diff --git a/.github/workflows/build.yaml b/.github/workflows/build.yaml
index d40fb9bb..f36bf6d1 100644
--- a/.github/workflows/build.yaml
+++ b/.github/workflows/build.yaml
@@ -174,7 +174,7 @@ jobs:
path: bioimageio_cache
key: ${{matrix.run-expensive-tests && needs.populate-cache.outputs.cache-key || needs.populate-cache.outputs.cache-key-light}}
- name: pytest
- run: pytest --disable-pytest-warnings
+ run: pytest --cov bioimageio --cov-report xml --cov-append --capture no --disable-pytest-warnings
env:
BIOIMAGEIO_CACHE_PATH: bioimageio_cache
RUN_EXPENSIVE_TESTS: ${{ matrix.run-expensive-tests && 'true' || 'false' }}
diff --git a/bioimageio/core/__init__.py b/bioimageio/core/__init__.py
index c7554372..d37be4d4 100644
--- a/bioimageio/core/__init__.py
+++ b/bioimageio/core/__init__.py
@@ -3,6 +3,7 @@
"""
from bioimageio.spec import (
+ ValidationSummary,
build_description,
dump_description,
load_dataset_description,
@@ -112,4 +113,5 @@
"test_model",
"test_resource",
"validate_format",
+ "ValidationSummary",
]
diff --git a/bioimageio/core/backends/keras_backend.py b/bioimageio/core/backends/keras_backend.py
index 1c10da7d..7b661f23 100644
--- a/bioimageio/core/backends/keras_backend.py
+++ b/bioimageio/core/backends/keras_backend.py
@@ -4,17 +4,18 @@
from loguru import logger
from numpy.typing import NDArray
-from bioimageio.spec._internal.io import download
-from bioimageio.spec._internal.type_guards import is_list, is_tuple
from bioimageio.spec.model import v0_4, v0_5
from bioimageio.spec.model.v0_5 import Version
+from bioimageio.spec.utils import download
from .._settings import settings
from ..digest_spec import get_axes_infos
+from ..utils._type_guards import is_list, is_tuple
from ._model_adapter import ModelAdapter
os.environ["KERAS_BACKEND"] = settings.keras_backend
+
# by default, we use the keras integrated with tensorflow
# TODO: check if we should prefer keras
try:
diff --git a/bioimageio/core/backends/onnx_backend.py b/bioimageio/core/backends/onnx_backend.py
index d5b89152..c81dc7d5 100644
--- a/bioimageio/core/backends/onnx_backend.py
+++ b/bioimageio/core/backends/onnx_backend.py
@@ -5,11 +5,11 @@
import onnxruntime as rt # pyright: ignore[reportMissingTypeStubs]
from numpy.typing import NDArray
-from bioimageio.spec._internal.type_guards import is_list, is_tuple
from bioimageio.spec.model import v0_4, v0_5
from bioimageio.spec.utils import download
from ..model_adapters import ModelAdapter
+from ..utils._type_guards import is_list, is_tuple
class ONNXModelAdapter(ModelAdapter):
diff --git a/bioimageio/core/backends/pytorch_backend.py b/bioimageio/core/backends/pytorch_backend.py
index af1ea85d..c0c11717 100644
--- a/bioimageio/core/backends/pytorch_backend.py
+++ b/bioimageio/core/backends/pytorch_backend.py
@@ -11,12 +11,13 @@
from torch import nn
from typing_extensions import assert_never
-from bioimageio.spec._internal.type_guards import is_list, is_ndarray, is_tuple
+from bioimageio.spec._internal.version_type import Version
from bioimageio.spec.common import ZipPath
from bioimageio.spec.model import AnyModelDescr, v0_4, v0_5
from bioimageio.spec.utils import download
from ..digest_spec import import_callable
+from ..utils._type_guards import is_list, is_ndarray, is_tuple
from ._model_adapter import ModelAdapter
@@ -143,7 +144,10 @@ def load_torch_state_dict(
model = model.to(devices[0])
with path.open("rb") as f:
assert not isinstance(f, TextIOWrapper)
- state = torch.load(f, map_location=devices[0], weights_only=True)
+ if Version(str(torch.__version__)) < Version("1.13"):
+ state = torch.load(f, map_location=devices[0])
+ else:
+ state = torch.load(f, map_location=devices[0], weights_only=True)
incompatible = model.load_state_dict(state)
if (
diff --git a/bioimageio/core/backends/torchscript_backend.py b/bioimageio/core/backends/torchscript_backend.py
index ce3ba131..a9801401 100644
--- a/bioimageio/core/backends/torchscript_backend.py
+++ b/bioimageio/core/backends/torchscript_backend.py
@@ -6,11 +6,11 @@
import torch
from numpy.typing import NDArray
-from bioimageio.spec._internal.type_guards import is_list, is_tuple
from bioimageio.spec.model import v0_4, v0_5
from bioimageio.spec.utils import download
from ..model_adapters import ModelAdapter
+from ..utils._type_guards import is_list, is_tuple
class TorchscriptModelAdapter(ModelAdapter):
diff --git a/bioimageio/core/io.py b/bioimageio/core/io.py
index dc5b70db..5c643b15 100644
--- a/bioimageio/core/io.py
+++ b/bioimageio/core/io.py
@@ -1,7 +1,6 @@
import collections.abc
import warnings
import zipfile
-from io import TextIOWrapper
from pathlib import Path, PurePosixPath
from shutil import copyfileobj
from typing import (
@@ -15,15 +14,16 @@
)
import h5py # pyright: ignore[reportMissingTypeStubs]
-import numpy as np
from imageio.v3 import imread, imwrite # type: ignore
from loguru import logger
from numpy.typing import NDArray
from pydantic import BaseModel, ConfigDict, TypeAdapter
from typing_extensions import assert_never
-from bioimageio.spec._internal.io import interprete_file_source
+from bioimageio.spec._internal.io import get_reader, interprete_file_source
+from bioimageio.spec._internal.type_guards import is_ndarray
from bioimageio.spec.common import (
+ FileSource,
HttpUrl,
PermissiveFileSource,
RelativeFilePath,
@@ -65,51 +65,51 @@ def load_image(
else:
src = parsed_source
- # FIXME: why is pyright complaining about giving the union to _split_dataset_path?
if isinstance(src, Path):
- file_source, subpath = _split_dataset_path(src)
+ file_source, suffix, subpath = _split_dataset_path(src)
elif isinstance(src, HttpUrl):
- file_source, subpath = _split_dataset_path(src)
+ file_source, suffix, subpath = _split_dataset_path(src)
elif isinstance(src, ZipPath):
- file_source, subpath = _split_dataset_path(src)
+ file_source, suffix, subpath = _split_dataset_path(src)
else:
assert_never(src)
- path = download(file_source).path
-
- if path.suffix == ".npy":
+ if suffix == ".npy":
if subpath is not None:
- raise ValueError(f"Unexpected subpath {subpath} for .npy path {path}")
- return load_array(path)
- elif path.suffix in SUFFIXES_WITH_DATAPATH:
+ logger.warning(
+ "Unexpected subpath {} for .npy source {}", subpath, file_source
+ )
+
+ image = load_array(file_source)
+ elif suffix in SUFFIXES_WITH_DATAPATH:
if subpath is None:
dataset_path = DEFAULT_H5_DATASET_PATH
else:
dataset_path = str(subpath)
- with h5py.File(path, "r") as f:
+ reader = download(file_source)
+
+ with h5py.File(reader, "r") as f:
h5_dataset = f.get( # pyright: ignore[reportUnknownVariableType]
dataset_path
)
if not isinstance(h5_dataset, h5py.Dataset):
raise ValueError(
- f"{path} is not of type {h5py.Dataset}, but has type "
+ f"{file_source} did not load as {h5py.Dataset}, but has type "
+ str(
type(h5_dataset) # pyright: ignore[reportUnknownArgumentType]
)
)
image: NDArray[Any]
image = h5_dataset[:] # pyright: ignore[reportUnknownVariableType]
- assert isinstance(image, np.ndarray), type(
- image # pyright: ignore[reportUnknownArgumentType]
- )
- return image # pyright: ignore[reportUnknownVariableType]
- elif isinstance(path, ZipPath):
- return imread(
- path.read_bytes(), extension=path.suffix
- ) # pyright: ignore[reportUnknownVariableType]
else:
- return imread(path) # pyright: ignore[reportUnknownVariableType]
+ reader = download(file_source)
+ image = imread( # pyright: ignore[reportUnknownVariableType]
+ reader.read(), extension=suffix
+ )
+
+ assert is_ndarray(image)
+ return image
def load_tensor(
@@ -123,19 +123,21 @@ def load_tensor(
_SourceT = TypeVar("_SourceT", Path, HttpUrl, ZipPath)
+Suffix = str
+
def _split_dataset_path(
source: _SourceT,
-) -> Tuple[_SourceT, Optional[PurePosixPath]]:
+) -> Tuple[_SourceT, Suffix, Optional[PurePosixPath]]:
"""Split off subpath (e.g. internal h5 dataset path)
from a file path following a file extension.
Examples:
>>> _split_dataset_path(Path("my_file.h5/dataset"))
- (...Path('my_file.h5'), PurePosixPath('dataset'))
+ (...Path('my_file.h5'), '.h5', PurePosixPath('dataset'))
>>> _split_dataset_path(Path("my_plain_file"))
- (...Path('my_plain_file'), None)
+ (...Path('my_plain_file'), '', None)
"""
if isinstance(source, RelativeFilePath):
@@ -148,42 +150,47 @@ def _split_dataset_path(
def separate_pure_path(path: PurePosixPath):
for p in path.parents:
if p.suffix in SUFFIXES_WITH_DATAPATH:
- return p, PurePosixPath(path.relative_to(p))
+ return p, p.suffix, PurePosixPath(path.relative_to(p))
- return path, None
+ return path, path.suffix, None
if isinstance(src, HttpUrl):
- file_path, data_path = separate_pure_path(PurePosixPath(src.path or ""))
+ file_path, suffix, data_path = separate_pure_path(PurePosixPath(src.path or ""))
if data_path is None:
- return src, None
+ return src, suffix, None
return (
HttpUrl(str(file_path).replace(f"/{data_path}", "")),
+ suffix,
data_path,
)
if isinstance(src, ZipPath):
- file_path, data_path = separate_pure_path(PurePosixPath(str(src)))
+ file_path, suffix, data_path = separate_pure_path(PurePosixPath(str(src)))
if data_path is None:
- return src, None
+ return src, suffix, None
return (
ZipPath(str(file_path).replace(f"/{data_path}", "")),
+ suffix,
data_path,
)
- file_path, data_path = separate_pure_path(PurePosixPath(src))
- return Path(file_path), data_path
+ file_path, suffix, data_path = separate_pure_path(PurePosixPath(src))
+ return Path(file_path), suffix, data_path
def save_tensor(path: Union[Path, str], tensor: Tensor) -> None:
# TODO: save axis meta data
- data: NDArray[Any] = tensor.data.to_numpy()
- file_path, subpath = _split_dataset_path(Path(path))
- if not file_path.suffix:
+ data: NDArray[Any] = ( # pyright: ignore[reportUnknownVariableType]
+ tensor.data.to_numpy()
+ )
+ assert is_ndarray(data)
+ file_path, suffix, subpath = _split_dataset_path(Path(path))
+ if not suffix:
raise ValueError(f"No suffix (needed to decide file format) found in {path}")
file_path.parent.mkdir(exist_ok=True, parents=True)
@@ -191,7 +198,7 @@ def save_tensor(path: Union[Path, str], tensor: Tensor) -> None:
if subpath is not None:
raise ValueError(f"Unexpected subpath {subpath} found in .npy path {path}")
save_array(file_path, data)
- elif file_path.suffix in (".h5", ".hdf", ".hdf5"):
+ elif suffix in (".h5", ".hdf", ".hdf5"):
if subpath is None:
dataset_path = DEFAULT_H5_DATASET_PATH
else:
@@ -275,22 +282,39 @@ def load_dataset_stat(path: Path):
def ensure_unzipped(source: Union[PermissiveFileSource, ZipPath], folder: Path):
"""unzip a (downloaded) **source** to a file in **folder** if source is a zip archive.
Always returns the path to the unzipped source (maybe source itself)"""
- local_weights_file = download(source).path
- if isinstance(local_weights_file, ZipPath):
- # source is inside a zip archive
- out_path = folder / local_weights_file.filename
- with local_weights_file.open("rb") as src, out_path.open("wb") as dst:
- assert not isinstance(src, TextIOWrapper)
- copyfileobj(src, dst)
-
- local_weights_file = out_path
-
- if zipfile.is_zipfile(local_weights_file):
+ weights_reader = get_reader(source)
+ out_path = folder / (
+ weights_reader.original_file_name or f"file{weights_reader.suffix}"
+ )
+
+ if zipfile.is_zipfile(weights_reader):
+ out_path = out_path.with_name(out_path.name + ".unzipped")
+ out_path.parent.mkdir(exist_ok=True, parents=True)
# source itself is a zipfile
- out_path = folder / local_weights_file.with_suffix(".unzipped").name
- with zipfile.ZipFile(local_weights_file, "r") as f:
+ with zipfile.ZipFile(weights_reader, "r") as f:
f.extractall(out_path)
- return out_path
else:
- return local_weights_file
+ out_path.parent.mkdir(exist_ok=True, parents=True)
+ with out_path.open("wb") as f:
+ copyfileobj(weights_reader, f)
+
+ return out_path
+
+
+def get_suffix(source: Union[ZipPath, FileSource]) -> str:
+ if isinstance(source, Path):
+ return source.suffix
+ elif isinstance(source, ZipPath):
+ return source.suffix
+ if isinstance(source, RelativeFilePath):
+ return source.path.suffix
+ elif isinstance(source, ZipPath):
+ return source.suffix
+ elif isinstance(source, HttpUrl):
+ if source.path is None:
+ return ""
+ else:
+ return PurePosixPath(source.path).suffix
+ else:
+ assert_never(source)
diff --git a/bioimageio/core/utils/_type_guards.py b/bioimageio/core/utils/_type_guards.py
new file mode 100644
index 00000000..3785c5e7
--- /dev/null
+++ b/bioimageio/core/utils/_type_guards.py
@@ -0,0 +1,8 @@
+"""use these type guards with caution!
+They widen the type to T[Any], which is not always correct."""
+
+from bioimageio.spec._internal import type_guards
+
+is_list = type_guards.is_list # pyright: ignore[reportPrivateImportUsage]
+is_ndarray = type_guards.is_ndarray # pyright: ignore[reportPrivateImportUsage]
+is_tuple = type_guards.is_tuple # pyright: ignore[reportPrivateImportUsage]
diff --git a/dev/env-dev.yaml b/dev/env-dev.yaml
index 13378376..38cbb289 100644
--- a/dev/env-dev.yaml
+++ b/dev/env-dev.yaml
@@ -5,10 +5,11 @@ channels:
- nodefaults
- pytorch
dependencies:
- - bioimageio.spec==0.5.4.1
+ - bioimageio.spec==0.5.4.3
- black
# - crick # currently requires python<=3.9
- h5py
+ - httpx
- imagecodecs
- imageio>=2.5
- jupyter
@@ -16,6 +17,7 @@ dependencies:
- keras>=3.0,<4
- loguru
- matplotlib
+ - napari
- numpy
- onnx
- onnxruntime
@@ -31,7 +33,7 @@ dependencies:
- pytest-cov
# - python=3.11 # removed
- pytorch>=2.1,<3
- - requests
+ - respx
- rich
- ruff
- ruyaml
diff --git a/dev/env-full.yaml b/dev/env-full.yaml
index a9dc0132..50c636c7 100644
--- a/dev/env-full.yaml
+++ b/dev/env-full.yaml
@@ -4,9 +4,9 @@ channels:
- nodefaults
- pytorch
dependencies:
- - bioimageio.spec==0.5.4.1
+ - bioimageio.spec==0.5.4.3
- black
- # - careamics # TODO: add careamics for model testing (currently pins pydantic to <2.9)
+ - careamics
- cellpose # for model testing
# - crick # currently requires python<=3.9
- h5py
@@ -18,6 +18,8 @@ dependencies:
- loguru
- matplotlib
- monai # for model testing
+ - napari
+
- numpy
- onnx
- onnxruntime
@@ -33,7 +35,8 @@ dependencies:
- pytest-cov
- python=3.11 # 3.12 not supported by cellpose->fastremap
- pytorch>=2.1,<3
- - requests
+ - httpx
+ - respx-mock
- rich
- ruff
- ruyaml
diff --git a/dev/env-gpu.yaml b/dev/env-gpu.yaml
index 7fc2123c..0cb97d73 100644
--- a/dev/env-gpu.yaml
+++ b/dev/env-gpu.yaml
@@ -4,7 +4,7 @@ channels:
- conda-forge
- nodefaults
dependencies:
- - bioimageio.spec==0.5.4.1
+ - bioimageio.spec==0.5.4.3
- black
- cellpose # for model testing
# - crick # currently requires python<=3.9
@@ -30,7 +30,8 @@ dependencies:
- pytest
- pytest-cov
- python=3.11
- - requests
+ - httpx
+ - respx-mock
- rich
- ruff
- ruyaml
diff --git a/dev/env-py38.yaml b/dev/env-py38.yaml
index 6fc6597a..490b8d50 100644
--- a/dev/env-py38.yaml
+++ b/dev/env-py38.yaml
@@ -5,7 +5,7 @@ channels:
- nodefaults
- pytorch
dependencies:
- - bioimageio.spec==0.5.4.1
+ - bioimageio.spec==0.5.4.3
- black
- crick # uncommented
- h5py
@@ -31,7 +31,8 @@ dependencies:
- pytest-cov
- python=3.8 # changed
- pytorch>=2.1,<3
- - requests
+ - httpx
+ - respx-mock
- rich
- ruff
- ruyaml
diff --git a/example/model_usage.ipynb b/example/model_usage.ipynb
index 8801b440..d6111a00 100644
--- a/example/model_usage.ipynb
+++ b/example/model_usage.ipynb
@@ -4,10 +4,11 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "# Bioimage Model Zoo Core Example notebook\n",
+ "# How to use bioimage.io models with bioimageio.core\n",
"\n",
- "This notebook shows how to interact with the `bioimageio.core` programmatically to explore, load, use, and export content from the [BioImage Model Zoo](https://bioimage.io).\n",
+ "This notebook shows how `bioimageio.core` can be used to load and deploy AI models shared in the [bioimage.io Model Zoo](https://bioimage.io).\n",
"\n",
+ "See the `bioimageio.spec` example notebook [load_model_and_create_your_own.ipynb](https://github.com/bioimage-io/spec-bioimage-io/blob/main/example/load_model_and_create_your_own.ipynb) for more details on loading and inspecting a model descriptions as well as how to create such descriptions in Python.\n",
"\n",
"quick links:\n",
"- [Create an input sample for a given model](#create_sample_for_model)"
@@ -17,43 +18,55 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "## 0. Activate human readable output error messages and load dependencies"
- ]
- },
- {
- "cell_type": "markdown",
- "metadata": {},
- "source": [
- "### 0.1. Install necessary dependencies"
+ "## 0. Setup\n",
+ "\n",
+ "### 0.1. Install dependencies"
]
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
+ "# type: ignore\n",
+ "dry_run: bool = True\n",
+ "\n",
+ "def install_missing(name: str, version: str):\n",
+ " if dry_run:\n",
+ " print(f\"Missing dependency {name}; would install {name}=={version}\")\n",
+ " else:\n",
+ " %pip install {name}=={version}\n",
+ "\n",
+ "\n",
+ "try:\n",
+ " import bioimageio.core\n",
+ "except ImportError:\n",
+ " install_missing(\"bioimageio.core\", \"0.8.0\")\n",
+ "\n",
"try:\n",
- " import matplotlib\n",
" import torch\n",
+ "except ImportError:\n",
+ " install_missing(\"torch\", \"2.5.1\")\n",
"\n",
- " import bioimageio.core\n",
+ "try:\n",
+ " import matplotlib\n",
"except ImportError:\n",
- " %pip install bioimageio.core==0.6.7 torch==2.3.1 matplotlib==3.9.0"
+ " install_missing(\"matplotlib\", \"3.9.0\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
- "### 0.2.Enable pretty_validation_errors\n",
+ "### 0.2 Prettify validation errors\n",
"\n",
- "This function displays validation errors in a human readable format."
+ "The `enable_pretty_validation_errors_in_ipynb` function registers an IPython exception handler to prettify raised validation errors."
]
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
@@ -68,51 +81,39 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "### 0.3. Load general dependencies"
+ "### 0.3. Utility functions\n",
+ "\n",
+ "Helper functions that are used in this notebook."
]
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
- "# Load general dependencies\n",
- "from pprint import pprint\n",
- "\n",
- "import matplotlib.pyplot as plt\n",
- "import numpy as np\n",
- "from imageio.v2 import imread\n",
- "\n",
+ "from bioimageio.core import Tensor\n",
"from bioimageio.spec.utils import download\n",
+ "from typing import Mapping\n",
"\n",
"\n",
- "# Function to display input and prediction output images\n",
- "def show_images(sample_tensor, prediction_tensor):\n",
- " input_array = sample_tensor.members[\"input0\"].data\n",
- "\n",
- " # Check for the number of channels to enable display\n",
- " input_array = np.squeeze(input_array)\n",
- " if len(input_array.shape) > 2:\n",
- " input_array = input_array[0]\n",
- "\n",
- " output_array = prediction_tensor.members[\"output0\"].data\n",
+ "try:\n",
+ " import napari\n",
+ "except ImportError:\n",
"\n",
- " # Check for the number of channels to enable display\n",
- " output_array = np.squeeze(output_array)\n",
- " if len(output_array.shape) > 2:\n",
- " output_array = output_array[0]\n",
+ " def show_images(images: Mapping[str, Tensor]):\n",
+ " for name, im in images.items():\n",
+ " im = im.data\n",
+ " print(f\"{name}: {im.shape}\")\n",
+ " # TODO: add plt.imshow\n",
"\n",
- " plt.figure()\n",
- " ax1 = plt.subplot(1, 2, 1)\n",
- " ax1.set_title(\"Input\")\n",
- " ax1.axis(\"off\")\n",
- " plt.imshow(input_array)\n",
- " ax2 = plt.subplot(1, 2, 2)\n",
- " ax2.set_title(\"Prediction\")\n",
- " ax2.axis(\"off\")\n",
- " plt.imshow(output_array)\n",
- " plt.show()"
+ "else:\n",
+ " def show_images(images: Mapping[str, Tensor]):\n",
+ " v = napari.Viewer()\n",
+ " for name, tensor in images.items():\n",
+ " im = tensor.data.data\n",
+ " print(f\"napari viewer: adding {name}\")\n",
+ " _ = v.add_image(im, name=name)\n"
]
},
{
@@ -151,18 +152,19 @@
"\n",
"Both of these options may be version specific (\"affable-shark/1\" or a version specific [__Zenodo__](https://zenodo.org/) backup [__DOI__](https://doi.org/)).\n",
"\n",
- "Alternatively, any rdf.yaml source, single file or in a .zip, may be loaded by providing its __local path__ or __URL__."
+ "Alternatively, any bioimageio.yaml source --- a single YAML file or a .zip package, may be loaded by providing its __local path__ or __URL__."
]
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
- "BMZ_MODEL_ID = \"\" # \"affable-shark\"\n",
- "BMZ_MODEL_DOI = \"\" # \"10.5281/zenodo.6287342\"\n",
- "BMZ_MODEL_URL = \"https://uk1s3.embassy.ebi.ac.uk/public-datasets/bioimage.io/affable-shark/draft/files/rdf.yaml\""
+ "MODEL_SOURCE = \"affable-shark\" # bioimageio ID\n",
+ "# MODEL_SOURCE = \"10.5281/zenodo.6287342\" # DOI of the backup hosted on Zenodo\n",
+ "# MODEL_SOURCE = \"https://uk1s3.embassy.ebi.ac.uk/public-datasets/bioimage.io/affable-shark/1.1/files/rdf.yaml\" # URL to bioimageio.yaml file\n",
+ "# MODEL_SOURCE = \"bioimageio_package.zip\" # path to a local zip file"
]
},
{
@@ -176,70 +178,573 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 5,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ " 0%| | 0.00/1.57k [00:00ModelDescr(\n",
+ " name='NucleiSegmentationBoundaryModel',\n",
+ " description='Nucleus segmentation for fluorescence microscopy',\n",
+ " covers=[RelativeFilePath(root=PureWindowsPath('cover.png'))],\n",
+ " id_emoji='๐ฆ',\n",
+ " authors=[\n",
+ " Author(\n",
+ " affiliation='EMBL Heidelberg',\n",
+ " email=None,\n",
+ " orcid=None,\n",
+ " name='Constantin Pape',\n",
+ " github_user='constantinpape'\n",
+ " )\n",
+ " ],\n",
+ " attachments=[\n",
+ " FileDescr(\n",
+ " source=RelativeFilePath(root=PureWindowsPath('zero_mean_unit_variance.ijm')),\n",
+ " sha256='767f2c3a50e36365c30b9e46e57fcf82e606d337e8a48d4a2440dc512813d186'\n",
+ " )\n",
+ " ],\n",
+ " cite=[\n",
+ " CiteEntry(text='training library', doi='10.5281/zenodo.5108853', url=None),\n",
+ " CiteEntry(text='architecture', doi='10.1007/978-3-319-24574-4_28', url=None),\n",
+ " CiteEntry(text='segmentation algorithm', doi='10.1038/nmeth.4151', url=None),\n",
+ " CiteEntry(text='data', doi=None, url='https://www.nature.com/articles/s41592-019-0612-7')\n",
+ " ],\n",
+ " license='CC-BY-4.0',\n",
+ " git_repo=None,\n",
+ " icon=None,\n",
+ " links=[\n",
+ " 'ilastik/stardist_dsb_training_data',\n",
+ " 'ilastik/ilastik',\n",
+ " 'deepimagej/deepimagej',\n",
+ " 'imjoy/BioImageIO-Packager'\n",
+ " ],\n",
+ " uploader=Uploader(email='thefynnbe@gmail.com', name='Fynn Beuttenmรผller'),\n",
+ " maintainers=[\n",
+ " Maintainer(\n",
+ " affiliation=None,\n",
+ " email=None,\n",
+ " orcid=None,\n",
+ " name='Constantin Pape',\n",
+ " github_user='constantinpape'\n",
+ " )\n",
+ " ],\n",
+ " tags=['fluorescence-light-microscopy', 'nuclei', 'instance-segmentation', 'unet', '2d'],\n",
+ " version=Version(root=1.1),\n",
+ " format_version='0.5.4',\n",
+ " type='model',\n",
+ " id='10.5281/zenodo.5764892/6647674',\n",
+ " documentation=RelativeFilePath(root=PureWindowsPath('documentation.md')),\n",
+ " inputs=[\n",
+ " InputTensorDescr(\n",
+ " id='input0',\n",
+ " description='',\n",
+ " axes=[\n",
+ " BatchAxis(id='batch', description='', type='batch', size=None),\n",
+ " ChannelAxis(id='channel', description='', type='channel', channel_names=['channel0']),\n",
+ " SpaceInputAxis(\n",
+ " size=ParameterizedSize(min=64, step=16),\n",
+ " id='y',\n",
+ " description='',\n",
+ " type='space',\n",
+ " unit=None,\n",
+ " scale=1.0,\n",
+ " concatenable=False\n",
+ " ),\n",
+ " SpaceInputAxis(\n",
+ " size=ParameterizedSize(min=64, step=16),\n",
+ " id='x',\n",
+ " description='',\n",
+ " type='space',\n",
+ " unit=None,\n",
+ " scale=1.0,\n",
+ " concatenable=False\n",
+ " )\n",
+ " ],\n",
+ " test_tensor=FileDescr(\n",
+ " source=RelativeFilePath(root=PureWindowsPath('test_input_0.npy')),\n",
+ " sha256='c29bd6e16e3f7856217b407ba948222b1c2a0da41922a0f79297e25588614fe2'\n",
+ " ),\n",
+ " sample_tensor=FileDescr(\n",
+ " source=RelativeFilePath(root=PureWindowsPath('sample_input_0.tif')),\n",
+ " sha256='a24b3c708b6ca6825494eb7c5a4d221335fb3eef5eb9d03f4108907cdaad2bf9'\n",
+ " ),\n",
+ " data=IntervalOrRatioDataDescr(\n",
+ " type='float32',\n",
+ " range=(None, None),\n",
+ " unit='arbitrary unit',\n",
+ " scale=1.0,\n",
+ " offset=None\n",
+ " ),\n",
+ " optional=False,\n",
+ " preprocessing=[\n",
+ " EnsureDtypeDescr(id='ensure_dtype', kwargs=EnsureDtypeKwargs(dtype='float32')),\n",
+ " ZeroMeanUnitVarianceDescr(\n",
+ " id='zero_mean_unit_variance',\n",
+ " kwargs=ZeroMeanUnitVarianceKwargs(axes=['channel', 'y', 'x'], eps=1e-06)\n",
+ " ),\n",
+ " EnsureDtypeDescr(id='ensure_dtype', kwargs=EnsureDtypeKwargs(dtype='float32'))\n",
+ " ]\n",
+ " )\n",
+ " ],\n",
+ " outputs=[\n",
+ " OutputTensorDescr(\n",
+ " id='output0',\n",
+ " description='',\n",
+ " axes=[\n",
+ " BatchAxis(id='batch', description='', type='batch', size=None),\n",
+ " ChannelAxis(id='channel', description='', type='channel', channel_names=['channel0', 'channel1']),\n",
+ " SpaceOutputAxisWithHalo(\n",
+ " halo=16,\n",
+ " size=SizeReference(tensor_id='input0', axis_id='y', offset=0),\n",
+ " id='y',\n",
+ " description='',\n",
+ " type='space',\n",
+ " unit=None,\n",
+ " scale=1.0\n",
+ " ),\n",
+ " SpaceOutputAxisWithHalo(\n",
+ " halo=16,\n",
+ " size=SizeReference(tensor_id='input0', axis_id='x', offset=0),\n",
+ " id='x',\n",
+ " description='',\n",
+ " type='space',\n",
+ " unit=None,\n",
+ " scale=1.0\n",
+ " )\n",
+ " ],\n",
+ " test_tensor=FileDescr(\n",
+ " source=RelativeFilePath(root=PureWindowsPath('test_output_0.npy')),\n",
+ " sha256='510181f38930e59e4fd8ecc03d6ea7c980eb6609759655f2d4a41fe36108d5f5'\n",
+ " ),\n",
+ " sample_tensor=FileDescr(\n",
+ " source=RelativeFilePath(root=PureWindowsPath('sample_output_0.tif')),\n",
+ " sha256='e8f99aabe8405427f515eba23a49f58ba50302f57d1fdfd07026e1984f836c5e'\n",
+ " ),\n",
+ " data=IntervalOrRatioDataDescr(\n",
+ " type='float32',\n",
+ " range=(None, None),\n",
+ " unit='arbitrary unit',\n",
+ " scale=1.0,\n",
+ " offset=None\n",
+ " ),\n",
+ " postprocessing=[EnsureDtypeDescr(id='ensure_dtype', kwargs=EnsureDtypeKwargs(dtype='float32'))]\n",
+ " )\n",
+ " ],\n",
+ " packaged_by=[],\n",
+ " parent=None,\n",
+ " run_mode=None,\n",
+ " timestamp=Datetime(root=datetime.datetime(2022, 6, 15, 20, 6, 22, 658325, tzinfo=datetime.timezone.utc)),\n",
+ " training_data=LinkedDataset(version=None, id='ilastik/stradist_dsb_training_data'),\n",
+ " weights=WeightsDescr(\n",
+ " keras_hdf5=None,\n",
+ " onnx=OnnxWeightsDescr(\n",
+ " source=RelativeFilePath(root=PureWindowsPath('weights.onnx')),\n",
+ " sha256='df913b85947f5132bcdaf81d91af0963f60d44f4caf8a4fec672d96a2f327b44',\n",
+ " authors=None,\n",
+ " parent='pytorch_state_dict',\n",
+ " comment='',\n",
+ " opset_version=12\n",
+ " ),\n",
+ " pytorch_state_dict=PytorchStateDictWeightsDescr(\n",
+ " source=RelativeFilePath(root=PureWindowsPath('weights.pt')),\n",
+ " sha256='608f52cd7f5119f7a7b8272395b0c169714e8be34536eaf159820f72a1d6a5b7',\n",
+ " authors=None,\n",
+ " parent=None,\n",
+ " comment='',\n",
+ " architecture=ArchitectureFromFileDescr(\n",
+ " source=RelativeFilePath(root=PureWindowsPath('unet.py')),\n",
+ " sha256='7f5b15948e8e2c91f78dcff34fbf30af517073e91ba487f3edb982b948d099b3',\n",
+ " callable='UNet2d',\n",
+ " kwargs={\n",
+ " 'depth': 4,\n",
+ " 'final_activation': 'Sigmoid',\n",
+ " 'gain': 2,\n",
+ " 'in_channels': 1,\n",
+ " 'initial_features': 64,\n",
+ " 'out_channels': 2,\n",
+ " 'postprocessing': None,\n",
+ " 'return_side_outputs': False\n",
+ " }\n",
+ " ),\n",
+ " pytorch_version=Version(root='1.10'),\n",
+ " dependencies=None\n",
+ " ),\n",
+ " tensorflow_js=None,\n",
+ " tensorflow_saved_model_bundle=None,\n",
+ " torchscript=TorchscriptWeightsDescr(\n",
+ " source=RelativeFilePath(root=PureWindowsPath('weights-torchscript.pt')),\n",
+ " sha256='8410950508655a300793b389c815dc30b1334062fc1dadb1e15e55a93cbb99a0',\n",
+ " authors=None,\n",
+ " parent='pytorch_state_dict',\n",
+ " comment='',\n",
+ " pytorch_version=Version(root='1.10')\n",
+ " )\n",
+ " ),\n",
+ " config=Config(\n",
+ " bioimageio=BioimageioConfig(\n",
+ " reproducibility_tolerance=(),\n",
+ " nickname='affable-shark',\n",
+ " nickname_icon='๐ฆ',\n",
+ " thumbnails={'cover.png': 'cover.thumbnail.png'}\n",
+ " ),\n",
+ " _conceptdoi='10.5281/zenodo.5764892',\n",
+ " deepimagej={\n",
+ " 'allow_tiling': True,\n",
+ " 'model_keys': None,\n",
+ " 'prediction': {\n",
+ " 'postprocess': [{'spec': None}],\n",
+ " 'preprocess': [{'kwargs': 'zero_mean_unit_variance.ijm', 'spec': 'ij.IJ::runMacroFile'}]\n",
+ " },\n",
+ " 'pyramidal_model': False,\n",
+ " 'test_information': {\n",
+ " 'inputs': [\n",
+ " {\n",
+ " 'name': 'sample_input_0.tif',\n",
+ " 'pixel_size': {'x': 1, 'y': 1, 'z': 1},\n",
+ " 'size': '256 x 256 x 1 x 1'\n",
+ " }\n",
+ " ],\n",
+ " 'memory_peak': None,\n",
+ " 'outputs': [{'name': 'sample_output_0.tif', 'size': '256 x 256 x 1 x 2', 'type': 'image'}],\n",
+ " 'runtime': None\n",
+ " }\n",
+ " }\n",
+ " )\n",
+ ")\n",
+ "\n"
+ ],
+ "text/plain": [
+ "\u001b[1;35mModelDescr\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33mname\u001b[0m=\u001b[32m'NucleiSegmentationBoundaryModel'\u001b[0m,\n",
+ " \u001b[33mdescription\u001b[0m=\u001b[32m'Nucleus segmentation for fluorescence microscopy'\u001b[0m,\n",
+ " \u001b[33mcovers\u001b[0m=\u001b[1m[\u001b[0m\u001b[1;35mRelativeFilePath\u001b[0m\u001b[1m(\u001b[0m\u001b[33mroot\u001b[0m=\u001b[1;35mPureWindowsPath\u001b[0m\u001b[1m(\u001b[0m\u001b[32m'cover.png'\u001b[0m\u001b[1m)\u001b[0m\u001b[1m)\u001b[0m\u001b[1m]\u001b[0m,\n",
+ " \u001b[33mid_emoji\u001b[0m=\u001b[32m'๐ฆ'\u001b[0m,\n",
+ " \u001b[33mauthors\u001b[0m=\u001b[1m[\u001b[0m\n",
+ " \u001b[1;35mAuthor\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33maffiliation\u001b[0m=\u001b[32m'EMBL Heidelberg'\u001b[0m,\n",
+ " \u001b[33memail\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33morcid\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33mname\u001b[0m=\u001b[32m'Constantin Pape'\u001b[0m,\n",
+ " \u001b[33mgithub_user\u001b[0m=\u001b[32m'constantinpape'\u001b[0m\n",
+ " \u001b[1m)\u001b[0m\n",
+ " \u001b[1m]\u001b[0m,\n",
+ " \u001b[33mattachments\u001b[0m=\u001b[1m[\u001b[0m\n",
+ " \u001b[1;35mFileDescr\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33msource\u001b[0m=\u001b[1;35mRelativeFilePath\u001b[0m\u001b[1m(\u001b[0m\u001b[33mroot\u001b[0m=\u001b[1;35mPureWindowsPath\u001b[0m\u001b[1m(\u001b[0m\u001b[32m'zero_mean_unit_variance.ijm'\u001b[0m\u001b[1m)\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33msha256\u001b[0m=\u001b[32m'767f2c3a50e36365c30b9e46e57fcf82e606d337e8a48d4a2440dc512813d186'\u001b[0m\n",
+ " \u001b[1m)\u001b[0m\n",
+ " \u001b[1m]\u001b[0m,\n",
+ " \u001b[33mcite\u001b[0m=\u001b[1m[\u001b[0m\n",
+ " \u001b[1;35mCiteEntry\u001b[0m\u001b[1m(\u001b[0m\u001b[33mtext\u001b[0m=\u001b[32m'training library'\u001b[0m, \u001b[33mdoi\u001b[0m=\u001b[32m'10.5281/zenodo.5108853'\u001b[0m, \u001b[33murl\u001b[0m=\u001b[3;35mNone\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[1;35mCiteEntry\u001b[0m\u001b[1m(\u001b[0m\u001b[33mtext\u001b[0m=\u001b[32m'architecture'\u001b[0m, \u001b[33mdoi\u001b[0m=\u001b[32m'10.1007/978-3-319-24574-4_28'\u001b[0m, \u001b[33murl\u001b[0m=\u001b[3;35mNone\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[1;35mCiteEntry\u001b[0m\u001b[1m(\u001b[0m\u001b[33mtext\u001b[0m=\u001b[32m'segmentation algorithm'\u001b[0m, \u001b[33mdoi\u001b[0m=\u001b[32m'10.1038/nmeth.4151'\u001b[0m, \u001b[33murl\u001b[0m=\u001b[3;35mNone\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[1;35mCiteEntry\u001b[0m\u001b[1m(\u001b[0m\u001b[33mtext\u001b[0m=\u001b[32m'data'\u001b[0m, \u001b[33mdoi\u001b[0m=\u001b[3;35mNone\u001b[0m, \u001b[33murl\u001b[0m=\u001b[32m'https://www.nature.com/articles/s41592-019-0612-7'\u001b[0m\u001b[1m)\u001b[0m\n",
+ " \u001b[1m]\u001b[0m,\n",
+ " \u001b[33mlicense\u001b[0m=\u001b[32m'CC-BY-4.0'\u001b[0m,\n",
+ " \u001b[33mgit_repo\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33micon\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33mlinks\u001b[0m=\u001b[1m[\u001b[0m\n",
+ " \u001b[32m'ilastik/stardist_dsb_training_data'\u001b[0m,\n",
+ " \u001b[32m'ilastik/ilastik'\u001b[0m,\n",
+ " \u001b[32m'deepimagej/deepimagej'\u001b[0m,\n",
+ " \u001b[32m'imjoy/BioImageIO-Packager'\u001b[0m\n",
+ " \u001b[1m]\u001b[0m,\n",
+ " \u001b[33muploader\u001b[0m=\u001b[1;35mUploader\u001b[0m\u001b[1m(\u001b[0m\u001b[33memail\u001b[0m=\u001b[32m'thefynnbe@gmail.com'\u001b[0m, \u001b[33mname\u001b[0m=\u001b[32m'Fynn Beuttenmรผller'\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33mmaintainers\u001b[0m=\u001b[1m[\u001b[0m\n",
+ " \u001b[1;35mMaintainer\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33maffiliation\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33memail\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33morcid\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33mname\u001b[0m=\u001b[32m'Constantin Pape'\u001b[0m,\n",
+ " \u001b[33mgithub_user\u001b[0m=\u001b[32m'constantinpape'\u001b[0m\n",
+ " \u001b[1m)\u001b[0m\n",
+ " \u001b[1m]\u001b[0m,\n",
+ " \u001b[33mtags\u001b[0m=\u001b[1m[\u001b[0m\u001b[32m'fluorescence-light-microscopy'\u001b[0m, \u001b[32m'nuclei'\u001b[0m, \u001b[32m'instance-segmentation'\u001b[0m, \u001b[32m'unet'\u001b[0m, \u001b[32m'2d'\u001b[0m\u001b[1m]\u001b[0m,\n",
+ " \u001b[33mversion\u001b[0m=\u001b[1;35mVersion\u001b[0m\u001b[1m(\u001b[0m\u001b[33mroot\u001b[0m=\u001b[1;36m1\u001b[0m\u001b[1;36m.1\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33mformat_version\u001b[0m=\u001b[32m'0.5.4'\u001b[0m,\n",
+ " \u001b[33mtype\u001b[0m=\u001b[32m'model'\u001b[0m,\n",
+ " \u001b[33mid\u001b[0m=\u001b[32m'10.5281/zenodo.5764892/6647674'\u001b[0m,\n",
+ " \u001b[33mdocumentation\u001b[0m=\u001b[1;35mRelativeFilePath\u001b[0m\u001b[1m(\u001b[0m\u001b[33mroot\u001b[0m=\u001b[1;35mPureWindowsPath\u001b[0m\u001b[1m(\u001b[0m\u001b[32m'documentation.md'\u001b[0m\u001b[1m)\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33minputs\u001b[0m=\u001b[1m[\u001b[0m\n",
+ " \u001b[1;35mInputTensorDescr\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33mid\u001b[0m=\u001b[32m'input0'\u001b[0m,\n",
+ " \u001b[33mdescription\u001b[0m=\u001b[32m''\u001b[0m,\n",
+ " \u001b[33maxes\u001b[0m=\u001b[1m[\u001b[0m\n",
+ " \u001b[1;35mBatchAxis\u001b[0m\u001b[1m(\u001b[0m\u001b[33mid\u001b[0m=\u001b[32m'batch'\u001b[0m, \u001b[33mdescription\u001b[0m=\u001b[32m''\u001b[0m, \u001b[33mtype\u001b[0m=\u001b[32m'batch'\u001b[0m, \u001b[33msize\u001b[0m=\u001b[3;35mNone\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[1;35mChannelAxis\u001b[0m\u001b[1m(\u001b[0m\u001b[33mid\u001b[0m=\u001b[32m'channel'\u001b[0m, \u001b[33mdescription\u001b[0m=\u001b[32m''\u001b[0m, \u001b[33mtype\u001b[0m=\u001b[32m'channel'\u001b[0m, \u001b[33mchannel_names\u001b[0m=\u001b[1m[\u001b[0m\u001b[32m'channel0'\u001b[0m\u001b[1m]\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[1;35mSpaceInputAxis\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33msize\u001b[0m=\u001b[1;35mParameterizedSize\u001b[0m\u001b[1m(\u001b[0m\u001b[33mmin\u001b[0m=\u001b[1;36m64\u001b[0m, \u001b[33mstep\u001b[0m=\u001b[1;36m16\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33mid\u001b[0m=\u001b[32m'y'\u001b[0m,\n",
+ " \u001b[33mdescription\u001b[0m=\u001b[32m''\u001b[0m,\n",
+ " \u001b[33mtype\u001b[0m=\u001b[32m'space'\u001b[0m,\n",
+ " \u001b[33munit\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33mscale\u001b[0m=\u001b[1;36m1\u001b[0m\u001b[1;36m.0\u001b[0m,\n",
+ " \u001b[33mconcatenable\u001b[0m=\u001b[3;91mFalse\u001b[0m\n",
+ " \u001b[1m)\u001b[0m,\n",
+ " \u001b[1;35mSpaceInputAxis\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33msize\u001b[0m=\u001b[1;35mParameterizedSize\u001b[0m\u001b[1m(\u001b[0m\u001b[33mmin\u001b[0m=\u001b[1;36m64\u001b[0m, \u001b[33mstep\u001b[0m=\u001b[1;36m16\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33mid\u001b[0m=\u001b[32m'x'\u001b[0m,\n",
+ " \u001b[33mdescription\u001b[0m=\u001b[32m''\u001b[0m,\n",
+ " \u001b[33mtype\u001b[0m=\u001b[32m'space'\u001b[0m,\n",
+ " \u001b[33munit\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33mscale\u001b[0m=\u001b[1;36m1\u001b[0m\u001b[1;36m.0\u001b[0m,\n",
+ " \u001b[33mconcatenable\u001b[0m=\u001b[3;91mFalse\u001b[0m\n",
+ " \u001b[1m)\u001b[0m\n",
+ " \u001b[1m]\u001b[0m,\n",
+ " \u001b[33mtest_tensor\u001b[0m=\u001b[1;35mFileDescr\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33msource\u001b[0m=\u001b[1;35mRelativeFilePath\u001b[0m\u001b[1m(\u001b[0m\u001b[33mroot\u001b[0m=\u001b[1;35mPureWindowsPath\u001b[0m\u001b[1m(\u001b[0m\u001b[32m'test_input_0.npy'\u001b[0m\u001b[1m)\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33msha256\u001b[0m=\u001b[32m'c29bd6e16e3f7856217b407ba948222b1c2a0da41922a0f79297e25588614fe2'\u001b[0m\n",
+ " \u001b[1m)\u001b[0m,\n",
+ " \u001b[33msample_tensor\u001b[0m=\u001b[1;35mFileDescr\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33msource\u001b[0m=\u001b[1;35mRelativeFilePath\u001b[0m\u001b[1m(\u001b[0m\u001b[33mroot\u001b[0m=\u001b[1;35mPureWindowsPath\u001b[0m\u001b[1m(\u001b[0m\u001b[32m'sample_input_0.tif'\u001b[0m\u001b[1m)\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33msha256\u001b[0m=\u001b[32m'a24b3c708b6ca6825494eb7c5a4d221335fb3eef5eb9d03f4108907cdaad2bf9'\u001b[0m\n",
+ " \u001b[1m)\u001b[0m,\n",
+ " \u001b[33mdata\u001b[0m=\u001b[1;35mIntervalOrRatioDataDescr\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33mtype\u001b[0m=\u001b[32m'float32'\u001b[0m,\n",
+ " \u001b[33mrange\u001b[0m=\u001b[1m(\u001b[0m\u001b[3;35mNone\u001b[0m, \u001b[3;35mNone\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33munit\u001b[0m=\u001b[32m'arbitrary unit'\u001b[0m,\n",
+ " \u001b[33mscale\u001b[0m=\u001b[1;36m1\u001b[0m\u001b[1;36m.0\u001b[0m,\n",
+ " \u001b[33moffset\u001b[0m=\u001b[3;35mNone\u001b[0m\n",
+ " \u001b[1m)\u001b[0m,\n",
+ " \u001b[33moptional\u001b[0m=\u001b[3;91mFalse\u001b[0m,\n",
+ " \u001b[33mpreprocessing\u001b[0m=\u001b[1m[\u001b[0m\n",
+ " \u001b[1;35mEnsureDtypeDescr\u001b[0m\u001b[1m(\u001b[0m\u001b[33mid\u001b[0m=\u001b[32m'ensure_dtype'\u001b[0m, \u001b[33mkwargs\u001b[0m=\u001b[1;35mEnsureDtypeKwargs\u001b[0m\u001b[1m(\u001b[0m\u001b[33mdtype\u001b[0m=\u001b[32m'float32'\u001b[0m\u001b[1m)\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[1;35mZeroMeanUnitVarianceDescr\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33mid\u001b[0m=\u001b[32m'zero_mean_unit_variance'\u001b[0m,\n",
+ " \u001b[33mkwargs\u001b[0m=\u001b[1;35mZeroMeanUnitVarianceKwargs\u001b[0m\u001b[1m(\u001b[0m\u001b[33maxes\u001b[0m=\u001b[1m[\u001b[0m\u001b[32m'channel'\u001b[0m, \u001b[32m'y'\u001b[0m, \u001b[32m'x'\u001b[0m\u001b[1m]\u001b[0m, \u001b[33meps\u001b[0m=\u001b[1;36m1e\u001b[0m\u001b[1;36m-06\u001b[0m\u001b[1m)\u001b[0m\n",
+ " \u001b[1m)\u001b[0m,\n",
+ " \u001b[1;35mEnsureDtypeDescr\u001b[0m\u001b[1m(\u001b[0m\u001b[33mid\u001b[0m=\u001b[32m'ensure_dtype'\u001b[0m, \u001b[33mkwargs\u001b[0m=\u001b[1;35mEnsureDtypeKwargs\u001b[0m\u001b[1m(\u001b[0m\u001b[33mdtype\u001b[0m=\u001b[32m'float32'\u001b[0m\u001b[1m)\u001b[0m\u001b[1m)\u001b[0m\n",
+ " \u001b[1m]\u001b[0m\n",
+ " \u001b[1m)\u001b[0m\n",
+ " \u001b[1m]\u001b[0m,\n",
+ " \u001b[33moutputs\u001b[0m=\u001b[1m[\u001b[0m\n",
+ " \u001b[1;35mOutputTensorDescr\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33mid\u001b[0m=\u001b[32m'output0'\u001b[0m,\n",
+ " \u001b[33mdescription\u001b[0m=\u001b[32m''\u001b[0m,\n",
+ " \u001b[33maxes\u001b[0m=\u001b[1m[\u001b[0m\n",
+ " \u001b[1;35mBatchAxis\u001b[0m\u001b[1m(\u001b[0m\u001b[33mid\u001b[0m=\u001b[32m'batch'\u001b[0m, \u001b[33mdescription\u001b[0m=\u001b[32m''\u001b[0m, \u001b[33mtype\u001b[0m=\u001b[32m'batch'\u001b[0m, \u001b[33msize\u001b[0m=\u001b[3;35mNone\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[1;35mChannelAxis\u001b[0m\u001b[1m(\u001b[0m\u001b[33mid\u001b[0m=\u001b[32m'channel'\u001b[0m, \u001b[33mdescription\u001b[0m=\u001b[32m''\u001b[0m, \u001b[33mtype\u001b[0m=\u001b[32m'channel'\u001b[0m, \u001b[33mchannel_names\u001b[0m=\u001b[1m[\u001b[0m\u001b[32m'channel0'\u001b[0m, \u001b[32m'channel1'\u001b[0m\u001b[1m]\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[1;35mSpaceOutputAxisWithHalo\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33mhalo\u001b[0m=\u001b[1;36m16\u001b[0m,\n",
+ " \u001b[33msize\u001b[0m=\u001b[1;35mSizeReference\u001b[0m\u001b[1m(\u001b[0m\u001b[33mtensor_id\u001b[0m=\u001b[32m'input0'\u001b[0m, \u001b[33maxis_id\u001b[0m=\u001b[32m'y'\u001b[0m, \u001b[33moffset\u001b[0m=\u001b[1;36m0\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33mid\u001b[0m=\u001b[32m'y'\u001b[0m,\n",
+ " \u001b[33mdescription\u001b[0m=\u001b[32m''\u001b[0m,\n",
+ " \u001b[33mtype\u001b[0m=\u001b[32m'space'\u001b[0m,\n",
+ " \u001b[33munit\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33mscale\u001b[0m=\u001b[1;36m1\u001b[0m\u001b[1;36m.0\u001b[0m\n",
+ " \u001b[1m)\u001b[0m,\n",
+ " \u001b[1;35mSpaceOutputAxisWithHalo\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33mhalo\u001b[0m=\u001b[1;36m16\u001b[0m,\n",
+ " \u001b[33msize\u001b[0m=\u001b[1;35mSizeReference\u001b[0m\u001b[1m(\u001b[0m\u001b[33mtensor_id\u001b[0m=\u001b[32m'input0'\u001b[0m, \u001b[33maxis_id\u001b[0m=\u001b[32m'x'\u001b[0m, \u001b[33moffset\u001b[0m=\u001b[1;36m0\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33mid\u001b[0m=\u001b[32m'x'\u001b[0m,\n",
+ " \u001b[33mdescription\u001b[0m=\u001b[32m''\u001b[0m,\n",
+ " \u001b[33mtype\u001b[0m=\u001b[32m'space'\u001b[0m,\n",
+ " \u001b[33munit\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33mscale\u001b[0m=\u001b[1;36m1\u001b[0m\u001b[1;36m.0\u001b[0m\n",
+ " \u001b[1m)\u001b[0m\n",
+ " \u001b[1m]\u001b[0m,\n",
+ " \u001b[33mtest_tensor\u001b[0m=\u001b[1;35mFileDescr\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33msource\u001b[0m=\u001b[1;35mRelativeFilePath\u001b[0m\u001b[1m(\u001b[0m\u001b[33mroot\u001b[0m=\u001b[1;35mPureWindowsPath\u001b[0m\u001b[1m(\u001b[0m\u001b[32m'test_output_0.npy'\u001b[0m\u001b[1m)\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33msha256\u001b[0m=\u001b[32m'510181f38930e59e4fd8ecc03d6ea7c980eb6609759655f2d4a41fe36108d5f5'\u001b[0m\n",
+ " \u001b[1m)\u001b[0m,\n",
+ " \u001b[33msample_tensor\u001b[0m=\u001b[1;35mFileDescr\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33msource\u001b[0m=\u001b[1;35mRelativeFilePath\u001b[0m\u001b[1m(\u001b[0m\u001b[33mroot\u001b[0m=\u001b[1;35mPureWindowsPath\u001b[0m\u001b[1m(\u001b[0m\u001b[32m'sample_output_0.tif'\u001b[0m\u001b[1m)\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33msha256\u001b[0m=\u001b[32m'e8f99aabe8405427f515eba23a49f58ba50302f57d1fdfd07026e1984f836c5e'\u001b[0m\n",
+ " \u001b[1m)\u001b[0m,\n",
+ " \u001b[33mdata\u001b[0m=\u001b[1;35mIntervalOrRatioDataDescr\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33mtype\u001b[0m=\u001b[32m'float32'\u001b[0m,\n",
+ " \u001b[33mrange\u001b[0m=\u001b[1m(\u001b[0m\u001b[3;35mNone\u001b[0m, \u001b[3;35mNone\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33munit\u001b[0m=\u001b[32m'arbitrary unit'\u001b[0m,\n",
+ " \u001b[33mscale\u001b[0m=\u001b[1;36m1\u001b[0m\u001b[1;36m.0\u001b[0m,\n",
+ " \u001b[33moffset\u001b[0m=\u001b[3;35mNone\u001b[0m\n",
+ " \u001b[1m)\u001b[0m,\n",
+ " \u001b[33mpostprocessing\u001b[0m=\u001b[1m[\u001b[0m\u001b[1;35mEnsureDtypeDescr\u001b[0m\u001b[1m(\u001b[0m\u001b[33mid\u001b[0m=\u001b[32m'ensure_dtype'\u001b[0m, \u001b[33mkwargs\u001b[0m=\u001b[1;35mEnsureDtypeKwargs\u001b[0m\u001b[1m(\u001b[0m\u001b[33mdtype\u001b[0m=\u001b[32m'float32'\u001b[0m\u001b[1m)\u001b[0m\u001b[1m)\u001b[0m\u001b[1m]\u001b[0m\n",
+ " \u001b[1m)\u001b[0m\n",
+ " \u001b[1m]\u001b[0m,\n",
+ " \u001b[33mpackaged_by\u001b[0m=\u001b[1m[\u001b[0m\u001b[1m]\u001b[0m,\n",
+ " \u001b[33mparent\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33mrun_mode\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33mtimestamp\u001b[0m=\u001b[1;35mDatetime\u001b[0m\u001b[1m(\u001b[0m\u001b[33mroot\u001b[0m=\u001b[1;35mdatetime\u001b[0m\u001b[1;35m.datetime\u001b[0m\u001b[1m(\u001b[0m\u001b[1;36m2022\u001b[0m, \u001b[1;36m6\u001b[0m, \u001b[1;36m15\u001b[0m, \u001b[1;36m20\u001b[0m, \u001b[1;36m6\u001b[0m, \u001b[1;36m22\u001b[0m, \u001b[1;36m658325\u001b[0m, \u001b[33mtzinfo\u001b[0m=\u001b[35mdatetime\u001b[0m.timezone.utc\u001b[1m)\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33mtraining_data\u001b[0m=\u001b[1;35mLinkedDataset\u001b[0m\u001b[1m(\u001b[0m\u001b[33mversion\u001b[0m=\u001b[3;35mNone\u001b[0m, \u001b[33mid\u001b[0m=\u001b[32m'ilastik/stradist_dsb_training_data'\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33mweights\u001b[0m=\u001b[1;35mWeightsDescr\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33mkeras_hdf5\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33monnx\u001b[0m=\u001b[1;35mOnnxWeightsDescr\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33msource\u001b[0m=\u001b[1;35mRelativeFilePath\u001b[0m\u001b[1m(\u001b[0m\u001b[33mroot\u001b[0m=\u001b[1;35mPureWindowsPath\u001b[0m\u001b[1m(\u001b[0m\u001b[32m'weights.onnx'\u001b[0m\u001b[1m)\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33msha256\u001b[0m=\u001b[32m'df913b85947f5132bcdaf81d91af0963f60d44f4caf8a4fec672d96a2f327b44'\u001b[0m,\n",
+ " \u001b[33mauthors\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33mparent\u001b[0m=\u001b[32m'pytorch_state_dict'\u001b[0m,\n",
+ " \u001b[33mcomment\u001b[0m=\u001b[32m''\u001b[0m,\n",
+ " \u001b[33mopset_version\u001b[0m=\u001b[1;36m12\u001b[0m\n",
+ " \u001b[1m)\u001b[0m,\n",
+ " \u001b[33mpytorch_state_dict\u001b[0m=\u001b[1;35mPytorchStateDictWeightsDescr\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33msource\u001b[0m=\u001b[1;35mRelativeFilePath\u001b[0m\u001b[1m(\u001b[0m\u001b[33mroot\u001b[0m=\u001b[1;35mPureWindowsPath\u001b[0m\u001b[1m(\u001b[0m\u001b[32m'weights.pt'\u001b[0m\u001b[1m)\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33msha256\u001b[0m=\u001b[32m'608f52cd7f5119f7a7b8272395b0c169714e8be34536eaf159820f72a1d6a5b7'\u001b[0m,\n",
+ " \u001b[33mauthors\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33mparent\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33mcomment\u001b[0m=\u001b[32m''\u001b[0m,\n",
+ " \u001b[33marchitecture\u001b[0m=\u001b[1;35mArchitectureFromFileDescr\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33msource\u001b[0m=\u001b[1;35mRelativeFilePath\u001b[0m\u001b[1m(\u001b[0m\u001b[33mroot\u001b[0m=\u001b[1;35mPureWindowsPath\u001b[0m\u001b[1m(\u001b[0m\u001b[32m'unet.py'\u001b[0m\u001b[1m)\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33msha256\u001b[0m=\u001b[32m'7f5b15948e8e2c91f78dcff34fbf30af517073e91ba487f3edb982b948d099b3'\u001b[0m,\n",
+ " \u001b[33mcallable\u001b[0m=\u001b[32m'UNet2d'\u001b[0m,\n",
+ " \u001b[33mkwargs\u001b[0m=\u001b[1m{\u001b[0m\n",
+ " \u001b[32m'depth'\u001b[0m: \u001b[1;36m4\u001b[0m,\n",
+ " \u001b[32m'final_activation'\u001b[0m: \u001b[32m'Sigmoid'\u001b[0m,\n",
+ " \u001b[32m'gain'\u001b[0m: \u001b[1;36m2\u001b[0m,\n",
+ " \u001b[32m'in_channels'\u001b[0m: \u001b[1;36m1\u001b[0m,\n",
+ " \u001b[32m'initial_features'\u001b[0m: \u001b[1;36m64\u001b[0m,\n",
+ " \u001b[32m'out_channels'\u001b[0m: \u001b[1;36m2\u001b[0m,\n",
+ " \u001b[32m'postprocessing'\u001b[0m: \u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[32m'return_side_outputs'\u001b[0m: \u001b[3;91mFalse\u001b[0m\n",
+ " \u001b[1m}\u001b[0m\n",
+ " \u001b[1m)\u001b[0m,\n",
+ " \u001b[33mpytorch_version\u001b[0m=\u001b[1;35mVersion\u001b[0m\u001b[1m(\u001b[0m\u001b[33mroot\u001b[0m=\u001b[32m'1.10'\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33mdependencies\u001b[0m=\u001b[3;35mNone\u001b[0m\n",
+ " \u001b[1m)\u001b[0m,\n",
+ " \u001b[33mtensorflow_js\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33mtensorflow_saved_model_bundle\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33mtorchscript\u001b[0m=\u001b[1;35mTorchscriptWeightsDescr\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33msource\u001b[0m=\u001b[1;35mRelativeFilePath\u001b[0m\u001b[1m(\u001b[0m\u001b[33mroot\u001b[0m=\u001b[1;35mPureWindowsPath\u001b[0m\u001b[1m(\u001b[0m\u001b[32m'weights-torchscript.pt'\u001b[0m\u001b[1m)\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33msha256\u001b[0m=\u001b[32m'8410950508655a300793b389c815dc30b1334062fc1dadb1e15e55a93cbb99a0'\u001b[0m,\n",
+ " \u001b[33mauthors\u001b[0m=\u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[33mparent\u001b[0m=\u001b[32m'pytorch_state_dict'\u001b[0m,\n",
+ " \u001b[33mcomment\u001b[0m=\u001b[32m''\u001b[0m,\n",
+ " \u001b[33mpytorch_version\u001b[0m=\u001b[1;35mVersion\u001b[0m\u001b[1m(\u001b[0m\u001b[33mroot\u001b[0m=\u001b[32m'1.10'\u001b[0m\u001b[1m)\u001b[0m\n",
+ " \u001b[1m)\u001b[0m\n",
+ " \u001b[1m)\u001b[0m,\n",
+ " \u001b[33mconfig\u001b[0m=\u001b[1;35mConfig\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33mbioimageio\u001b[0m=\u001b[1;35mBioimageioConfig\u001b[0m\u001b[1m(\u001b[0m\n",
+ " \u001b[33mreproducibility_tolerance\u001b[0m=\u001b[1m(\u001b[0m\u001b[1m)\u001b[0m,\n",
+ " \u001b[33mnickname\u001b[0m=\u001b[32m'affable-shark'\u001b[0m,\n",
+ " \u001b[33mnickname_icon\u001b[0m=\u001b[32m'๐ฆ'\u001b[0m,\n",
+ " \u001b[33mthumbnails\u001b[0m=\u001b[1m{\u001b[0m\u001b[32m'cover.png'\u001b[0m: \u001b[32m'cover.thumbnail.png'\u001b[0m\u001b[1m}\u001b[0m\n",
+ " \u001b[1m)\u001b[0m,\n",
+ " \u001b[33m_conceptdoi\u001b[0m=\u001b[32m'10.5281/zenodo.5764892'\u001b[0m,\n",
+ " \u001b[33mdeepimagej\u001b[0m=\u001b[1m{\u001b[0m\n",
+ " \u001b[32m'allow_tiling'\u001b[0m: \u001b[3;92mTrue\u001b[0m,\n",
+ " \u001b[32m'model_keys'\u001b[0m: \u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[32m'prediction'\u001b[0m: \u001b[1m{\u001b[0m\n",
+ " \u001b[32m'postprocess'\u001b[0m: \u001b[1m[\u001b[0m\u001b[1m{\u001b[0m\u001b[32m'spec'\u001b[0m: \u001b[3;35mNone\u001b[0m\u001b[1m}\u001b[0m\u001b[1m]\u001b[0m,\n",
+ " \u001b[32m'preprocess'\u001b[0m: \u001b[1m[\u001b[0m\u001b[1m{\u001b[0m\u001b[32m'kwargs'\u001b[0m: \u001b[32m'zero_mean_unit_variance.ijm'\u001b[0m, \u001b[32m'spec'\u001b[0m: \u001b[32m'ij.IJ::runMacroFile'\u001b[0m\u001b[1m}\u001b[0m\u001b[1m]\u001b[0m\n",
+ " \u001b[1m}\u001b[0m,\n",
+ " \u001b[32m'pyramidal_model'\u001b[0m: \u001b[3;91mFalse\u001b[0m,\n",
+ " \u001b[32m'test_information'\u001b[0m: \u001b[1m{\u001b[0m\n",
+ " \u001b[32m'inputs'\u001b[0m: \u001b[1m[\u001b[0m\n",
+ " \u001b[1m{\u001b[0m\n",
+ " \u001b[32m'name'\u001b[0m: \u001b[32m'sample_input_0.tif'\u001b[0m,\n",
+ " \u001b[32m'pixel_size'\u001b[0m: \u001b[1m{\u001b[0m\u001b[32m'x'\u001b[0m: \u001b[1;36m1\u001b[0m, \u001b[32m'y'\u001b[0m: \u001b[1;36m1\u001b[0m, \u001b[32m'z'\u001b[0m: \u001b[1;36m1\u001b[0m\u001b[1m}\u001b[0m,\n",
+ " \u001b[32m'size'\u001b[0m: \u001b[32m'256 x 256 x 1 x 1'\u001b[0m\n",
+ " \u001b[1m}\u001b[0m\n",
+ " \u001b[1m]\u001b[0m,\n",
+ " \u001b[32m'memory_peak'\u001b[0m: \u001b[3;35mNone\u001b[0m,\n",
+ " \u001b[32m'outputs'\u001b[0m: \u001b[1m[\u001b[0m\u001b[1m{\u001b[0m\u001b[32m'name'\u001b[0m: \u001b[32m'sample_output_0.tif'\u001b[0m, \u001b[32m'size'\u001b[0m: \u001b[32m'256 x 256 x 1 x 2'\u001b[0m, \u001b[32m'type'\u001b[0m: \u001b[32m'image'\u001b[0m\u001b[1m}\u001b[0m\u001b[1m]\u001b[0m,\n",
+ " \u001b[32m'runtime'\u001b[0m: \u001b[3;35mNone\u001b[0m\n",
+ " \u001b[1m}\u001b[0m\n",
+ " \u001b[1m}\u001b[0m\n",
+ " \u001b[1m)\u001b[0m\n",
+ "\u001b[1m)\u001b[0m\n"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
"source": [
- "pprint(model)"
+ "from rich import print\n",
+ "\n",
+ "print(model)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "We can use this metadata programmatically to explore and deploy (next section) the model. \n",
+ "For example we can checkout the model's cover images.\n",
+ "For a step-by-step inspection of other metadata fields, please refer to the [bioimageio.spec example notebook](https://github.com/bioimage-io/spec-bioimage-io/blob/main/example/load_model_and_create_your_own.ipynb)."
]
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 7,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ " 0%| | 0.00/89.1k [00:00"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
"source": [
- "print(f\"\\n Covers of the model '{model.name}' are: \")\n",
- "for cover in model.covers:\n",
- " cover_data = imread(download(cover).path)\n",
- " plt.figure(figsize=(10, 10))\n",
- " plt.imshow(cover_data)\n",
- " plt.xticks([])\n",
- " plt.yticks([])\n",
- " plt.show()"
+ "from bioimageio.core.io import load_image\n",
+ "import matplotlib.pyplot as plt\n",
+ "\n",
+ "for i, cover in enumerate(model.covers):\n",
+ " cover_data = load_image(cover)\n",
+ " _ = plt.figure(figsize=(10, 10))\n",
+ " _ = plt.imshow(cover_data)\n",
+ " _ = plt.xticks([])\n",
+ " _ = plt.yticks([])\n",
+ " _ = plt.title(f\"Cover image {i+1} of '{model.name}'\")\n",
+ " _ = plt.show()"
]
},
{
@@ -260,29 +765,798 @@
"\n",
"----\n",
"\n",
- "`bioimageio.core.test_model` returns a validation dictionary with 'status'='passed'/'failed' and other detailed information that can be inspected by calling `.display()` on it.\n",
+ "`bioimageio.core.test_model` returns a validation dictionary with `status='passed'/'failed'` and other detailed information that can be inspected by calling `.display()` on it.\n",
+ "The validation sumary may also be saved (`.save()`) as JSON, Markdown or HTML file.\n",
+ "Validation summaries may be loaded from json (`.load_json()`).\n",
"\n",
"The validation summary will indicate:\n",
"- the versions of the `bioimageio.spec` and `bioimageio.core` libraries used to run the validation\n",
"- the status of several validation steps\n",
" - โ๏ธ: Success\n",
- " - ๐: information about the validation context\n",
" - โ : Warning\n",
" - โ: Error"
]
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 8,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "\u001b[32m2025-04-02 16:57:44.264\u001b[0m | \u001b[34m\u001b[1mDEBUG \u001b[0m | \u001b[36mbioimageio.core._resource_tests\u001b[0m:\u001b[36m_test_model_inference\u001b[0m:\u001b[36m593\u001b[0m - \u001b[34m\u001b[1mstarting 'Reproduce test outputs from test inputs (onnx)'\u001b[0m\n",
+ " 0%| | 0.00/116M [00:00โญโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ Traceback (most recent call last) โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฎ\n",
+ "โ C:\\repos\\bioimage-io\\core-bioimage-io-python\\bioimageio\\core\\_resource_tests.py:768 in _test_mod โ\n",
+ "โ โ\n",
+ "โ 765 โ โ ) as prediction_pipeline: โ\n",
+ "โ 766 โ โ โ for n, batch_size, inputs, exptected_output_shape in generate_test_cases(): โ\n",
+ "โ 767 โ โ โ โ error: Optional[str] = None โ\n",
+ "โ โฑ 768 โ โ โ โ result = prediction_pipeline.predict_sample_without_blocking(inputs) โ\n",
+ "โ 769 โ โ โ โ if len(result.members) != len(exptected_output_shape): โ\n",
+ "โ 770 โ โ โ โ โ error = ( โ\n",
+ "โ 771 โ โ โ โ โ โ f\"Expected {len(exptected_output_shape)} outputs,\" โ\n",
+ "โ โ\n",
+ "โ C:\\repos\\bioimage-io\\core-bioimage-io-python\\bioimageio\\core\\_prediction_pipeline.py:160 in pred โ\n",
+ "โ โ\n",
+ "โ 157 โ โ if not skip_preprocessing: โ\n",
+ "โ 158 โ โ โ self.apply_preprocessing(sample) โ\n",
+ "โ 159 โ โ โ\n",
+ "โ โฑ 160 โ โ output = self._adapter.forward(sample) โ\n",
+ "โ 161 โ โ if not skip_postprocessing: โ\n",
+ "โ 162 โ โ โ self.apply_postprocessing(output) โ\n",
+ "โ 163 โ\n",
+ "โ โ\n",
+ "โ C:\\repos\\bioimage-io\\core-bioimage-io-python\\bioimageio\\core\\backends\\_model_adapter.py:205 in f โ\n",
+ "โ โ\n",
+ "โ 202 โ โ โ ) โ\n",
+ "โ 203 โ โ โ for in_id, in_order in zip(self._input_ids, self._input_axes) โ\n",
+ "โ 204 โ โ ] โ\n",
+ "โ โฑ 205 โ โ output_arrays = self._forward_impl(input_arrays) โ\n",
+ "โ 206 โ โ assert len(output_arrays) <= len(self._output_ids) โ\n",
+ "โ 207 โ โ output_tensors = [ โ\n",
+ "โ 208 โ โ โ None if a is None else Tensor(a, dims=d) โ\n",
+ "โ โ\n",
+ "โ C:\\repos\\bioimage-io\\core-bioimage-io-python\\bioimageio\\core\\backends\\onnx_backend.py:40 in _for โ\n",
+ "โ โ\n",
+ "โ 37 โ def _forward_impl( โ\n",
+ "โ 38 โ โ self, input_arrays: Sequence[Optional[NDArray[Any]]] โ\n",
+ "โ 39 โ ) -> List[Optional[NDArray[Any]]]: โ\n",
+ "โ โฑ 40 โ โ result: Any = self._session.run( โ\n",
+ "โ 41 โ โ โ None, dict(zip(self._input_names, input_arrays)) โ\n",
+ "โ 42 โ โ ) โ\n",
+ "โ 43 โ โ if is_list(result) or is_tuple(result): โ\n",
+ "โ โ\n",
+ "โ C:\\Users\\fbeut\\miniforge3\\envs\\core\\Lib\\site-packages\\onnxruntime\\capi\\onnxruntime_inference_col โ\n",
+ "โ โ\n",
+ "โ 263 โ โ if not output_names: โ\n",
+ "โ 264 โ โ โ output_names = [output.name for output in self._outputs_meta] โ\n",
+ "โ 265 โ โ try: โ\n",
+ "โ โฑ 266 โ โ โ return self._sess.run(output_names, input_feed, run_options) โ\n",
+ "โ 267 โ โ except C.EPFail as err: โ\n",
+ "โ 268 โ โ โ if self._enable_fallback: โ\n",
+ "โ 269 โ โ โ โ print(f\"EP Error: {err!s} using {self._providers}\") โ\n",
+ "โฐโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฏ\n",
+ "InvalidArgument: [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Got invalid dimensions for input: input.1 for the following indices\n",
+ " index: 2 Got: 64 Expected: 256\n",
+ " index: 3 Got: 64 Expected: 256\n",
+ " Please fix either the inputs/outputs or the model.\n",
+ "\n"
+ ],
+ "text/plain": [
+ "\u001b[31mโญโ\u001b[0m\u001b[31mโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\u001b[0m\u001b[31m \u001b[0m\u001b[1;31mTraceback \u001b[0m\u001b[1;2;31m(most recent call last)\u001b[0m\u001b[31m \u001b[0m\u001b[31mโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ\u001b[0m\u001b[31mโโฎ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[33mC:\\repos\\bioimage-io\\core-bioimage-io-python\\bioimageio\\core\\_resource_tests.py\u001b[0m:\u001b[94m768\u001b[0m in \u001b[92m_test_mod\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m765 \u001b[0m\u001b[2mโ โ \u001b[0m) \u001b[94mas\u001b[0m prediction_pipeline: \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m766 \u001b[0m\u001b[2mโ โ โ \u001b[0m\u001b[94mfor\u001b[0m n, batch_size, inputs, exptected_output_shape \u001b[95min\u001b[0m generate_test_cases(): \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m767 \u001b[0m\u001b[2mโ โ โ โ \u001b[0merror: Optional[\u001b[96mstr\u001b[0m] = \u001b[94mNone\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[31mโฑ \u001b[0m768 \u001b[2mโ โ โ โ \u001b[0mresult = prediction_pipeline.predict_sample_without_blocking(inputs) \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m769 \u001b[0m\u001b[2mโ โ โ โ \u001b[0m\u001b[94mif\u001b[0m \u001b[96mlen\u001b[0m(result.members) != \u001b[96mlen\u001b[0m(exptected_output_shape): \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m770 \u001b[0m\u001b[2mโ โ โ โ โ \u001b[0merror = ( \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m771 \u001b[0m\u001b[2mโ โ โ โ โ โ \u001b[0m\u001b[33mf\u001b[0m\u001b[33m\"\u001b[0m\u001b[33mExpected \u001b[0m\u001b[33m{\u001b[0m\u001b[96mlen\u001b[0m(exptected_output_shape)\u001b[33m}\u001b[0m\u001b[33m outputs,\u001b[0m\u001b[33m\"\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[33mC:\\repos\\bioimage-io\\core-bioimage-io-python\\bioimageio\\core\\_prediction_pipeline.py\u001b[0m:\u001b[94m160\u001b[0m in \u001b[92mpred\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m157 \u001b[0m\u001b[2mโ โ \u001b[0m\u001b[94mif\u001b[0m \u001b[95mnot\u001b[0m skip_preprocessing: \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m158 \u001b[0m\u001b[2mโ โ โ \u001b[0m\u001b[96mself\u001b[0m.apply_preprocessing(sample) \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m159 \u001b[0m\u001b[2mโ โ \u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[31mโฑ \u001b[0m160 \u001b[2mโ โ \u001b[0moutput = \u001b[1;4;96mself\u001b[0m\u001b[1;4m._adapter.forward(sample)\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m161 \u001b[0m\u001b[2mโ โ \u001b[0m\u001b[94mif\u001b[0m \u001b[95mnot\u001b[0m skip_postprocessing: \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m162 \u001b[0m\u001b[2mโ โ โ \u001b[0m\u001b[96mself\u001b[0m.apply_postprocessing(output) \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m163 \u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[33mC:\\repos\\bioimage-io\\core-bioimage-io-python\\bioimageio\\core\\backends\\_model_adapter.py\u001b[0m:\u001b[94m205\u001b[0m in \u001b[92mf\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m202 \u001b[0m\u001b[2mโ โ โ \u001b[0m) \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m203 \u001b[0m\u001b[2mโ โ โ \u001b[0m\u001b[94mfor\u001b[0m in_id, in_order \u001b[95min\u001b[0m \u001b[96mzip\u001b[0m(\u001b[96mself\u001b[0m._input_ids, \u001b[96mself\u001b[0m._input_axes) \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m204 \u001b[0m\u001b[2mโ โ \u001b[0m] \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[31mโฑ \u001b[0m205 \u001b[2mโ โ \u001b[0moutput_arrays = \u001b[1;4;96mself\u001b[0m\u001b[1;4m._forward_impl(input_arrays)\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m206 \u001b[0m\u001b[2mโ โ \u001b[0m\u001b[94massert\u001b[0m \u001b[96mlen\u001b[0m(output_arrays) <= \u001b[96mlen\u001b[0m(\u001b[96mself\u001b[0m._output_ids) \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m207 \u001b[0m\u001b[2mโ โ \u001b[0moutput_tensors = [ \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m208 \u001b[0m\u001b[2mโ โ โ \u001b[0m\u001b[94mNone\u001b[0m \u001b[94mif\u001b[0m a \u001b[95mis\u001b[0m \u001b[94mNone\u001b[0m \u001b[94melse\u001b[0m Tensor(a, dims=d) \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[33mC:\\repos\\bioimage-io\\core-bioimage-io-python\\bioimageio\\core\\backends\\onnx_backend.py\u001b[0m:\u001b[94m40\u001b[0m in \u001b[92m_for\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m37 \u001b[0m\u001b[2mโ \u001b[0m\u001b[94mdef\u001b[0m \u001b[92m_forward_impl\u001b[0m( \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m38 \u001b[0m\u001b[2mโ โ \u001b[0m\u001b[96mself\u001b[0m, input_arrays: Sequence[Optional[NDArray[Any]]] \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m39 \u001b[0m\u001b[2mโ \u001b[0m) -> List[Optional[NDArray[Any]]]: \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[31mโฑ \u001b[0m40 \u001b[2mโ โ \u001b[0mresult: Any = \u001b[1;4;96mself\u001b[0m\u001b[1;4m._session.run(\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m41 \u001b[0m\u001b[1;2;4mโ โ โ \u001b[0m\u001b[1;4;94mNone\u001b[0m\u001b[1;4m, \u001b[0m\u001b[1;4;96mdict\u001b[0m\u001b[1;4m(\u001b[0m\u001b[1;4;96mzip\u001b[0m\u001b[1;4m(\u001b[0m\u001b[1;4;96mself\u001b[0m\u001b[1;4m._input_names, input_arrays))\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m42 \u001b[0m\u001b[1;2;4mโ โ \u001b[0m\u001b[1;4m)\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m43 \u001b[0m\u001b[2mโ โ \u001b[0m\u001b[94mif\u001b[0m is_list(result) \u001b[95mor\u001b[0m is_tuple(result): \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[33mC:\\Users\\fbeut\\miniforge3\\envs\\core\\Lib\\site-packages\\onnxruntime\\capi\\onnxruntime_inference_col\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m 263 \u001b[0m\u001b[2mโ โ \u001b[0m\u001b[94mif\u001b[0m \u001b[95mnot\u001b[0m output_names: \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m 264 \u001b[0m\u001b[2mโ โ โ \u001b[0moutput_names = [output.name \u001b[94mfor\u001b[0m output \u001b[95min\u001b[0m \u001b[96mself\u001b[0m._outputs_meta] \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m 265 \u001b[0m\u001b[2mโ โ \u001b[0m\u001b[94mtry\u001b[0m: \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[31mโฑ \u001b[0m 266 \u001b[2mโ โ โ \u001b[0m\u001b[94mreturn\u001b[0m \u001b[96mself\u001b[0m._sess.run(output_names, input_feed, run_options) \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m 267 \u001b[0m\u001b[2mโ โ \u001b[0m\u001b[94mexcept\u001b[0m C.EPFail \u001b[94mas\u001b[0m err: \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m 268 \u001b[0m\u001b[2mโ โ โ \u001b[0m\u001b[94mif\u001b[0m \u001b[96mself\u001b[0m._enable_fallback: \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโ\u001b[0m \u001b[2m 269 \u001b[0m\u001b[2mโ โ โ โ \u001b[0m\u001b[96mprint\u001b[0m(\u001b[33mf\u001b[0m\u001b[33m\"\u001b[0m\u001b[33mEP Error: \u001b[0m\u001b[33m{\u001b[0merr\u001b[33m!s}\u001b[0m\u001b[33m using \u001b[0m\u001b[33m{\u001b[0m\u001b[96mself\u001b[0m._providers\u001b[33m}\u001b[0m\u001b[33m\"\u001b[0m) \u001b[31mโ\u001b[0m\n",
+ "\u001b[31mโฐโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฏ\u001b[0m\n",
+ "\u001b[1;91mInvalidArgument: \u001b[0m\u001b[1m[\u001b[0mONNXRuntimeError\u001b[1m]\u001b[0m : \u001b[1;36m2\u001b[0m : INVALID_ARGUMENT : Got invalid dimensions for input: input.\u001b[1;36m1\u001b[0m for the following indices\n",
+ " index: \u001b[1;36m2\u001b[0m Got: \u001b[1;36m64\u001b[0m Expected: \u001b[1;36m256\u001b[0m\n",
+ " index: \u001b[1;36m3\u001b[0m Got: \u001b[1;36m64\u001b[0m Expected: \u001b[1;36m256\u001b[0m\n",
+ " Please fix either the inputs/outputs or the model.\n"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "\u001b[32m2025-04-02 16:58:06.465\u001b[0m | \u001b[34m\u001b[1mDEBUG \u001b[0m | \u001b[36mbioimageio.core._resource_tests\u001b[0m:\u001b[36m_test_model_inference\u001b[0m:\u001b[36m593\u001b[0m - \u001b[34m\u001b[1mstarting 'Reproduce test outputs from test inputs (pytorch_state_dict)'\u001b[0m\n",
+ "\u001b[32m2025-04-02 16:58:09.841\u001b[0m | \u001b[1mINFO \u001b[0m | \u001b[36mbioimageio.core._resource_tests\u001b[0m:\u001b[36m_test_model_inference_parametrized\u001b[0m:\u001b[36m705\u001b[0m - \u001b[1mTesting inference with 6 different inputs (B, N): {(1, 2), (2, 1), (1, 1), (2, 0), (2, 2), (1, 0)}\u001b[0m\n",
+ "\u001b[32m2025-04-02 16:58:13.088\u001b[0m | \u001b[34m\u001b[1mDEBUG \u001b[0m | \u001b[36mbioimageio.core._resource_tests\u001b[0m:\u001b[36m_test_model_inference\u001b[0m:\u001b[36m593\u001b[0m - \u001b[34m\u001b[1mstarting 'Reproduce test outputs from test inputs (torchscript)'\u001b[0m\n",
+ " 0%| | 0.00/116M [00:00\n",
+ " \n",
+ " | โ | \n",
+ " bioimageio format validation | \n",
+ "
\n",
+ " \n",
+ " | status | \n",
+ " failed | \n",
+ "
\n",
+ " \n",
+ " | source | \n",
+ " https://hypha.aicell.io/bioimage-io/artifacts/affable-shark/files/rdf.yaml | \n",
+ "
\n",
+ " \n",
+ " | id | \n",
+ " 10.5281/zenodo.5764892/6647674 | \n",
+ "
\n",
+ " \n",
+ " | format version | \n",
+ " model 0.5.4 | \n",
+ "
\n",
+ " \n",
+ " | bioimageio.core | \n",
+ " 0.8.0 | \n",
+ "
\n",
+ " \n",
+ " | bioimageio.spec | \n",
+ " 0.5.4.2 | \n",
+ "
\n",
+ "\n",
+ "\n",
+ " \n",
+ " | \n",
+ " Location | \n",
+ " Details | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " | \n",
+ " Successfully created `ModelDescr` instance. | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " | \n",
+ " bioimageio.spec format validation model 0.5.4 | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " type | \n",
+ " Has expected resource type | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " weights.onnx | \n",
+ " Reproduce test outputs from test inputs (onnx) | \n",
+ "
\n",
+ " \n",
+ " | \n",
+ " weights.onnx | \n",
+ " recommended conda environment (Reproduce test outputs from test inputs (onnx)) | \n",
+ "
\n",
+ " \n",
+ " | \n",
+ " | \n",
+ " %YAML 1.2\n",
+ "---\n",
+ "channels:\n",
+ " - conda-forge\n",
+ " - nodefaults\n",
+ "dependencies:\n",
+ " - conda-forge::bioimageio.core\n",
+ " - onnxruntime\n",
+ " - pip\n",
+ " | \n",
+ "
\n",
+ " \n",
+ " | \n",
+ " weights.onnx | \n",
+ " conda compare (Reproduce test outputs from test inputs (onnx)) | \n",
+ "
\n",
+ " \n",
+ " | \n",
+ " | \n",
+ " bioimageio.core not found | \n",
+ "
\n",
+ " \n",
+ " | โ | \n",
+ " weights.onnx | \n",
+ " Run onnx inference for parametrized inputs | \n",
+ "
\n",
+ " \n",
+ " | โ | \n",
+ " weights.onnx | \n",
+ " [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Got invalid dimensions for input: input.1 for the following indices\n",
+ " index: 2 Got: 64 Expected: 256\n",
+ " index: 3 Got: 64 Expected: 256\n",
+ " Please fix either the inputs/outputs or the model. | \n",
+ "
\n",
+ " \n",
+ " | \n",
+ " | \n",
+ " See Traceback 1. | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " weights.pytorch_state_dict | \n",
+ " Reproduce test outputs from test inputs (pytorch_state_dict) | \n",
+ "
\n",
+ " \n",
+ " | \n",
+ " weights.pytorch_state_dict | \n",
+ " recommended conda environment (Reproduce test outputs from test inputs (pytorch_state_dict)) | \n",
+ "
\n",
+ " \n",
+ " | \n",
+ " | \n",
+ " %YAML 1.2\n",
+ "---\n",
+ "channels:\n",
+ " - pytorch\n",
+ " - conda-forge\n",
+ " - nodefaults\n",
+ "dependencies:\n",
+ " - conda-forge::bioimageio.core\n",
+ " - mkl ==2024.0.0\n",
+ " - numpy <2\n",
+ " - pip\n",
+ " - pytorch==1.10.0\n",
+ " - setuptools <70.0.0\n",
+ " - torchaudio==0.10.0\n",
+ " - torchvision==0.11.0\n",
+ " | \n",
+ "
\n",
+ " \n",
+ " | \n",
+ " weights.pytorch_state_dict | \n",
+ " conda compare (Reproduce test outputs from test inputs (pytorch_state_dict)) | \n",
+ "
\n",
+ " \n",
+ " | \n",
+ " | \n",
+ " bioimageio.core not found mkl found but mismatch. Specification pkg: mkl==2024.0.0, Running pkg:\n",
+ "mkl==2024.2.2=h66d3029_15 pytorch found but mismatch. Specification pkg: pytorch==1.10.0,\n",
+ "Running pkg: pytorch==2.6.0=cpu_mkl_py311_hde8219b_101 setuptools found but mismatch.\n",
+ "Specification pkg: setuptools[version='<70.0.0'], Running pkg: setuptools==75.3.0=pyhd8ed1ab_0\n",
+ "torchaudio not found torchvision found but mismatch. Specification pkg: torchvision==0.11.0,\n",
+ "Running pkg: torchvision==0.21.0=cpu_py311_hc601fa1_0 | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " weights.pytorch_state_dict | \n",
+ " Run pytorch_state_dict inference for inputs with batch_size: 1 and size parameter n: 0 | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " weights.pytorch_state_dict | \n",
+ " Run pytorch_state_dict inference for inputs with batch_size: 1 and size parameter n: 1 | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " weights.pytorch_state_dict | \n",
+ " Run pytorch_state_dict inference for inputs with batch_size: 1 and size parameter n: 2 | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " weights.pytorch_state_dict | \n",
+ " Run pytorch_state_dict inference for inputs with batch_size: 2 and size parameter n: 0 | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " weights.pytorch_state_dict | \n",
+ " Run pytorch_state_dict inference for inputs with batch_size: 2 and size parameter n: 1 | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " weights.pytorch_state_dict | \n",
+ " Run pytorch_state_dict inference for inputs with batch_size: 2 and size parameter n: 2 | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " weights.torchscript | \n",
+ " Reproduce test outputs from test inputs (torchscript) | \n",
+ "
\n",
+ " \n",
+ " | \n",
+ " weights.torchscript | \n",
+ " recommended conda environment (Reproduce test outputs from test inputs (torchscript)) | \n",
+ "
\n",
+ " \n",
+ " | \n",
+ " | \n",
+ " %YAML 1.2\n",
+ "---\n",
+ "channels:\n",
+ " - pytorch\n",
+ " - conda-forge\n",
+ " - nodefaults\n",
+ "dependencies:\n",
+ " - conda-forge::bioimageio.core\n",
+ " - mkl ==2024.0.0\n",
+ " - numpy <2\n",
+ " - pip\n",
+ " - pytorch==1.10.0\n",
+ " - setuptools <70.0.0\n",
+ " - torchaudio==0.10.0\n",
+ " - torchvision==0.11.0\n",
+ " | \n",
+ "
\n",
+ " \n",
+ " | \n",
+ " weights.torchscript | \n",
+ " conda compare (Reproduce test outputs from test inputs (torchscript)) | \n",
+ "
\n",
+ " \n",
+ " | \n",
+ " | \n",
+ " bioimageio.core not found mkl found but mismatch. Specification pkg: mkl==2024.0.0, Running pkg:\n",
+ "mkl==2024.2.2=h66d3029_15 pytorch found but mismatch. Specification pkg: pytorch==1.10.0,\n",
+ "Running pkg: pytorch==2.6.0=cpu_mkl_py311_hde8219b_101 setuptools found but mismatch.\n",
+ "Specification pkg: setuptools[version='<70.0.0'], Running pkg: setuptools==75.3.0=pyhd8ed1ab_0\n",
+ "torchaudio not found torchvision found but mismatch. Specification pkg: torchvision==0.11.0,\n",
+ "Running pkg: torchvision==0.21.0=cpu_py311_hc601fa1_0 | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " weights.torchscript | \n",
+ " Run torchscript inference for inputs with batch_size: 1 and size parameter n: 0 | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " weights.torchscript | \n",
+ " Run torchscript inference for inputs with batch_size: 1 and size parameter n: 1 | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " weights.torchscript | \n",
+ " Run torchscript inference for inputs with batch_size: 1 and size parameter n: 2 | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " weights.torchscript | \n",
+ " Run torchscript inference for inputs with batch_size: 2 and size parameter n: 0 | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " weights.torchscript | \n",
+ " Run torchscript inference for inputs with batch_size: 2 and size parameter n: 1 | \n",
+ "
\n",
+ " \n",
+ " | โ๏ธ | \n",
+ " weights.torchscript | \n",
+ " Run torchscript inference for inputs with batch_size: 2 and size parameter n: 2 | \n",
+ "
\n",
+ "
\n",
+ "Traceback 1
\n",
+ "\n",
+ "\n",
+ "\n",
+ "\n",
+ "\n",
+ "\n",
+ "\n",
+ " โญโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ Traceback (most recent call last) โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฎ\n",
+ "โ C:\\repos\\bioimage-io\\core-bioimage-io-python\\bioimageio\\core\\_resource_tests.py:768 in _test_mod โ\n",
+ "โ โ\n",
+ "โ 765 โ โ ) as prediction_pipeline: โ\n",
+ "โ 766 โ โ โ for n, batch_size, inputs, exptected_output_shape in generate_test_cases(): โ\n",
+ "โ 767 โ โ โ โ error: Optional[str] = None โ\n",
+ "โ โฑ 768 โ โ โ โ result = prediction_pipeline.predict_sample_without_blocking(inputs) โ\n",
+ "โ 769 โ โ โ โ if len(result.members) != len(exptected_output_shape): โ\n",
+ "โ 770 โ โ โ โ โ error = ( โ\n",
+ "โ 771 โ โ โ โ โ โ f"Expected {len(exptected_output_shape)} outputs," โ\n",
+ "โ โ\n",
+ "โ C:\\repos\\bioimage-io\\core-bioimage-io-python\\bioimageio\\core\\_prediction_pipeline.py:160 in pred โ\n",
+ "โ โ\n",
+ "โ 157 โ โ if not skip_preprocessing: โ\n",
+ "โ 158 โ โ โ self.apply_preprocessing(sample) โ\n",
+ "โ 159 โ โ โ\n",
+ "โ โฑ 160 โ โ output = self._adapter.forward(sample) โ\n",
+ "โ 161 โ โ if not skip_postprocessing: โ\n",
+ "โ 162 โ โ โ self.apply_postprocessing(output) โ\n",
+ "โ 163 โ\n",
+ "โ โ\n",
+ "โ C:\\repos\\bioimage-io\\core-bioimage-io-python\\bioimageio\\core\\backends\\_model_adapter.py:205 in f โ\n",
+ "โ โ\n",
+ "โ 202 โ โ โ ) โ\n",
+ "โ 203 โ โ โ for in_id, in_order in zip(self._input_ids, self._input_axes) โ\n",
+ "โ 204 โ โ ] โ\n",
+ "โ โฑ 205 โ โ output_arrays = self._forward_impl(input_arrays) โ\n",
+ "โ 206 โ โ assert len(output_arrays) <= len(self._output_ids) โ\n",
+ "โ 207 โ โ output_tensors = [ โ\n",
+ "โ 208 โ โ โ None if a is None else Tensor(a, dims=d) โ\n",
+ "โ โ\n",
+ "โ C:\\repos\\bioimage-io\\core-bioimage-io-python\\bioimageio\\core\\backends\\onnx_backend.py:40 in _for โ\n",
+ "โ โ\n",
+ "โ 37 โ def _forward_impl( โ\n",
+ "โ 38 โ โ self, input_arrays: Sequence[Optional[NDArray[Any]]] โ\n",
+ "โ 39 โ ) -> List[Optional[NDArray[Any]]]: โ\n",
+ "โ โฑ 40 โ โ result: Any = self._session.run( โ\n",
+ "โ 41 โ โ โ None, dict(zip(self._input_names, input_arrays)) โ\n",
+ "โ 42 โ โ ) โ\n",
+ "โ 43 โ โ if is_list(result) or is_tuple(result): โ\n",
+ "โ โ\n",
+ "โ C:\\Users\\fbeut\\miniforge3\\envs\\core\\Lib\\site-packages\\onnxruntime\\capi\\onnxruntime_inference_col โ\n",
+ "โ โ\n",
+ "โ 263 โ โ if not output_names: โ\n",
+ "โ 264 โ โ โ output_names = [output.name for output in self._outputs_meta] โ\n",
+ "โ 265 โ โ try: โ\n",
+ "โ โฑ 266 โ โ โ return self._sess.run(output_names, input_feed, run_options) โ\n",
+ "โ 267 โ โ except C.EPFail as err: โ\n",
+ "โ 268 โ โ โ if self._enable_fallback: โ\n",
+ "โ 269 โ โ โ โ print(f"EP Error: {err!s} using {self._providers}") โ\n",
+ "โฐโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโฏ\n",
+ "InvalidArgument: [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Got invalid dimensions for input: input.1 for the following indices\n",
+ " index: 2 Got: 64 Expected: 256\n",
+ " index: 3 Got: 64 Expected: 256\n",
+ " Please fix either the inputs/outputs or the model.\n",
+ "
\n",
+ "\n",
+ "\n",
+ "
"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
"test_summary.display()"
]
},
+ {
+ "cell_type": "code",
+ "execution_count": 10,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stderr",
+ "output_type": "stream",
+ "text": [
+ "\u001b[32m2025-04-02 20:48:09.875\u001b[0m | \u001b[1mINFO \u001b[0m | \u001b[36mbioimageio.spec.summary\u001b[0m:\u001b[36msave_json\u001b[0m:\u001b[36m443\u001b[0m - \u001b[1mSaved summary to C:\\repos\\bioimage-io\\core-bioimage-io-python\\example\\summary\\summary.json\u001b[0m\n",
+ "\u001b[32m2025-04-02 20:48:09.898\u001b[0m | \u001b[1mINFO \u001b[0m | \u001b[36mbioimageio.spec.summary\u001b[0m:\u001b[36msave_markdown\u001b[0m:\u001b[36m450\u001b[0m - \u001b[1mSaved Markdown formatted summary to C:\\repos\\bioimage-io\\core-bioimage-io-python\\example\\summary\\summary.md\u001b[0m\n",
+ "\u001b[32m2025-04-02 20:48:09.962\u001b[0m | \u001b[1mINFO \u001b[0m | \u001b[36mbioimageio.spec.summary\u001b[0m:\u001b[36msave_html\u001b[0m:\u001b[36m458\u001b[0m - \u001b[1mSaved HTML formatted summary to C:\\repos\\bioimage-io\\core-bioimage-io-python\\example\\summary\\summary.html\u001b[0m\n"
+ ]
+ }
+ ],
+ "source": [
+ "_ = test_summary.save(\"summary\")"
+ ]
+ },
{
"cell_type": "markdown",
"metadata": {},
@@ -298,14 +1572,28 @@
"\n",
"This includes functions to run predictions on `numpy.ndarray`/`xarray.DataArray` as input and convenience functions to run predictions for images stored on disc.\n",
"\n",
- "### 3.1. Load the test image and convert into a tensor"
+ "### 3.1. Create an input sample"
]
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 11,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "array shape: (1, 1, 256, 256)\n",
+ "
\n"
+ ],
+ "text/plain": [
+ "array shape: \u001b[1m(\u001b[0m\u001b[1;36m1\u001b[0m, \u001b[1;36m1\u001b[0m, \u001b[1;36m256\u001b[0m, \u001b[1;36m256\u001b[0m\u001b[1m)\u001b[0m\n"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
"source": [
"from bioimageio.spec.model import v0_5\n",
"from bioimageio.spec.utils import load_array\n",
@@ -328,9 +1616,23 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 12,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "tensor shape: Frozen({'batch': 1, 'channel': 1, 'y': 256, 'x': 256})\n",
+ "\n"
+ ],
+ "text/plain": [
+ "tensor shape: \u001b[1;35mFrozen\u001b[0m\u001b[1m(\u001b[0m\u001b[1m{\u001b[0m\u001b[32m'batch'\u001b[0m: \u001b[1;36m1\u001b[0m, \u001b[32m'channel'\u001b[0m: \u001b[1;36m1\u001b[0m, \u001b[32m'y'\u001b[0m: \u001b[1;36m256\u001b[0m, \u001b[32m'x'\u001b[0m: \u001b[1;36m256\u001b[0m\u001b[1m}\u001b[0m\u001b[1m)\u001b[0m\n"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
"source": [
"from bioimageio.core import Tensor\n",
"\n",
@@ -351,9 +1653,20 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 13,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "Sample(members={'raw': }, stat=None, id='sample-from-numpy')"
+ ]
+ },
+ "execution_count": 13,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
"source": [
"from bioimageio.core import Sample\n",
"\n",
@@ -366,15 +1679,44 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "`bioimageio.core` provides the helper function `create_sample_for_model` to automatically create the `Sample` for the given model.\n",
- ""
+ "#### 3.1.1 Create an input sample with create_sample_for_model\n",
+ "\n",
+ "`bioimageio.core` provides the helper function `create_sample_for_model` to automatically create the `Sample` for the given model."
]
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 14,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "data": {
+ "text/html": [
+ "input paths: {'input0': \n",
+ "WindowsPath('C:/Users/fbeut/AppData/Local/bioimageio/bioimageio/Cache/277e1abf8d7287b159459f2288fbcfa1-test_input_0\n",
+ ".npy')}\n",
+ "\n"
+ ],
+ "text/plain": [
+ "input paths: \u001b[1m{\u001b[0m\u001b[32m'input0'\u001b[0m: \n",
+ "\u001b[1;35mWindowsPath\u001b[0m\u001b[1m(\u001b[0m\u001b[32m'C:/Users/fbeut/AppData/Local/bioimageio/bioimageio/Cache/277e1abf8d7287b159459f2288fbcfa1-test_input_0\u001b[0m\n",
+ "\u001b[32m.npy'\u001b[0m\u001b[1m)\u001b[0m\u001b[1m}\u001b[0m\n"
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ },
+ {
+ "data": {
+ "text/plain": [
+ "Sample(members={'input0': }, stat={}, id='my_demo_sample')"
+ ]
+ },
+ "execution_count": 14,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
"source": [
"from bioimageio.core.digest_spec import create_sample_for_model\n",
"from bioimageio.spec.utils import download\n",
@@ -393,14 +1735,26 @@
"cell_type": "markdown",
"metadata": {},
"source": [
- "There is also a helper function `get_test_inputs` to directly import the test input sample for a given model."
+ "#### 3.1.2 Get the test input sample\n",
+ "There is also a helper function `get_test_inputs` to directly import the test input sample for a given model."
]
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 15,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "data": {
+ "text/plain": [
+ "Sample(members={'input0': }, stat={}, id='test-sample')"
+ ]
+ },
+ "execution_count": 15,
+ "metadata": {},
+ "output_type": "execute_result"
+ }
+ ],
"source": [
"from bioimageio.core.digest_spec import get_test_inputs\n",
"\n",
@@ -432,7 +1786,7 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 16,
"metadata": {},
"outputs": [],
"source": [
@@ -457,9 +1811,21 @@
},
{
"cell_type": "code",
- "execution_count": null,
+ "execution_count": 17,
"metadata": {},
- "outputs": [],
+ "outputs": [
+ {
+ "ename": "TypeError",
+ "evalue": "show_images() takes 1 positional argument but 2 were given",
+ "output_type": "error",
+ "traceback": [
+ "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
+ "\u001b[1;31mTypeError\u001b[0m Traceback (most recent call last)",
+ "Cell \u001b[1;32mIn[17], line 4\u001b[0m\n\u001b[0;32m 1\u001b[0m prediction: Sample \u001b[38;5;241m=\u001b[39m prediction_pipeline\u001b[38;5;241m.\u001b[39mpredict_sample_without_blocking(sample)\n\u001b[0;32m 3\u001b[0m \u001b[38;5;66;03m# show the prediction result\u001b[39;00m\n\u001b[1;32m----> 4\u001b[0m \u001b[43mshow_images\u001b[49m\u001b[43m(\u001b[49m\u001b[43msample\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mprediction\u001b[49m\u001b[43m)\u001b[49m\n",
+ "\u001b[1;31mTypeError\u001b[0m: show_images() takes 1 positional argument but 2 were given"
+ ]
+ }
+ ],
"source": [
"prediction: Sample = prediction_pipeline.predict_sample_without_blocking(sample)\n",
"\n",
@@ -544,7 +1910,7 @@
],
"metadata": {
"kernelspec": {
- "display_name": "bioimageio-core",
+ "display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
@@ -558,7 +1924,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
- "version": "3.9.19"
+ "version": "3.11.11"
}
},
"nbformat": 4,
diff --git a/pyproject.toml b/pyproject.toml
index 5d58fe72..db264984 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -40,7 +40,7 @@ typeCheckingMode = "strict"
useLibraryCodeForTypes = true
[tool.pytest.ini_options]
-addopts = "--cov bioimageio --cov-report xml --cov-append --capture no --doctest-modules --failed-first --ignore dogfood --ignore bioimageio/core/backends --ignore bioimageio/core/weight_converters"
+addopts = "--doctest-modules --failed-first --ignore dogfood --ignore bioimageio/core/backends --ignore bioimageio/core/weight_converters"
testpaths = ["bioimageio/core", "tests"]
[tool.ruff]
diff --git a/setup.py b/setup.py
index f4e975a5..bfaa4f4b 100644
--- a/setup.py
+++ b/setup.py
@@ -30,7 +30,7 @@
],
packages=find_namespace_packages(exclude=["tests"]),
install_requires=[
- "bioimageio.spec ==0.5.4.1",
+ "bioimageio.spec ==0.5.4.3",
"h5py",
"imagecodecs",
"imageio>=2.10",
@@ -38,7 +38,6 @@
"numpy",
"pydantic-settings>=2.5,<3",
"pydantic>=2.7.0,<3",
- "requests",
"ruyaml",
"tqdm",
"typing-extensions",
@@ -58,6 +57,7 @@
+ [
"black",
"cellpose", # for model testing
+ "httpx",
"jupyter-black",
"jupyter",
"matplotlib",
@@ -67,7 +67,7 @@
"packaging>=17.0",
"pdoc",
"pre-commit",
- "pyright==1.1.396",
+ "pyright==1.1.401",
"segment-anything", # for model testing
"timm", # for model testing
# "crick", # currently requires python<=3.9
diff --git a/tests/test_bioimageio_collection.py b/tests/test_bioimageio_collection.py
index fc81034c..d01868e6 100644
--- a/tests/test_bioimageio_collection.py
+++ b/tests/test_bioimageio_collection.py
@@ -1,8 +1,8 @@
import os
from typing import Any, Collection, Dict, Iterable, Mapping, Tuple
+import httpx
import pytest
-import requests
from pydantic import HttpUrl
from bioimageio.spec import InvalidDescr
@@ -13,7 +13,7 @@
def _get_latest_rdf_sources():
- entries: Any = requests.get(BASE_URL + "all_versions.json").json()["entries"]
+ entries: Any = httpx.get(BASE_URL + "all_versions.json").json()["entries"]
ret: Dict[str, Tuple[HttpUrl, Sha256]] = {}
for entry in entries:
version = entry["versions"][0]
@@ -40,7 +40,6 @@ def yield_bioimageio_yaml_urls() -> Iterable[ParameterSet]:
KNOWN_INVALID: Collection[str] = {
- "affable-shark/1.1", # onnx weights expect fixed input shape
"affectionate-cow/0.1.0", # custom dependencies
"ambitious-sloth/1.2", # requires inferno
"committed-turkey/1.2", # error deserializing VarianceScaling