `example_data/dataset.yaml`.
 
 If you need help creating the dataset config, follow the dataset section in
 {doc}`train-a-custom-model`.
-For more detail on dataset source formats, see {doc}`../reference/data-sources`.
+For more detail on dataset source formats, see
+{doc}`../reference/configs/data/data-sources`.
 
 Use a dataset that was not used for training or tuning.
 
@@ -56,7 +57,7 @@ batdetect2 evaluate \
 
 If you do not pass `--model`, BatDetect2 uses the built-in default UK model.
 If you want to choose a different checkpoint, alias, or Hugging Face model, see
-{doc}`../how_to/choose-a-model`.
+{doc}`../how_to/inference/choose-a-model`.
 
 If you want to save the results somewhere else, add `--output-dir`:
 
@@ -121,17 +122,17 @@ So, depending on your evaluation config, you may see files such as:
 - saved prediction files.
 
 If you want to control which tasks run and which plots are generated, see
-{doc}`../reference/evaluation-config` and
-{doc}`../how_to/choose-and-configure-evaluation-tasks`.
+{doc}`../reference/configs/evaluation/evaluation-config` and
+{doc}`../how_to/evaluation/choose-and-configure-evaluation-tasks`.
 
 ## Common next steps
 
 - Choose a different model:
-  {doc}`../how_to/choose-a-model`
+  {doc}`../how_to/inference/choose-a-model`
 - Configure evaluation tasks:
-  {doc}`../how_to/choose-and-configure-evaluation-tasks`
+  {doc}`../how_to/evaluation/choose-and-configure-evaluation-tasks`
 - Interpret evaluation artifacts:
-  {doc}`../how_to/interpret-evaluation-outputs`
+  {doc}`../how_to/evaluation/interpret-evaluation-outputs`
 - Learn the evaluation concepts:
   {doc}`../explanation/evaluation-concepts-and-matching`
 - Check full evaluate options:

docs/source/tutorials/integrate-with-a-python-pipeline.md

@@ -1,10 +1,12 @@
 # Integrate with a Python pipeline
 
-This tutorial shows a simple Python workflow for loading audio, running BatDetect2, and inspecting the detections.
+This tutorial shows a simple Python workflow for loading audio, running
+BatDetect2, and inspecting the detections.
 
 Use it when you want to work directly in Python rather than through the CLI.
 
-If you mainly want to run the model on recordings, start with {doc}`run-inference-on-folder` instead.
+If you mainly want to run the model on recordings, start with
+{doc}`run-inference-on-folder` instead.
 
 ## Before you start
 
@@ -21,7 +23,8 @@ By the end of this tutorial you will have:
 - run inference on one file,
 - inspected detections, scores, and features,
 - used lower-level audio and spectrogram methods for more control,
-- identified the next API workflows for batch processing, training, fine-tuning, and evaluation.
+- identified the next API workflows for batch processing, training, fine-tuning,
+  and evaluation.
 
 ## 1. Create the API instance
 
@@ -34,11 +37,13 @@ from batdetect2 import BatDetect2API
 api = BatDetect2API.from_checkpoint()
 ```
 
-If you want to use a different checkpoint later, see {doc}`../how_to/choose-a-model`.
+If you want to use a different checkpoint later, see
+{doc}`../how_to/inference/choose-a-model`.
 
 ## 2. Run inference on one file
 
-`process_file` is the simplest Python entry point when you want one prediction object per recording.
+`process_file` is the simplest Python entry point when you want one prediction
+object per recording.
 
 ```python
 from batdetect2 import BatDetect2API
@@ -57,7 +62,8 @@ for detection in prediction.detections:
 `prediction` is a `ClipDetections` object.
 See {doc}`../reference/detections` for the full reference.
 
-Very briefly, `ClipDetections` represents all detections for one processed clip or recording.
+Very briefly, `ClipDetections` represents all detections for one processed clip
+or recording.
 It includes:
 
 - the clip metadata,
@@ -76,10 +82,13 @@ Each `Detection` includes:
 
 The detection score and the class scores answer different questions.
 
-- `detection_score` is about whether the model thinks there is a call at that time-frequency location.
-- `class_scores` are about which class the model prefers for that detected event.
+- `detection_score` is about whether the model thinks there is a call at that
+  time-frequency location.
+- `class_scores` are about which class the model prefers for that detected
+  event.
 
-So a detection can have a fairly strong detection score, but still have a more uncertain class ranking.
+So a detection can have a fairly strong detection score, but still have a more
+uncertain class ranking.
 
 ```python
 for detection in prediction.detections:
@@ -90,7 +99,8 @@ for detection in prediction.detections:
         print(f"  {class_name}: {score:.3f}")
 ```
 
-If you want more detail on class-score inspection, see {doc}`../how_to/inspect-class-scores-in-python`.
+If you want more detail on class-score inspection, see
+{doc}`../how_to/analysis/inspect-class-scores-in-python`.
 
 ## 5. Inspect the detection features
 
@@ -104,15 +114,20 @@ They can be useful for things like:
 - comparing detections across files,
 - building downstream analysis pipelines.
 
-They are useful descriptors, but they are not direct ecological labels by themselves.
+They are useful descriptors, but they are not direct ecological labels by
+themselves.
 
-For more detail, see {doc}`../how_to/inspect-detection-features-in-python` and {doc}`../explanation/extracted-features-and-embeddings`.
+For more detail, see
+{doc}`../how_to/analysis/inspect-detection-features-in-python` and
+{doc}`../explanation/extracted-features-and-embeddings`.
 
 ## 6. Use lower-level audio and spectrogram methods for more control
 
-If you want finer control over what gets processed and when, the API also lets you work step by step.
+If you want finer control over what gets processed and when, the API also lets
+you work step by step.
 
-For example, you can load the audio yourself, inspect the waveform length, generate the spectrogram, and then run detection on that spectrogram:
+For example, you can load the audio yourself, inspect the waveform length,
+generate the spectrogram, and then run detection on that spectrogram:
 
 ```python
 from batdetect2 import BatDetect2API
@@ -136,12 +151,14 @@ This is helpful when you want to:
 - control which audio segment is processed,
 - run only part of the pipeline in custom code.
 
-You can also call `process_audio(audio)` directly if you already have the waveform array in memory.
+You can also call `process_audio(audio)` directly if you already have the
+waveform array in memory.
 
 ## 7. Use the wider API workflows
 
 The Python API is not only for single-file inference.
-It also exposes methods for batch processing, training, evaluation, and fine-tuning.
+It also exposes methods for batch processing, training, evaluation, and
+fine-tuning.
 
 Examples:
 
@@ -152,9 +169,15 @@ Examples:
 
 Useful next pages:
 
-- Choose a different model: {doc}`../how_to/choose-a-model`
-- Run batch predictions: {doc}`../how_to/run-batch-predictions`
-- Train a custom model: {doc}`train-a-custom-model`
-- Evaluate on a test set: {doc}`evaluate-on-a-test-set`
-- Fine-tune from a checkpoint: {doc}`../how_to/fine-tune-from-a-checkpoint`
-- API reference: {doc}`../reference/api`
+- Choose a different model:
+  {doc}`../how_to/inference/choose-a-model`
+- Run batch predictions:
+  {doc}`../how_to/inference/run-batch-predictions`
+- Train a custom model:
+  {doc}`train-a-custom-model`
+- Evaluate on a test set:
+  {doc}`evaluate-on-a-test-set`
+- Fine-tune from a checkpoint:
+  {doc}`../how_to/training/fine-tune-from-a-checkpoint`
+- API reference:
+  {doc}`../reference/api`

docs/source/tutorials/run-inference-on-folder.md

@@ -1,10 +1,13 @@
 # Run BatDetect2 on a folder of audio files
 
-This tutorial shows how to run BatDetect2 on a folder of recordings from the command line.
+This tutorial shows how to run BatDetect2 on a folder of recordings from the
+command line.
 
-Use it when you want a first pass over a folder of audio recordings and want to see what BatDetect2 finds.
+Use it when you want a first pass over a folder of audio recordings and want to
+see what BatDetect2 finds.
 
-If you want to follow the tutorial exactly, you can use the example recordings that come with the repository.
+If you want to follow the tutorial exactly, you can use the example recordings
+that come with the repository.
 
 ## Before you start
 
@@ -18,7 +21,8 @@ If you have not installed BatDetect2 yet, start with {doc}`../getting_started`.
 
 ## Optional: use the repository example files
 
-If you want to follow the steps with the same paths shown here, clone the repository and move into it:
+If you want to follow the steps with the same paths shown here, clone the
+repository and move into it:
 
 ```bash
 git clone https://github.com/macaodha/batdetect2.git
@@ -57,7 +61,8 @@ project/
 If `outputs/` does not exist yet, that is fine.
 BatDetect2 can create it.
 
-If you are using the repository example files, your layout already looks like this:
+If you are using the repository example files, your layout already looks like
+this:
 
 ```text
 batdetect2/
@@ -95,7 +100,8 @@ What this does:
 You do not need to choose a model for this first run.
 If you do nothing, BatDetect2 uses the built-in default UK model.
 
-If you want to use a different model later, see {doc}`../how_to/choose-a-model`.
+If you want to use a different model later, see
+{doc}`../how_to/inference/choose-a-model`.
 
 ## 3. Check the output files
 
@@ -104,7 +110,8 @@ After the command finishes, look in your output folder.
 By default, the CLI writes predictions in the `batdetect2` output format.
 This is a JSON-based format used for BatDetect2-style outputs.
 
-With the default settings, you will usually see one `.json` file and one `_detections.csv` file per recording.
+With the default settings, you will usually see one `.json` file and one
+`_detections.csv` file per recording.
 
 For the repository example run, that means files like:
 
@@ -141,16 +148,20 @@ One of the JSON files will look roughly like this:
 
 Very briefly:
 
-- `annotated: false` means this is a prediction file, not a reviewed annotation file.
+- `annotated:
+  false` means this is a prediction file, not a reviewed annotation file.
 - `annotation` holds the list of detections.
-- Each detection includes a predicted class, detection score, class score, time bounds, and frequency bounds.
+- Each detection includes a predicted class, detection score, class score, time
+  bounds, and frequency bounds.
 
 For more detail, see {doc}`../explanation/interpreting-formatted-outputs`.
-If you want to save results in another format, see {doc}`../how_to/save-predictions-in-different-output-formats`.
+If you want to save results in another format, see
+{doc}`../how_to/inference/save-predictions-in-different-output-formats`.
 
 ## 4. Run the same folder with a higher threshold
 
-If you want, you can also run the same folder again with a higher detection threshold and save that run in a separate output folder.
+If you want, you can also run the same folder again with a higher detection
+threshold and save that run in a separate output folder.
 
 ```bash
 batdetect2 process directory \
@@ -206,12 +217,18 @@ batdetect2 process file_list \
     example_outputs/selected_outputs
 ```
 
-This is useful when your recordings are spread across folders, or when you only want to run a chosen subset.
+This is useful when your recordings are spread across folders, or when you only
+want to run a chosen subset.
 
 ## Common next steps
 
-- If your recordings are not all in one folder, or you want to compare input modes, see {doc}`../how_to/choose-an-inference-input-mode`.
-- If you want to save results in another format, see {doc}`../how_to/save-predictions-in-different-output-formats`.
-- If you want to choose a different model, see {doc}`../how_to/choose-a-model`.
-- If you already write code and want more control from Python, see {doc}`integrate-with-a-python-pipeline`.
-- If you want the full command reference, including `--model`, see {doc}`../reference/cli/predict`.
+- If your recordings are not all in one folder, or you want to compare input
+  modes, see {doc}`../how_to/inference/choose-an-inference-input-mode`.
+- If you want to save results in another format, see
+  {doc}`../how_to/inference/save-predictions-in-different-output-formats`.
+- If you want to choose a different model, see
+  {doc}`../how_to/inference/choose-a-model`.
+- If you already write code and want more control from Python, see
+  {doc}`integrate-with-a-python-pipeline`.
+- If you want the full command reference, including `--model`, see
+  {doc}`../reference/cli/predict`.

docs/source/tutorials/train-a-custom-model.md

@@ -1,8 +1,10 @@
 # Train a custom model
 
-This tutorial walks through a first custom training run using your own annotations.
+This tutorial walks through a first custom training run using your own
+annotations.
 
-Use it when you already have labelled recordings and want to train a model for your own data.
+Use it when you already have labelled recordings and want to train a model for
+your own data.
 
 ## Before you start
 
@@ -19,7 +21,8 @@ Use {doc}`run-inference-on-folder` for that.
 
 ## Optional: use the repository example files
 
-If you want to follow the steps with the same files shown here, clone the repository and move into it:
+If you want to follow the steps with the same files shown here, clone the
+repository and move into it:
 
 ```bash
 git clone https://github.com/macaodha/batdetect2.git
@@ -40,10 +43,12 @@ By the end of this tutorial you will have:
 
 The dataset config explicitly declares what data you want to use for training.
 It is a YAML file.
-If YAML is new to you, see [Learn YAML in Y Minutes](https://learnxinyminutes.com/yaml/).
+If YAML is new to you, see
+[Learn YAML in Y Minutes](https://learnxinyminutes.com/yaml/).
 
 In the dataset config, you list one or more data sources.
-Each source tells `batdetect2` where the audio recordings live and where the matching annotations are stored.
+Each source tells `batdetect2` where the audio recordings live and where the
+matching annotations are stored.
 
 BatDetect2 can read annotations from different source formats.
 In this example, we use the example data in the `batdetect2` format.
@@ -62,21 +67,25 @@ sources:
 ```
 
 For your own project, the main thing to change is the file paths.
-If you have several collections of recordings, you can add more than one source to the same dataset config.
+If you have several collections of recordings, you can add more than one source
+to the same dataset config.
 That lets you describe the full training data you want to use in one place.
 
-If you need more detail on dataset source formats, see {doc}`../reference/data-sources`.
+If you need more detail on dataset source formats, see
+{doc}`../reference/configs/data/data-sources`.
 
 ## 2. Define a targets config
 
-The targets config tells BatDetect2 how to turn your annotations into training targets.
+The targets config tells BatDetect2 how to turn your annotations into training
+targets.
 
 It defines two main things:
 
 - what should count as a detection,
 - which classes the model should learn to predict.
 
-In practice, this means the targets config maps the labels in your annotations to the detection and classification outputs used during training.
+In practice, this means the targets config maps the labels in your annotations
+to the detection and classification outputs used during training.
 
 Use `example_data/targets.yaml` as a reference:
 
@@ -107,16 +116,21 @@ classification_targets:
         value: Pipistrellus pipistrellus
 ```
 
-For your own project, update the matching rules and class definitions so they fit your labels.
+For your own project, update the matching rules and class definitions so they
+fit your labels.
 
 In this example:
 
-- `detection_target` says that echolocation calls should be treated as detections,
+- `detection_target` says that echolocation calls should be treated as
+  detections,
 - `classification_targets` define the classes the model should predict,
 
-It is worth taking a bit of time over this file, because your targets config decides what the model is actually being asked to learn.
+It is worth taking a bit of time over this file, because your targets config
+decides what the model is actually being asked to learn.
 
-If you need help with that, see {doc}`../how_to/configure-target-definitions` and {doc}`../reference/targets-config-workflow`.
+If you need help with that, see
+{doc}`../how_to/data/configure-target-definitions` and
+{doc}`../reference/configs/data/targets-config-workflow`.
 
 ## 3. Run a first training command
 
@@ -138,12 +152,16 @@ batdetect2 train \
   --targets example_data/targets.yaml
 ```
 
-This uses the same dataset for training and validation only to keep the example simple.
-For real training runs, you usually want separate training and validation datasets.
+This uses the same dataset for training and validation only to keep the example
+simple.
+For real training runs, you usually want separate training and validation
+datasets.
 
 This uses the built-in default model and training settings.
-If you want to change the model architecture later, see {doc}`../reference/model-config`.
-If you want to change optimiser settings, batch size, epochs, or checkpoint behaviour, see {doc}`../reference/training-config`.
+If you want to change the model architecture later, see
+{doc}`../reference/configs/training/model-config`.
+If you want to change optimiser settings, batch size, epochs, or checkpoint
+behaviour, see {doc}`../reference/configs/training/training-config`.
 
 ## 4. Check the training outputs
 
@@ -170,10 +188,13 @@ outputs/
       val_class_summary.csv
 ```
 
-The checkpoint is the trained model you can use later for inference, evaluation, or sharing with someone else.
+The checkpoint is the trained model you can use later for inference, evaluation,
+or sharing with someone else.
 
-The files in `training_artifacts/` record which datasets and targets were used for the run.
-The `hparams.yaml` file records the full training setup, including the configs used for the model, training, and other parts of the run.
+The files in `training_artifacts/` record which datasets and targets were used
+for the run.
+The `hparams.yaml` file records the full training setup, including the configs
+used for the model, training, and other parts of the run.
 
 The `metrics.csv` file stores one row per validation epoch.
 It includes training losses as well as validation losses and metrics such as:
@@ -190,19 +211,29 @@ You may also see class-specific metrics in extra columns.
 The more detailed metrics are computed from the validation set.
 If you do not provide `--val-dataset`, those validation metrics will not appear.
 
-Other logger backends are also supported, including TensorBoard, MLflow, and DVCLive.
-See {doc}`../reference/logging-config` if you want to change that.
+Other logger backends are also supported, including TensorBoard, MLflow, and
+DVCLive.
+See {doc}`../reference/configs/training/logging-config` if you want to change
+that.
 
 ## Use the trained model
 
-You can now use the trained checkpoint in BatDetect2, or share it with someone else to use in their own runs.
-If you want to load it for inference or evaluation, see {doc}`../how_to/choose-a-model`.
+You can now use the trained checkpoint in BatDetect2, or share it with someone
+else to use in their own runs.
+If you want to load it for inference or evaluation, see
+{doc}`../how_to/inference/choose-a-model`.
 
 ## Common next steps
 
-- Evaluate the trained checkpoint: {doc}`evaluate-on-a-test-set`
-- Fine-tune from a checkpoint: {doc}`../how_to/fine-tune-from-a-checkpoint`
-- Configure targets in more detail: {doc}`../how_to/configure-target-definitions`
-- Configure audio preprocessing: {doc}`../how_to/configure-audio-preprocessing`
-- Configure spectrogram preprocessing: {doc}`../how_to/configure-spectrogram-preprocessing`
-- Check full train options: {doc}`../reference/cli/train`
+- Evaluate the trained checkpoint:
+  {doc}`evaluate-on-a-test-set`
+- Fine-tune from a checkpoint:
+  {doc}`../how_to/training/fine-tune-from-a-checkpoint`
+- Configure targets in more detail:
+  {doc}`../how_to/data/configure-target-definitions`
+- Configure audio preprocessing:
+  {doc}`../how_to/data/configure-audio-preprocessing`
+- Configure spectrogram preprocessing:
+  {doc}`../how_to/data/configure-spectrogram-preprocessing`
+- Check full train options:
+  {doc}`../reference/cli/train`