Muhammad Faizan
Muhammad Faizan
Thanks Glenn, I have read the paper. However, there is no such thing they mentioned for joint training for classification and detection in their repository implementation. Can't we play with...
Yes, I have trained detectors on male and female head classes, which gave reasonably good results.
@aezakmi99 you can use yolov7 for localizing objects, so yes if you trained it on digits or alphabets dataset. It will be detect them. You can then crop those digits...
Here is a pipe line you can try: digits detector->digits classifier->label-text-from-prediction
maybe you can train detector on a large digits dataset to detect and label digits. As for as YOLOv7, you can just use it to detect an object for instance...
please reduce batch size it seems like you don't have an enough GPU memory.
Please post link to the notebook and provide more information about the error
@XinyiYing `self.length` is defined for dimension 2 and 1, however, the dimensions for 'hwd' is 3. I don't understand that.
@XinyiYing thank you so much for the detailed explanation
Thank you so much. I have one more question. I have built the engine file with half precision and I am getting about 8.5 FPS on Jetson Orin Nano. How...