Muhammed Sezer comments

Results 6 comments of


                                            Muhammed Sezer

What about a fake training model?

Hi thanks for answer. I could not express my idea. I just want to see all train batches in exp/train folder to see training images without training. If there is...

What about a fake training model?

Yes, I made it first 150 but I need all of them, and If I make it 150 I cannot git pull since it will be zeroed. It just my...

Feature request - ROS2

Yes ROS2 support would be awesome, Thanks for repo

face_recognition performance on Raspberry PI and Jetson Nano

Hi, Did you solve this problem? I am using yolov5n model for inference. its much faster (0.06) seconds And for face recognition, I extract one random ID per Loop. I...

Nan loss during first epoch in custom dataset training

I also have this problem. I tried to decrease LR to very very low numbers or using Adam it did not change. But only cls_loss becomes inf, others fall as...

Roboflow for Datasets, Labeling, and Active Learning 🌟

> @Jacobsolawetz pinging you again about the broken links above it seems you need to ping them again