New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Make run_graph() can be run in training mode #1065
base: master
Are you sure you want to change the base?
Conversation
The variable "inputs" is declared but not used in run_graph().
1. The argument become inputs and outputs 2. It can now run in training mode(useful in debugging), but only works when batch size equals to 1.
Update: Make run_graph() can be run in training mode
|
I found a place in inspect_nucleus_model.ipynb that uses the image_metas param. Which would break with this update. I wonder if we can change that notebook to pass the original image, and if so, then image_meta param is not needed anymore and can be removed to simplify. I'm hoping to get a bit of time to look into that in the next few days. |
@waleedka Yes, I noticed that point. I just want to make sure this PR will work and then start to apply changes to other notebooks. Maybe I should add a WIP flag. |
@waleedka There is only one notebook use the argument "image_metas", which is inspect_nucleus_model.ipynb. I have already made it compatible with the new run_graph. |
The variable
inputs
is declared but not used inrun_graph()
.