Error in slide-graph inference pipeline collab, when using pregenerate_grap=False

[cupcakes04]

• TIA Toolbox version: 6.1
• Python version: 3.10.15
• Operating System: Windows 11 -- WSL 2

Description

(RuntimeError: dictionary changed size during iteration)
this multi-processing step error occurred when running the below notebook, but not using pre-generated graphs

https://colab.research.google.com/github/TissueImageAnalytics/tiatoolbox/blob/develop/examples/inference-pipelines/slide-graph.ipynb

What I Did

inst_segmentor = NucleusInstanceSegmentor(
        pretrained_model="hovernet_fast-pannuke",
        batch_size=16,
        num_postproc_workers=4,
        num_loader_workers=4,
    )

i commented the workers, and it works fine, but slower.

i have identified the issue to be the merging process of workers at models/engine/nucleu_instance_segmentor.py

@NucleusInstanceSegmentor._merge_post_process_results

here's the file for the runtime error output:
bug.txt

thank you for your time and work on the toolbox!