search

constantinpape / torch-em

Deep-learning based semantic and instance segmentation for 3D Electron Microscopy and other bioimage analysis problems based on pytorch.
MIT License
69 stars 19 forks source link

request for more examples #225

Open PG-Gu opened 4 months ago

PG-Gu commented 4 months ago

Thank you for the effort building this repo. I'm wondering if it is possible to add more experiment examples such as some neural segmentaion scripts written in out-dated Python2.7 For example, the neural network training for mutex-watershed and multicut. It would not only provide more context for users of torch-em but also help maintain the previous algorithms' utility Any kind of feedback from you would be appreciated. :)

constantinpape commented 4 months ago

Hi @PG-Gu , there are several examples that would fit these applications in https://github.com/constantinpape/torch-em/tree/main/experiments/unet-segmentation/neuron-segmentation. For example, https://github.com/constantinpape/torch-em/blob/main/experiments/unet-segmentation/neuron-segmentation/cremi/train_affinities.py shows how to train a network to predict affinities on CREMI, or https://github.com/constantinpape/torch-em/blob/main/experiments/unet-segmentation/neuron-segmentation/isbi2012/train_affinities_3d.py how to do this on ISBI.