example to reproduce issue with hitting limit of pool

lightly_studio/Makefile

@@ -40,6 +40,11 @@ start-e2e: build
 	@echo "Starting server for e2e tests..."
 	uv run e2e-tests/index_dataset_for_end2end_ui_tests.py
 
+.PHONY: start-coco-10k
+start-coco-10k: build
+	@echo "Starting server for e2e tests with COCO 10k dataset..."
+	uv run e2e-tests/index_coco_10k.py
+
 .PHONY: start-e2e-with-captions
 start-e2e-with-captions: build
 	@echo "Starting server for e2e tests with captions..."

lightly_studio/e2e-tests/index_coco_10k.py

@@ -0,0 +1,25 @@
+"""End-to-end demonstration of the lightly_studio dataset loading and UI.
+
+This module provides a simple example of how to load a COCO instance
+segmentation dataset and launch the UI application for exploration and
+visualization.
+"""
+
+from lightly_studio import AnnotationType, Dataset, db_manager, start_gui
+
+# Clean up an existing database
+db_manager.connect(cleanup_existing=True)
+
+# Create a Dataset instance
+dataset = Dataset.create()
+
+# We point to the annotations json file and the input images folder.
+# Defined dataset is processed here to be available for the UI application.
+dataset.add_samples_from_coco(
+    annotations_json="datasets/coco-10k/annotations/instances_train2017.json",
+    images_path="datasets/coco-10k/images",
+    annotation_type=AnnotationType.INSTANCE_SEGMENTATION,
+)
+
+# We start the UI application on port 8001
+start_gui()