How to deploy a pathology pytorch model on MONAI label apps? #1165
Replies: 2 comments 2 replies
-
Hi @OmarAshkar , yes, you could start creating a new model or customized the pathology app by getting familiar with the current template app. Pretrained model is set here: Finally, adjust the integration in the Thanks. |
Beta Was this translation helpful? Give feedback.
-
Would be super helpful to have a step-by-step-video instruction on how to implement own models. Still not getting it 100%. My focus is currently on the endoscopy sample app since my use case is similar. I am trying to follow the steps in the documentation (https://github.com/Project-MONAI/MONAILabel/tree/main/sample-apps/endoscopy#how-to-add-new-model) but I don't really understand the strucuture of the MONAI Label App and how the internal workflow of inference looks like. Additionally, the referenced endoscopy/lib/trainers/tooltracking.py does not hold much information that could be used as a template. |
Beta Was this translation helpful? Give feedback.
-
Hello,
I am wondering how may I customize the pathology sample apps with my own model, so I can interact with the model more and fix it if needed.
My best guess is the start will be where the pretrained model is loaded in monai label.
Thanks for your great efforts!
Omar
Beta Was this translation helpful? Give feedback.
All reactions