Prediction failure of images from disks with varying z-depths

[aidsouza]

Describe the bug

When predicting for images with varying z-depths on a disk, the program iterates through all images, but fails to load the predicted images into napari with an array concatenation error

To Reproduce

A folder with two image stacks (CZYX) , with identical X and Y shapes, but different Z shapes.

Error message:

# np = <module 'numpy' from \anaconda3\\envs\\careamics-napari-env\\Lib\\site-packages\\numpy\\__init__.py'>    395     else:
    396         samples = update.value

ValueError: all the input array dimensions except for the concatenation axis must match exactly, but along dimension 2, the array at index 0 has size 61 and the array at index 1 has size 41

System

Important

• OS: Windows
• Python version == 3.12.13
• PyTorch version == 2.9.1+cu126
• PyTorch lightning version == 2.6.1
• CAREamics version == 0.0.22

Environment

name: careamics-napari-env
channels:
  - pkgs/main
dependencies:
  - bzip2=1.0.8=h2bbff1b_6
  - ca-certificates=2026.3.19=haa95532_0
  - libexpat=2.7.5=hd7fb8db_0
  - libffi=3.4.4=hd77b12b_1
  - libzlib=1.3.1=h02ab6af_0
  - openssl=3.5.5=hbb43b14_0
  - packaging=26.0=py312haa95532_0
  - pip=26.0.1=pyhc872135_1
  - python=3.12.13=h63b1a2d_1
  - setuptools=82.0.1=py312haa95532_0
  - sqlite=3.51.2=hee5a0db_0
  - tk=8.6.15=hf199647_0
  - tzdata=2026a=he532380_0
  - ucrt=10.0.22621.0=haa95532_0
  - vc=14.3=h2df5915_10
  - vc14_runtime=14.44.35208=h4927774_10
  - vs2015_runtime=14.44.35208=ha6b5a95_10
  - wheel=0.46.3=py312haa95532_0
  - xz=5.8.2=h53af0af_0
  - zlib=1.3.1=h02ab6af_0
  - pip:
    - Deprecated==1.3.1
    - HeapDict==1.0.1
    - ImageIO==2.37.3
    - Jinja2==3.1.6
    - Markdown==3.10.2
    - MarkupSafe==3.0.3
    - Pint==0.25.3
    - PyOpenGL==3.1.10
    - PyQt6==6.11.0
    - PyQt6-Qt6==6.11.0
    - PyQt6_sip==13.11.1
    - PyYAML==6.0.3
    - Pygments==2.20.0
    - QtPy==2.4.3
    - Werkzeug==3.1.8
    - absl-py==2.4.0
    - aiohappyeyeballs==2.6.1
    - aiohttp==3.13.5
    - aiosignal==1.4.0
    - annotated-doc==0.0.4
    - annotated-types==0.7.0
    - anyio==4.13.0
    - app-model==0.5.1
    - appdirs==1.4.4
    - asttokens==3.0.1
    - attrs==26.1.0
    - bermuda==0.1.7
    - bioimageio.core==0.10.0
    - bioimageio.spec==0.5.9.0
    - build==1.4.2
    - cachey==0.2.1
    - careamics==0.0.22
    - careamics-napari==0.0.20
    - careamics-portfolio==0.0.15
    - certifi==2026.2.25
    - charset-normalizer==3.4.7
    - click==8.3.2
    - cloudpickle==3.1.2
    - colorama==0.4.6
    - comm==0.2.3
    - contourpy==1.3.3
    - cycler==0.12.1
    - dask==2026.3.0
    - debugpy==1.8.20
    - decorator==5.2.1
    - distro==1.9.0
    - dnspython==2.8.0
    - docstring_parser==0.17.0
    - donfig==0.8.1.post1
    - email-validator==2.3.0
    - exceptiongroup==1.3.1
    - executing==2.2.1
    - filelock==3.25.2
    - flexcache==0.3
    - flexparser==0.4
    - fonttools==4.62.1
    - freetype-py==2.5.1
    - frozenlist==1.8.0
    - fsspec==2026.2.0
    - genericache==0.5.2
    - google-crc32c==1.8.0
    - grpcio==1.80.0
    - h11==0.16.0
    - hsluv==5.0.4
    - httpcore==1.0.9
    - httpx==0.28.1
    - idna==3.11
    - imagecodecs==2026.3.6
    - in-n-out==0.2.1
    - ipykernel==6.31.0
    - ipython==9.12.0
    - ipython_pygments_lexers==1.1.1
    - jedi==0.19.2
    - jsonschema==4.26.0
    - jsonschema-specifications==2025.9.1
    - jupyter_client==8.8.0
    - jupyter_core==5.9.1
    - kiwisolver==1.5.0
    - lazy-loader==0.5
    - lightning-utilities==0.15.3
    - llvmlite==0.47.0
    - locket==1.0.0
    - loguru==0.7.3
    - magicgui==0.10.1
    - markdown-it-py==4.0.0
    - matplotlib==3.10.8
    - matplotlib-inline==0.2.1
    - mdurl==0.1.2
    - microssim==0.0.3
    - mpmath==1.3.0
    - multidict==6.7.1
    - napari==0.7.0
    - napari-console==0.1.4
    - napari-metadata==0.3.0
    - napari-plugin-engine==0.2.1
    - napari-plugin-manager==0.1.11
    - napari-svg==0.2.1
    - nest-asyncio==1.6.0
    - networkx==3.6.1
    - npe2==0.8.2
    - numba==0.65.0
    - numcodecs==0.15.1
    - numpy==2.4.3
    - pandas==3.0.2
    - parso==0.8.6
    - partd==1.4.2
    - pillow==12.1.1
    - platformdirs==4.9.6
    - pooch==1.9.0
    - prompt_toolkit==3.0.52
    - propcache==0.4.1
    - protobuf==5.29.1
    - psutil==7.2.2
    - psygnal==0.15.1
    - pure_eval==0.2.3
    - pyconify==0.2.1
    - pydantic==2.12.5
    - pydantic-extra-types==2.11.2
    - pydantic-settings==2.13.1
    - pydantic_core==2.41.5
    - pyparsing==3.3.2
    - pyproject_hooks==1.2.0
    - pyqtgraph==0.14.0
    - python-dateutil==2.9.0.post0
    - python-dotenv==1.2.2
    - pytorch-lightning==2.6.1
    - pywin32==311
    - pyzmq==27.1.0
    - qtconsole==5.7.2
    - referencing==0.37.0
    - requests==2.33.1
    - rich==14.3.3
    - rpds-py==0.30.0
    - ruyaml==0.91.0
    - scikit-image==0.26.0
    - scipy==1.17.0
    - setuptools==70.2.0
    - shellingham==1.5.4
    - six==1.17.0
    - stack-data==0.6.3
    - superqt==0.8.1
    - sympy==1.14.0
    - tensorboard==2.20.0
    - tensorboard-data-server==0.7.2
    - tifffile==2026.2.15
    - tomli_w==1.2.0
    - toolz==1.1.0
    - torch==2.9.1+cu126
    - torchmetrics==1.4.3
    - torchvision==0.24.1+cu126
    - tornado==6.5.5
    - tqdm==4.67.1
    - traitlets==5.14.3
    - triangle==20250106
    - typer==0.23.1
    - typing-inspection==0.4.2
    - typing_extensions==4.15.0
    - tzdata==2026.1
    - urllib3==2.6.3
    - validators==0.35.0
    - vispy==0.16.1
    - wcwidth==0.6.0
    - win32_setctime==1.2.0
    - wrapt==2.1.2
    - xarray==2026.2.0
    - yarl==1.23.0
    - zarr==3.1.6
    - zipp==3.23.0

[jdeschamps]

Hi @aidsouza,

Thanks for the report! We are finalizing the new version of the plugin (for the next version of CAREamics, v0.2.0), we will have the following:

• checkbox to predict directly to the disk
• a check on the sizes of the returned prediction to catch any mismatch, following your issue!

We should be able to release it this week or the next. Since we will release it "early", there will probably be some small problems here and there, so don't hesitate to keep opening issues so we can fix them asap.

Thanks again!