tdf1995
tdf1995
![image](https://user-images.githubusercontent.com/39071957/131487789-5a97711f-25b7-4117-9693-0ab74b3012de.png) ![image](https://user-images.githubusercontent.com/39071957/131488008-e6454839-8999-44e2-95fb-796d77349791.png) ![image](https://user-images.githubusercontent.com/39071957/131488058-57c23c61-6e42-4e67-8a7d-c4f4358dded2.png) ![image](https://user-images.githubusercontent.com/39071957/131488104-13bac1ae-dd79-4f67-bc99-8d0b7482e1c5.png) the white box is predicted by yolo,color box is predicted by Kalman filter
![image](https://user-images.githubusercontent.com/39071957/131487789-5a97711f-25b7-4117-9693-0ab74b3012de.png) ![image](https://user-images.githubusercontent.com/39071957/131488008-e6454839-8999-44e2-95fb-796d77349791.png) ![image](https://user-images.githubusercontent.com/39071957/131488058-57c23c61-6e42-4e67-8a7d-c4f4358dded2.png) ![image](https://user-images.githubusercontent.com/39071957/131488104-13bac1ae-dd79-4f67-bc99-8d0b7482e1c5.png) the white box is predicted by yolo,color box is predicted by Kalman filter
![3](https://user-images.githubusercontent.com/39071957/54654704-05bc0b80-4afa-11e9-8feb-0b5ade617b7a.jpg) 我想生成上图背景的白字图像,我在设置reverse_color后发现生成的图像背景与想要的背景还是不一样 ![00000200](https://user-images.githubusercontent.com/39071957/54654747-313ef600-4afa-11e9-81fb-15cfe367642f.jpg) 请问要怎么设置
train_set: ![image](https://user-images.githubusercontent.com/39071957/174749833-f02645a7-46ba-4e2d-82c2-aeda99dfeec9.png) generated images: ![image](https://user-images.githubusercontent.com/39071957/174749451-b533a298-7fa7-4f12-8e42-038acf91e339.png) @click.option('--outdir',default='/work/ai_lab/miner/data-efficient-gans/output', help='Where to save the results', required=True, metavar='DIR') @click.option('--gpus',default=2, help='Number of GPUs to use [default: 1]', type=int, metavar='INT') @click.option('--snap',default=10, help='Snapshot interval [default: 50 ticks]', type=int,...
多类别训练问题
无论我训练camvid还是自己的数据集,总是只找到两类图像,以下是训练camvid时的信息: Epoch 1/30 Found 367 images belonging to 1 classes. Found 1402 images belonging to 6 classes. Found 367 images belonging to 1 classes. Found 1402 images belonging to 6...
i changed the dafault parser.add_argument( "--restore_path", help="If not None, resume training of a given " "checkpoint (mode 'train').", default="resources/triplet-20200109T013728Z-001/triplet/mars.ckpt") and got ValueError: Checkpoint is missing variable [global_step]
### System Info python3.9,flagai1.7.3 ### Information - [ ] The official example scripts - [ ] My own modified scripts ### Tasks - [ ] An officially supported task in...
WARNING: The argument `dynamic_input_shape=True` is not needed any more, onnxsim can now support dynamic input shapes natively, please refer to the latest documentation. An error will be raised in the...