Great job！I meet a problem is that I wanna to use it for muti-object detection, but It looks like cannot location them. #7
Description
Thank you for the great job! Now I meet some problems. The PNG contain some Cars and Truck, But the model can not output the location right,It seems like random output ,how could I sovle this? I think maybe I need to train the net on my dataset, but I don't have the train.py, could you heip me ? Thank you very much!
Warning: Flash attention is not available, using eager attention instead.
Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:05<00:00, 2.78s/it]
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
User:
Please describe the image shortly.
Assistant: The image shows wide open highway flanked by green trees on the left and green grass on the right. The sky is blue with clouds. There are several lines at the bottom, possibly indicating lane dividers on the road. On the right side of the image are a few modern buildings, among them a tall tower with electronic wires attached to it. There are poles with blue and white signs on the side of the road, and a few cars are driving down the road.
User:
There may be people, car, bus, lamp, motorcycle, and truck in this picture. Each type may also contain multiple detection targets. If there are corresponding categories, please detect all the targets and provide their precise positions in the image.For each object, output exactly ONE line in this format: [Category: xmin, ymin, xmax, ymax].
Assistant: [6.0, 1007, 16.0, 1000]
[2.0, 700, 16.0, 1000]
[8.0, 1310, 1600, 1000]
[0.0, 680, 1600, 1000]