Align code with polygon-only training

README.md

@@ -97,8 +97,6 @@ dataset_card.json
   and write normalized manifests.
 - `scripts/search_hf_marine_datasets.py`: search Hugging Face datasets and write
   normalized `hf://` manifest references without downloading full repositories.
-- `scripts/import_quality_gated_hf_dataset.py`: import only HF samples that pass
-  explicit quality adapters; unverified samples stay in review/reject manifests.
 - `scripts/prepare_polygon_dataset.py`: build the strict mask/polygon dataset
   for SAMPoly-style training and reject bbox-only samples.
 - `scripts/compose_task_profile.py`: compose Markdown capability cards into a
@@ -122,23 +120,6 @@ The target architecture is a unified, explainable, multi-task model:
 Current legacy green-tide weights can still be loaded for candidate generation,
 but final products should come from the registry-driven multi-task framework.
 
-## Hugging Face Dataset Discovery
-
-HF dataset search and standardization results are summarized in
-[docs/hf_dataset_discovery.md](docs/hf_dataset_discovery.md). The generated local
-manifests are under:
-
-```text
-C:\Users\HUAWEI\Documents\New project 2\hf_marine_feature_dataset
-```
-
-The discovery output is conservative: unknown elements and unpaired HF assets are
-kept for review and are not treated as training negatives.
-
-The current accepted import batch is summarized in
-[docs/quality_gated_import.md](docs/quality_gated_import.md). Accepted samples
-are stored under:
-
-```text
-C:\Users\HUAWEI\Documents\New project 2\marine_feature_dataset_imported
-```
+No bbox-only accepted dataset is retained. New training samples should be
+imported only when they provide mask or polygon annotations suitable for the
+SAMPoly-style head.