core-bioimage-io-python
core-bioimage-io-python copied to clipboard
Python libraries for loading, running and packaging bioimage.io models
There have been some recent changes to the weight converters and it seems not everything there is up-to-date. In `weight_converter/torch/_torchscript` the function converts the weights and returns the corresponding weight...
plan: - [x] pdoc documentation for bioimageio.core API users - [x] update README.md extend and point to pdoc docs for bioimageio.core API users - [ ] add CONTRIBUTING.md for new...
Pytorch lightning model format is just a `pytorch_state_dict` with more entries. It should be quite straightforward to make a model adaptor for it, or adapt the current adaptor. Potentially a...
Hi all, Due to my stuff not working anymore complaining about some pytorch internal issue and core-dumps, I re-installed my conda environment but still my previously working code won't work...
this came up in offline discussions and was mentioned in https://github.com/bioimage-io/spec-bioimage-io/issues/579
Halo! When building a model and adding the deepImageJ config with the IJ macros, if the pre and post-processing need to use the same pipeline (e.g., scale range), there is...
Our ONNX converters could be improved. To convert from pytorch [ONNX dynamo](https://pytorch.org/docs/stable/onnx_dynamo.html) looks promising.
The following code runs into an assertion error (extracted minimum example of a previously working implementation): ``` import bioimageio.core rdf_doi = "10.5281/zenodo.7274275" model_resource = bioimageio.core.load_resource_description(rdf_doi) ``` Output: ``` /home/julian/miniconda3/envs/cebra-em-all-env2/lib/python3.9/site-packages/bioimageio/spec/shared/_resolve_source.py:437: CacheWarning:...
We should improve our network weight converters and investigate which additional weight converters we can add.