slide2vec 4.0.3__tar.gz → 4.1.0__tar.gz

{slide2vec-4.0.3 → slide2vec-4.1.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: slide2vec
-Version: 4.0.3
+Version: 4.1.0
 Summary: Embedding of whole slide images with Foundation Models
 Author-email: Clément Grisi <clement.grisi@radboudumc.nl>
 License-Expression: Apache-2.0

{slide2vec-4.0.3 → slide2vec-4.1.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "slide2vec"
-version = "4.0.3"
+version = "4.1.0"
 description = "Embedding of whole slide images with Foundation Models"
 readme = "README.md"
 requires-python = ">=3.10"
@@ -154,7 +154,7 @@ no_implicit_reexport = true
 max-line-length = 160
 
 [tool.bumpver]
-current_version = "4.0.3"
+current_version = "4.1.0"
 version_pattern = "MAJOR.MINOR.PATCH"
 commit = false       # We do version bumping in CI, not as a commit
 tag = false          # Git tag already exists — we don't auto-tag

{slide2vec-4.0.3 → slide2vec-4.1.0}/slide2vec/__init__.py

@@ -2,7 +2,7 @@ from slide2vec.api import EmbeddedSlide, ExecutionOptions, Model, Pipeline, Prep
 from slide2vec.artifacts import HierarchicalEmbeddingArtifact, SlideEmbeddingArtifact, TileEmbeddingArtifact
 
 
-__version__ = "4.0.3"
+__version__ = "4.1.0"
 
 __all__ = [
     "Model",

{slide2vec-4.0.3 → slide2vec-4.1.0}/slide2vec/distributed/direct_embed_worker.py

@@ -25,6 +25,7 @@ def main(argv=None) -> int:
         _compute_hierarchical_embedding_shard_for_slide,
         _compute_tile_embeddings_for_slide,
         _is_hierarchical_preprocessing,
+        _resolve_hierarchical_geometry,
         deserialize_execution,
         deserialize_preprocessing,
         load_successful_tiled_slides,
@@ -74,9 +75,11 @@ def main(argv=None) -> int:
                 slide, tiling_result = paired_by_sample[sample_id]
                 loaded = model._load_backend()
                 if _is_hierarchical_preprocessing(preprocessing):
+                    geometry = _resolve_hierarchical_geometry(preprocessing, tiling_result)
                     index = _build_hierarchical_index(
                         tiling_result,
                         region_tile_multiple=int(preprocessing.region_tile_multiple),
+                        tile_size_lv0=int(geometry["tile_size_lv0"]),
                     )
                     flat_indices = np.array_split(index.flat_index, world_size)[global_rank]
                     shard_indices, tile_embeddings = _compute_hierarchical_embedding_shard_for_slide(

{slide2vec-4.0.3 → slide2vec-4.1.0}/slide2vec/inference.py

@@ -111,19 +111,23 @@ def _resolve_hierarchical_geometry(preprocessing: PreprocessingConfig, tiling_re
         raise ValueError("Hierarchical preprocessing requires target_region_size_px")
     target_tile_size_px = int(preprocessing.target_tile_size_px)
     target_region_size_px = int(preprocessing.target_region_size_px)
-    effective_region_size_px = int(getattr(tiling_result, "effective_tile_size_px"))
-    tile_size_lv0 = int(getattr(tiling_result, "tile_size_lv0"))
+    target_spacing_um = float(preprocessing.target_spacing_um)
     multiple = int(preprocessing.region_tile_multiple)
     if target_region_size_px % multiple != 0:
         raise ValueError("target_region_size_px must be divisible by region_tile_multiple")
+    effective_spacing_um = float(getattr(tiling_result, "effective_spacing_um"))
+    base_spacing_um = float(getattr(tiling_result, "base_spacing_um"))
+    effective_tile_size_px = int(round(target_tile_size_px * target_spacing_um / effective_spacing_um))
+    effective_region_size_px = effective_tile_size_px * multiple
+    tile_size_lv0 = int(round(target_tile_size_px * target_spacing_um / base_spacing_um))
     return {
         "region_tile_multiple": multiple,
         "tiles_per_region": multiple * multiple,
         "target_tile_size_px": target_tile_size_px,
-        "effective_tile_size_px": effective_region_size_px // multiple,
+        "effective_tile_size_px": effective_tile_size_px,
         "target_region_size_px": target_region_size_px,
         "effective_region_size_px": effective_region_size_px,
-        "tile_size_lv0": tile_size_lv0 // multiple,
+        "tile_size_lv0": tile_size_lv0,
     }
 
 
@@ -131,14 +135,18 @@ def _build_hierarchical_index(
     tiling_result,
     *,
     region_tile_multiple: int,
+    tile_size_lv0: int | None = None,
 ) -> HierarchicalIndex:
     x_values, y_values = coordinate_arrays(tiling_result)
     num_regions = int(len(x_values))
     multiple = int(region_tile_multiple)
     if multiple < 2:
         raise ValueError("region_tile_multiple must be at least 2")
-    tile_size_lv0 = int(getattr(tiling_result, "tile_size_lv0"))
-    subtile_size_lv0 = tile_size_lv0 // multiple
+    subtile_size_lv0 = (
+        int(tile_size_lv0)
+        if tile_size_lv0 is not None
+        else int(getattr(tiling_result, "tile_size_lv0")) // multiple
+    )
     tiles_per_region = multiple * multiple
     if num_regions == 0:
         empty = np.empty(0, dtype=np.int64)
@@ -1275,6 +1283,7 @@ def _compute_hierarchical_embeddings_for_slide(
     index = _build_hierarchical_index(
         tiling_result,
         region_tile_multiple=int(geometry["region_tile_multiple"]),
+        tile_size_lv0=int(geometry["tile_size_lv0"]),
     )
     resolved_indices = index.flat_index
     if flat_indices is not None:
@@ -1370,6 +1379,7 @@ def _compute_hierarchical_embedding_shard_for_slide(
     index = _build_hierarchical_index(
         tiling_result,
         region_tile_multiple=int(geometry["region_tile_multiple"]),
+        tile_size_lv0=int(geometry["tile_size_lv0"]),
     )
     resolved_indices = np.asarray(flat_indices, dtype=np.int64)
     collate_fn = OnTheFlyHierarchicalBatchCollator(
@@ -2261,11 +2271,12 @@ def _should_persist_tile_embeddings(model, execution: ExecutionOptions) -> bool:
 
 
 def _resolved_process_list_output_variant(model) -> str | None:
+    requested_output_variant = getattr(model, "_output_variant", None)
     if not hasattr(model, "name") or model.name not in encoder_registry:
-        return model._output_variant if hasattr(model, "_output_variant") else None
+        return requested_output_variant
     resolved = resolve_encoder_output(
         model.name,
-        requested_output_variant=model._output_variant,
+        requested_output_variant=requested_output_variant,
     )
     return str(resolved["output_variant"])
 

{slide2vec-4.0.3 → slide2vec-4.1.0}/slide2vec.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: slide2vec
-Version: 4.0.3
+Version: 4.1.0
 Summary: Embedding of whole slide images with Foundation Models
 Author-email: Clément Grisi <clement.grisi@radboudumc.nl>
 License-Expression: Apache-2.0

{slide2vec-4.0.3 → slide2vec-4.1.0}/tests/test_regression_inference.py

@@ -176,6 +176,8 @@ def test_collect_distributed_pipeline_artifacts_runs_stage_collects_and_updates(
         persist_tile_embeddings,
         persist_hierarchical_embeddings,
         include_slide_embeddings,
+        encoder_name,
+        output_variant,
         tile_artifacts,
         hierarchical_artifacts,
         slide_artifacts,
@@ -186,6 +188,8 @@ def test_collect_distributed_pipeline_artifacts_runs_stage_collects_and_updates(
             "persist_tile_embeddings": persist_tile_embeddings,
             "persist_hierarchical_embeddings": persist_hierarchical_embeddings,
             "include_slide_embeddings": include_slide_embeddings,
+            "encoder_name": encoder_name,
+            "output_variant": output_variant,
             "tile_artifacts": tile_artifacts,
             "hierarchical_artifacts": hierarchical_artifacts,
             "slide_artifacts": slide_artifacts,
@@ -220,6 +224,8 @@ def test_collect_distributed_pipeline_artifacts_runs_stage_collects_and_updates(
     assert captured["update"]["persist_tile_embeddings"] is True
     assert captured["update"]["persist_hierarchical_embeddings"] is False
     assert captured["update"]["include_slide_embeddings"] is True
+    assert captured["update"]["encoder_name"] == "prism"
+    assert captured["update"]["output_variant"] == "default"
     assert captured["update"]["tile_artifacts"] == ["tile-artifact"]
     assert captured["update"]["hierarchical_artifacts"] == []
     assert captured["update"]["slide_artifacts"] == ["slide-artifact"]
@@ -2802,6 +2808,49 @@ def test_build_hierarchical_index_is_region_major_and_row_major_within_region():
     )
 
 
+def test_resolve_hierarchical_geometry_scales_tile_first_under_spacing_mismatch():
+    import slide2vec.inference as inference
+
+    preprocessing = PreprocessingConfig(
+        target_spacing_um=0.5,
+        target_tile_size_px=224,
+        target_region_size_px=1792,
+        region_tile_multiple=8,
+    )
+    tiling_result = SimpleNamespace(
+        effective_tile_size_px=3319,
+        effective_spacing_um=0.27,
+        tile_size_lv0=3319,
+        base_spacing_um=0.27,
+    )
+
+    geometry = inference._resolve_hierarchical_geometry(preprocessing, tiling_result)
+
+    assert geometry["effective_tile_size_px"] == 415
+    assert geometry["effective_region_size_px"] == 3320
+    assert geometry["tile_size_lv0"] == 415
+    assert geometry["tiles_per_region"] == 64
+
+
+def test_build_hierarchical_index_uses_tile_first_level0_offsets_under_spacing_mismatch():
+    import slide2vec.inference as inference
+
+    tiling_result = SimpleNamespace(
+        x=np.array([100], dtype=np.int64),
+        y=np.array([200], dtype=np.int64),
+    )
+
+    index = inference._build_hierarchical_index(
+        tiling_result,
+        region_tile_multiple=8,
+        tile_size_lv0=415,
+    )
+
+    assert index.tiles_per_region == 64
+    np.testing.assert_array_equal(index.subtile_x[:8], np.array([100, 515, 930, 1345, 1760, 2175, 2590, 3005], dtype=np.int64))
+    np.testing.assert_array_equal(index.subtile_y[::8], np.array([200, 615, 1030, 1445, 1860, 2275, 2690, 3105], dtype=np.int64))
+
+
 def test_merge_hierarchical_embedding_shards_restores_original_region_shape():
     import slide2vec.inference as inference
 
@@ -2926,6 +2975,7 @@ def test_compute_hierarchical_embeddings_for_slide_encodes_flat_tile_batches_and
         tile_size_lv0=448,
         target_spacing_um=0.5,
         effective_spacing_um=0.5,
+        base_spacing_um=0.5,
         read_level=0,
     )