Robin Cole
Robin Cole
OK tried: ``` %load_ext tensorboard %tensorboard --logdir /content/deepstack-trainer/train-runs/ ``` But doesn't display results Possibly because the runs directory was empty when I ran tensorboard. Can it only be added after...
ok that was it
The part2 images are 42.94 GB on disk for 14,757 items. I guess 1 million images will be a gigantic dataset on disk
I had Tesla P100. Will try your suggestions. - decreasing batch_size from 4 to 2 and doubling (I assume?) ` accumulate_grad_batches=2 # default 1, increase to 2` results in immediate...
Increasing from 5 to `patience=20` I managed to get an additional epoch before the memory error
There are also many exotic metrics in https://piq.readthedocs.io/en/latest/overview.html
Tried deleting all the data under `SE320572405` but then I get the same error for the next item. Also deleting the `.iris` file doesn't clear it
added some debugging, it is trying to access a file not yet created: ``` DEBUG:root:Entering load_mask with image_id: SE320572405 DEBUG:root:Retrieved user_id from session: 1 ERROR:root:Exception encountered: [Errno 2] No such...
I restored an earlier version of the config, and now the app loads, although I still see that error in the logs
I can then ``` > iris demo Welcome to IRIS! No admin user was detected so please enter a new admin password. New admin password: ```