vugia truong

@m00nLi What made you so mad? From your image, I can see some improvements in inference speed, model size, flops.  

@m00nLi Yes I am. I clean up their report numbers a little bit. I only take the inference speed of Speed of (V100, fp32, batch=32). The input size is set...

> Let me show you: "YOLOv6-n": 11.1 GLOPs, YOLOv5n: 4.5 GFLOPs "YOLOv6-tiny": 36.7 GLOPs, YOLOv5s: 16.5 GFLOPs "YOLOv6-s": 44.2GLOPs, YOLOv5m: 49 GFLOPs Is that clear enough? That is clear for...

@MingChaoXu Your loss value is still very high. With those loss value, the accuracy should be equal 0 I think your learning rate is too large. By default the learning...

@MingChaoXu Would you mind add more detail? At least some training log and how to extract feature.

@MingChaoXu So in your case the feature is mean the "f-f----------------f---" and the result is "fff". Actually the "f-f--------f--" (26 characters) is not the feature, it is the raw prediction...

@ccnankai Hello, i am not chinese, so i guess you want to create train and test lmdb file. I the the tool available at https://github.com/bgshih/crnn/blob/master/tool/create_dataset.py . In summary, the script...

@maichm Can you post same of your train data ?

@maichm You image and label (imgpath.split("_")[-2] ) is correct so your dataset should be correct. Could you post your training log ? and training parameter too ?

@maichm Your loss is too high, so it mean that the model haven't converged yet. parser.add_argument('--lr', type=float, default=0.01, help='learning rate for Critic, default=0.00005') The default lr is very high, reduce...