π Geometric Computer Vision Library for Spatial AI
Python artificial-intelligencecomputer-visiondeep-learninghacktoberfestimage-processingmachine-learningneural-networkpythonpytorchroboticsspatial-ai
kornia.onnx
===========
.. meta::
:description: API reference for kornia.onnx: ONNXModule, ONNXSequential to chain ONNX models and Kornia operators into one graph, and ONNXLoader to fetch pre-exported operators and models from the Hugging Face Hub.
.. currentmodule:: kornia.onnx
Run and compose ONNX models with ONNX Runtime: :class:`~kornia.onnx.sequential.ONNXSequential` chains several
ONNX graphs (local files, ``onnx.ModelProto`` objects, or pre-exported Kornia operators and models from the
``kornia/ONNX_models`` Hugging Face repository) into one model with a PyTorch-like call interface. The
:doc:`ONNX support guide </get-started/onnx>` walks through installation, chaining, export and running on CUDA.
.. autoclass:: kornia.onnx.module.ONNXModule
:members:
.. autofunction:: kornia.onnx.load
.. autoclass:: kornia.onnx.sequential.ONNXSequential
:members:
.. autoclass:: kornia.onnx.utils.ONNXLoader
:members:
.. code-block:: python
from kornia.onnx.utils import ONNXLoader
# Load a Hugging Face operator
ONNXLoader.load_model("hf://operators/kornia.color.gray.GrayscaleToRgb")
# Load a downloaded operator from the cache: the same name, served from the cache directory
# (".kornia_hub/onnx_models" by default) without a download
ONNXLoader.load_model("hf://operators/kornia.color.gray.GrayscaleToRgb", download=False)
# Load a local file by its path
ONNXLoader.load_model(".kornia_hub/onnx_models/operators/kornia.color.gray.GrayscaleToRgb.onnx")
.. autofunction:: kornia.onnx.utils.add_metadata
.. autofunction:: kornia.onnx.utils.io_name_conversion
.. autoclass:: kornia.onnx.download.CachedDownloader
:members: