Max Lu

thank you! ill give it a try today.

@michalk8 hi, sorry I am not that familiar with dask, but after using dask_image.imread to read in my image I see that it only has one chunk: Array | Chunk...

Hi @michalk8, thanks a lot for the response! I have revised my workflow based on your suggestion: ```python # initialize cluster cluster = LocalCluster( n_workers=4, threads_per_worker=2, memory_limit='8GB' ) client =...

@michalk8 thanks! this is very helpful information! If I understand correctly, calling "compute()" on the ImageContainer loads the underlying image layers into persistent memory? Indeed I am able to run...

@michalk8 thanks so much for the explanation! So for starters I don't come from a software engineering background so I am generally not capable of identifying the most efficient implementation....

I am not familiar with tifffile - does that handle all common wsi formats such as .svs, .ndpi, etc.?

you can maybe try commenting out the if block to see if you can proceed without issue: https://github.com/mahmoodlab/CLAM/blob/1a92ef234411b44ec9ba27551307aea1143b5b4e/create_patches_fp.py#L174

> * `targets_as_params` - if you want to use some targets (arguments that you pass when call the augmentation pipeline) to produce some augmentation parameters on aug call, you need...

To give some context - this is the transform i want to implement (i want to pad each image up to the nearest multiple of "size_divisible", e.g. 32) - which...

> Have you tried [PadIfNeeded](https://albumentations.ai/docs/api_reference/augmentations/geometric/transforms/#albumentations.augmentations.geometric.transforms.PadIfNeeded) with `pad_height_divisor=32, pad_width_divisor=32`? > > Example: > > ```python > import numpy as np > import albumentations as A > > t = A.Compose([A.PadIfNeeded(pad_height_divisor=32, pad_width_divisor=32,...