Shalin Mehta

@ziw-liu @edyoshikun let's use the following. 2.5D UNet: 3D input -> 3D encoder -> 2D decoder -> 2D output 3D UNet: 3D input -> 3D encoder -> 3D decoder ->...

@ziw-liu > I have been thinking about using [dependency injection with lightning configs](https://lightning.ai/docs/pytorch/stable/cli/lightning_cli_advanced_3.html#models-with-multiple-submodules) to initialize (and type check!) each underlying model component. The model design and the config file both...

## Story 17: Registering multi-position acquisition 1. User collects 3D volumes in multiple channels at multiple positions in a 96-well plate. 2. User analyzes the volumes to register the volumes...

Thanks for summarizing your attempts @Christianfoley. 👋🏼 @pattonw @sheridana - any tips on how to handle our data? Some notes for context: * We are upgrading our image translation/virtual staining...

Hello @funkey , @pattonw! We have been using gunpowder for augmentation during virtual staining with data stored in ome-zarr's HCS format. Here is the [key method](https://github.com/mehta-lab/microDL/blob/053ad3f940c53659dd1aeec796ff912311cfddce/micro_dl/utils/gunpowder_utils.py#L145). We are at a...

We do need a unified reader module for the diverse datasets we have, which may or may not be a universal entry point. We have accumulated significant technical debt in...

Hi @ziw-liu, How about we divide this issue in two - one to unify the API for zarr formats and the other to unify API for tiff formats? This specific...

> My understanding was that we intentionally do not support writing NGFF v0.1 to promote transition to the new version. Is that still the plan? Yes! I am thinking that...

I added my input re: TIFF format to #53. I think we should release iohub 0.1.0 once we have unified the API for ome-zarr format and have converters in place.

@ziw-liu this CLI is now needed by multiple team members. Can you please take up #186?