Ishaan
Ishaan
This error is due to the code snippet of ```mmseg/models/data_preprocessor.py``` If you set size_divisor = None there, error will be resolved
yes I have done it On Mon, Jul 8, 2024 at 2:40 PM 肖遥 ***@***.***> wrote: > I have the same issue,have you solved it? > > — > Reply...
> I have the same issue,have you solved it? Yes I have resolved the issue. It was mainly related to the incompatibility of torch and torchvision in the newer updates...
Thanks @SkalskiP for pointing out the need to verify that change. I forgot to add the verification to it. I created a sample label file to notice how polygon's coordinates...
hello @SkalskiP please review the changes once you have time, thanks!
Yes @SkalskiP, it would be absolutely great if I am able to contribute towards maestro. Will go through the code base, develop the QLoRA approach for Florence-2 and then raise...
@SkalskiP, people would love to hop on a Roboflow's discord server in general. It would be great to discuss the fine-tunings and approaches using Maestro, people would be definitely interested!
@SkalskiP sure, I will go through it before starting to contribute. Will keep you in touch with all the changes that I come up with
saving the best model along with the optimizer state and parameter schedulers typically involves configuring the training pipeline and checkpointing mechanisms. suppose this is your sample training script which saves...
the main thing you can do is introduce dice loss along with IoU metric and then optimize your model based on results of both the losses.