slide2vec 3.2.1__tar.gz → 4.0.2__tar.gz

{slide2vec-3.2.1/slide2vec.egg-info → slide2vec-4.0.2}/PKG-INFO

@@ -1,16 +1,11 @@
 Metadata-Version: 2.4
 Name: slide2vec
-Version: 3.2.1
+Version: 4.0.2
 Summary: Embedding of whole slide images with Foundation Models
-Home-page: https://github.com/clemsgrs/slide2vec
-Author: Clément Grisi
-Author-email: clement.grisi@radboudumc.nl
+Author-email: Clément Grisi <clement.grisi@radboudumc.nl>
+License-Expression: Apache-2.0
+Project-URL: Homepage, https://github.com/clemsgrs/slide2vec
 Project-URL: Bug Tracker, https://github.com/clemsgrs/slide2vec/issues
-Platform: unix
-Platform: linux
-Platform: osx
-Platform: cygwin
-Platform: win32
 Classifier: Programming Language :: Python :: 3
 Classifier: Programming Language :: Python :: 3 :: Only
 Classifier: Programming Language :: Python :: 3.10
@@ -20,42 +15,86 @@ Classifier: Programming Language :: Python :: 3.13
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: hs2p<3,>=2.5.1
+Requires-Dist: hs2p[asap,cucim,openslide,vips]>=3.1.4
 Requires-Dist: omegaconf
-Requires-Dist: h5py
 Requires-Dist: matplotlib
 Requires-Dist: numpy<2
 Requires-Dist: pandas
 Requires-Dist: pillow
-Requires-Dist: PyTurboJPEG
 Requires-Dist: rich
-Requires-Dist: tqdm
 Requires-Dist: torch
 Requires-Dist: torchvision
 Requires-Dist: transformers
 Requires-Dist: wandb
-Requires-Dist: wholeslidedata<0.0.16
 Requires-Dist: einops
 Requires-Dist: timm
-Provides-Extra: cucim
-Requires-Dist: hs2p[cucim]<3,>=2.5.1; extra == "cucim"
-Provides-Extra: models
-Requires-Dist: huggingface-hub; extra == "models"
-Requires-Dist: sacremoses; extra == "models"
-Requires-Dist: einops-exts; extra == "models"
-Requires-Dist: xformers; extra == "models"
+Requires-Dist: huggingface_hub
+Provides-Extra: hoptimus
+Requires-Dist: torch>=2.0; extra == "hoptimus"
+Requires-Dist: torchvision>=0.15.0; extra == "hoptimus"
+Requires-Dist: xformers>=0.0.18; extra == "hoptimus"
+Provides-Extra: virchow
+Requires-Dist: timm>=0.9.11; extra == "virchow"
+Requires-Dist: torch>=2.0; extra == "virchow"
+Provides-Extra: uni
+Requires-Dist: torch>=2.0; extra == "uni"
+Requires-Dist: timm>=0.9.8; extra == "uni"
+Requires-Dist: xformers>=0.0.18; extra == "uni"
+Provides-Extra: prism
+Requires-Dist: transformers~=4.53.0; extra == "prism"
+Requires-Dist: torch<2.8,>=2.3; extra == "prism"
+Requires-Dist: einops==0.8.0; extra == "prism"
+Requires-Dist: environs==11.0.0; extra == "prism"
+Requires-Dist: sacremoses==0.1.1; extra == "prism"
+Requires-Dist: xformers==0.0.31; extra == "prism"
+Provides-Extra: hibou
+Requires-Dist: scipy~=1.8.1; extra == "hibou"
+Requires-Dist: scikit-image~=0.19.3; extra == "hibou"
+Provides-Extra: titan
+Requires-Dist: torch==2.0.1; extra == "titan"
+Requires-Dist: timm==1.0.3; extra == "titan"
+Requires-Dist: einops==0.6.1; extra == "titan"
+Requires-Dist: einops-exts==0.0.4; extra == "titan"
+Requires-Dist: transformers==4.46.0; extra == "titan"
+Provides-Extra: fm
+Requires-Dist: omegaconf>=2.3.0; extra == "fm"
+Requires-Dist: matplotlib; extra == "fm"
+Requires-Dist: numpy<2; extra == "fm"
+Requires-Dist: pandas; extra == "fm"
+Requires-Dist: pillow; extra == "fm"
+Requires-Dist: rich; extra == "fm"
+Requires-Dist: hs2p[asap,cucim,openslide,vips]>=3.1.4; extra == "fm"
+Requires-Dist: wandb; extra == "fm"
+Requires-Dist: torch<2.8,>=2.3; extra == "fm"
+Requires-Dist: torchvision>=0.18.0; extra == "fm"
+Requires-Dist: einops>=0.8.0; extra == "fm"
+Requires-Dist: timm>=1.0.3; extra == "fm"
+Requires-Dist: huggingface_hub<1.0,>=0.30.0; extra == "fm"
+Requires-Dist: environs; extra == "fm"
+Requires-Dist: einops-exts>=0.0.4; extra == "fm"
+Requires-Dist: transformers>=4.53; extra == "fm"
+Requires-Dist: sacremoses; extra == "fm"
+Requires-Dist: xformers>=0.0.31; extra == "fm"
+Requires-Dist: scipy>=1.8.1; extra == "fm"
+Requires-Dist: scikit-image>=0.19.3; extra == "fm"
+Requires-Dist: torchmetrics>=0.10.3; extra == "fm"
+Requires-Dist: fvcore; extra == "fm"
+Requires-Dist: iopath; extra == "fm"
+Requires-Dist: webdataset; extra == "fm"
+Requires-Dist: scikit-survival; extra == "fm"
+Requires-Dist: scikit-learn; extra == "fm"
+Requires-Dist: fairscale; extra == "fm"
+Requires-Dist: packaging==23.2; extra == "fm"
+Requires-Dist: ninja==1.11.1.1; extra == "fm"
+Requires-Dist: psutil<6; extra == "fm"
 Provides-Extra: testing
 Requires-Dist: pytest>=6.0; extra == "testing"
 Requires-Dist: pytest-cov>=2.0; extra == "testing"
 Requires-Dist: mypy>=0.910; extra == "testing"
 Requires-Dist: flake8>=3.9; extra == "testing"
+Requires-Dist: flake8-pyproject>=1.2.3; extra == "testing"
 Requires-Dist: tox>=3.24; extra == "testing"
-Dynamic: author-email
-Dynamic: description
-Dynamic: description-content-type
-Dynamic: home-page
 Dynamic: license-file
-Dynamic: project-url
 
 # slide2vec
 
@@ -67,16 +106,19 @@ Dynamic: project-url
 
 ```shell
 pip install slide2vec
+pip install "slide2vec[fm]"
 ```
 
-Install the full model runtime only when you need embedding/model execution:
+`slide2vec` keeps the base install focused on the core package surface. Use `slide2vec[fm]` when you want the PyPI-hosted FM dependencies.
+
+Some model backends still rely on upstream Git repositories that PyPI will not accept as package metadata. Install those separately when needed:
 
 ```shell
-pip install "slide2vec[models]"
+pip install git+https://github.com/lilab-stanford/MUSK.git
+pip install git+https://github.com/Mahmoodlab/CONCH.git
+pip install git+https://github.com/prov-gigapath/prov-gigapath.git
 ```
 
-`slide2vec` now keeps the base install focused on the core package surface and moves the heavier model stack into the optional `models` extra.
-
 ## Python API
 
 ```python
@@ -89,7 +131,8 @@ model = Model.from_preset("virchow2")
 embedded = model.embed_slide("/path/to/slide.svs")
 
 tile_embeddings = embedded.tile_embeddings
-coordinates = embedded.coordinates
+x = embedded.x
+y = embedded.y
 ```
 
 Use `Pipeline(...)` for manifest-driven batch processing when you want artifacts written to disk instead of only in-memory outputs:
@@ -111,6 +154,23 @@ result = pipeline.run(manifest_path="/path/to/slides.csv")
 
 By default, `ExecutionOptions()` uses all available GPUs. Set `ExecutionOptions(num_gpus=4)` when you want to cap the sharding explicitly.
 
+### Hierarchical Feature Extraction
+
+Tile embeddings can be spatially grouped into regions for downstream models that consume region-level structure. Enable it by setting `region_tile_multiple` on `PreprocessingConfig`:
+
+```python
+preprocessing = PreprocessingConfig(
+    target_spacing_um=0.5,
+    target_tile_size_px=224,
+    region_tile_multiple=6,  # 6x6 tiles per region
+)
+embedded = model.embed_slide("/path/to/slide.svs", preprocessing=preprocessing)
+```
+
+Hierarchical outputs have shape `(num_regions, tiles_per_region, feature_dim)` and are written to `hierarchical_embeddings/` when persisted.
+
+See [`docs/python-api.md`](docs/python-api.md) for details.
+
 ### Input Manifest
 
 Manifest-driven runs use the schema below. `mask_path` and `spacing_at_level_0` are optional.
@@ -131,6 +191,8 @@ The package writes explicit artifact directories:
 
 - `tile_embeddings/<sample_id>.pt` or `.npz`
 - `tile_embeddings/<sample_id>.meta.json`
+- `hierarchical_embeddings/<sample_id>.pt` or `.npz` (when `region_tile_multiple` is set)
+- `hierarchical_embeddings/<sample_id>.meta.json`
 - `slide_embeddings/<sample_id>.pt` or `.npz`
 - `slide_embeddings/<sample_id>.meta.json`
 - optional `slide_latents/<sample_id>.pt` or `.npz`
@@ -139,7 +201,7 @@ The package writes explicit artifact directories:
 
 ### Supported Models
 
-`slide2vec` currently ships preset configs for 20 tile-level models and 3 slide-level models.  
+`slide2vec` currently ships preset configs for 17 tile-level models and 3 slide-level models.  
 For the full catalog and preset names, see [`docs/models.md`](docs/models.md).
 
 ## CLI

{slide2vec-3.2.1 → slide2vec-4.0.2}/README.md

@@ -8,16 +8,19 @@
 
 ```shell
 pip install slide2vec
+pip install "slide2vec[fm]"
 ```
 
-Install the full model runtime only when you need embedding/model execution:
+`slide2vec` keeps the base install focused on the core package surface. Use `slide2vec[fm]` when you want the PyPI-hosted FM dependencies.
+
+Some model backends still rely on upstream Git repositories that PyPI will not accept as package metadata. Install those separately when needed:
 
 ```shell
-pip install "slide2vec[models]"
+pip install git+https://github.com/lilab-stanford/MUSK.git
+pip install git+https://github.com/Mahmoodlab/CONCH.git
+pip install git+https://github.com/prov-gigapath/prov-gigapath.git
 ```
 
-`slide2vec` now keeps the base install focused on the core package surface and moves the heavier model stack into the optional `models` extra.
-
 ## Python API
 
 ```python
@@ -30,7 +33,8 @@ model = Model.from_preset("virchow2")
 embedded = model.embed_slide("/path/to/slide.svs")
 
 tile_embeddings = embedded.tile_embeddings
-coordinates = embedded.coordinates
+x = embedded.x
+y = embedded.y
 ```
 
 Use `Pipeline(...)` for manifest-driven batch processing when you want artifacts written to disk instead of only in-memory outputs:
@@ -52,6 +56,23 @@ result = pipeline.run(manifest_path="/path/to/slides.csv")
 
 By default, `ExecutionOptions()` uses all available GPUs. Set `ExecutionOptions(num_gpus=4)` when you want to cap the sharding explicitly.
 
+### Hierarchical Feature Extraction
+
+Tile embeddings can be spatially grouped into regions for downstream models that consume region-level structure. Enable it by setting `region_tile_multiple` on `PreprocessingConfig`:
+
+```python
+preprocessing = PreprocessingConfig(
+    target_spacing_um=0.5,
+    target_tile_size_px=224,
+    region_tile_multiple=6,  # 6x6 tiles per region
+)
+embedded = model.embed_slide("/path/to/slide.svs", preprocessing=preprocessing)
+```
+
+Hierarchical outputs have shape `(num_regions, tiles_per_region, feature_dim)` and are written to `hierarchical_embeddings/` when persisted.
+
+See [`docs/python-api.md`](docs/python-api.md) for details.
+
 ### Input Manifest
 
 Manifest-driven runs use the schema below. `mask_path` and `spacing_at_level_0` are optional.
@@ -72,6 +93,8 @@ The package writes explicit artifact directories:
 
 - `tile_embeddings/<sample_id>.pt` or `.npz`
 - `tile_embeddings/<sample_id>.meta.json`
+- `hierarchical_embeddings/<sample_id>.pt` or `.npz` (when `region_tile_multiple` is set)
+- `hierarchical_embeddings/<sample_id>.meta.json`
 - `slide_embeddings/<sample_id>.pt` or `.npz`
 - `slide_embeddings/<sample_id>.meta.json`
 - optional `slide_latents/<sample_id>.pt` or `.npz`
@@ -80,7 +103,7 @@ The package writes explicit artifact directories:
 
 ### Supported Models
 
-`slide2vec` currently ships preset configs for 20 tile-level models and 3 slide-level models.  
+`slide2vec` currently ships preset configs for 17 tile-level models and 3 slide-level models.  
 For the full catalog and preset names, see [`docs/models.md`](docs/models.md).
 
 ## CLI

slide2vec-4.0.2/pyproject.toml

@@ -0,0 +1,174 @@
+[build-system]
+requires = ["setuptools>=61.0", "wheel"]
+build-backend = "setuptools.build_meta"
+
+[project]
+name = "slide2vec"
+version = "4.0.2"
+description = "Embedding of whole slide images with Foundation Models"
+readme = "README.md"
+requires-python = ">=3.10"
+license = "Apache-2.0"
+authors = [
+    { name = "Clément Grisi", email = "clement.grisi@radboudumc.nl" },
+]
+classifiers = [
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3 :: Only",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Programming Language :: Python :: 3.13",
+]
+dependencies = [
+    "hs2p[asap,cucim,openslide,vips]>=3.1.4",
+    "omegaconf",
+    "matplotlib",
+    "numpy<2",
+    "pandas",
+    "pillow",
+    "rich",
+    "torch",
+    "torchvision",
+    "transformers",
+    "wandb",
+    "einops",
+    "timm",
+    "huggingface_hub",
+]
+
+[project.urls]
+Homepage = "https://github.com/clemsgrs/slide2vec"
+"Bug Tracker" = "https://github.com/clemsgrs/slide2vec/issues"
+
+[project.scripts]
+slide2vec = "slide2vec.cli:main"
+
+[project.optional-dependencies]
+hoptimus = [
+    "torch>=2.0",
+    "torchvision>=0.15.0",
+    "xformers>=0.0.18",
+]
+virchow = [
+    "timm>=0.9.11",
+    "torch>=2.0",
+]
+uni = [
+    "torch>=2.0",
+    "timm>=0.9.8",
+    "xformers>=0.0.18",
+]
+prism = [
+    "transformers~=4.53.0",
+    "torch>=2.3,<2.8",
+    "einops==0.8.0",
+    "environs==11.0.0",
+    "sacremoses==0.1.1",
+    "xformers==0.0.31",
+]
+hibou = [
+    "scipy~=1.8.1",
+    "scikit-image~=0.19.3",
+]
+titan = [
+    "torch==2.0.1",
+    "timm==1.0.3",
+    "einops==0.6.1",
+    "einops-exts==0.0.4",
+    "transformers==4.46.0",
+]
+fm = [
+    "omegaconf>=2.3.0",
+    "matplotlib",
+    "numpy<2",
+    "pandas",
+    "pillow",
+    "rich",
+    "hs2p[asap,cucim,openslide,vips]>=3.1.4",
+    "wandb",
+    "torch>=2.3,<2.8",
+    "torchvision>=0.18.0",
+    "einops>=0.8.0",
+    "timm>=1.0.3",
+    "huggingface_hub>=0.30.0,<1.0",
+    "environs",
+    "einops-exts>=0.0.4",
+    "transformers>=4.53",
+    "sacremoses",
+    "xformers>=0.0.31",
+    "scipy>=1.8.1",
+    "scikit-image>=0.19.3",
+    "torchmetrics>=0.10.3",
+    "fvcore",
+    "iopath",
+    "webdataset",
+    "scikit-survival",
+    "scikit-learn",
+    "fairscale",
+    "packaging==23.2",
+    "ninja==1.11.1.1",
+    "psutil<6",
+]
+testing = [
+    "pytest>=6.0",
+    "pytest-cov>=2.0",
+    "mypy>=0.910",
+    "flake8>=3.9",
+    "flake8-pyproject>=1.2.3",
+    "tox>=3.24",
+]
+
+[tool.setuptools]
+include-package-data = true
+zip-safe = false
+
+[tool.setuptools.packages.find]
+include = ["slide2vec*"]
+
+[tool.setuptools.package-data]
+slide2vec = ["py.typed"]
+"slide2vec.configs" = ["*.yaml", "models/*.yaml", "preprocessing/*.yaml"]
+
+[tool.pytest.ini_options]
+addopts = "--cov=slide2vec"
+testpaths = [
+    "tests",
+]
+
+[tool.mypy]
+mypy_path = "."
+check_untyped_defs = true
+disallow_any_generics = true
+ignore_missing_imports = true
+no_implicit_optional = true
+show_error_codes = true
+strict_equality = true
+warn_redundant_casts = true
+warn_return_any = true
+warn_unreachable = true
+warn_unused_configs = true
+no_implicit_reexport = true
+
+[tool.flake8]
+max-line-length = 160
+
+[tool.bumpver]
+current_version = "4.0.2"
+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
+push = false         # Don't push anything in CI
+files = [
+    "pyproject.toml",
+    "slide2vec/__init__.py"
+]
+
+[tool.bumpver.file_patterns]
+"pyproject.toml" = [
+    '^current_version = "{version}"$',
+    '^version = "{version}"$',
+]
+"slide2vec/__init__.py" = [
+    '^__version__ = "{version}"$',
+]

slide2vec-4.0.2/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

{slide2vec-3.2.1 → slide2vec-4.0.2}/slide2vec/__init__.py

@@ -1,8 +1,8 @@
 from slide2vec.api import EmbeddedSlide, ExecutionOptions, Model, Pipeline, PreprocessingConfig, RunResult
-from slide2vec.artifacts import SlideEmbeddingArtifact, TileEmbeddingArtifact
+from slide2vec.artifacts import HierarchicalEmbeddingArtifact, SlideEmbeddingArtifact, TileEmbeddingArtifact
 
 
-__version__ = "3.2.1"
+__version__ = "4.0.2"
 
 __all__ = [
     "Model",
@@ -12,6 +12,7 @@ __all__ = [
     "RunResult",
     "EmbeddedSlide",
     "SlideEmbeddingArtifact",
+    "HierarchicalEmbeddingArtifact",
     "TileEmbeddingArtifact",
     "__version__",
 ]