Prateek Gupta
Prateek Gupta
> I want to use darknet for yolov7 model that is trained with my custom data on this github code. However, I cannot find any way to convert the pytorch...
@zgFighting You should raise this issue in Wandb repo- [https://github.com/wandb/wandb/issues](url)
> Try installing (restart the kernel beforehead) the latest nightly version of SG that lifts up pytorch requirements, so it will use what is already installed on kaggle. I believe...
> Hi @dsbyprateekg thanks for reporting this! There seems to be some stability issues in the server backend which causes this. Are you able to reproduce this error consistently for...
@adamp87 I tried your PR with my custom tolov7-tiny model and found that difference in bbox position. With the original model bbox was tight (close around the object) but with...
@adamp87 onnx2tf script is failing for yolov7-w6 based model with image size of 1280, it is not showing any error in the console. `!onnx2tf -i model_opt.onnx -o model_opt.tf --verbosity info...
@danielhanchen In the shared kaggle notebook, the only changed line of code is the installation of few packages instead of using pip-  Also, I tried but no success. Getting...
@danielhanchen Infact I have used the exact copy of the shared notebook and result was same. It's not about time. I just save and commit the notebook with P100 and...
Unfortunately, I am still facing the same issue after I rerunning the kaggle notebook.
@danielhanchen thanks! I checked and found with T4x2 GPU, I can run successfully but for P100 GPU issue is still there. I can finally work further with T4x2 GPU :)