PanYang

当我更换了自己定义的一个数据集时，出现了上述bug，以下是我的参数， 2022-03-02 22:06:41,743 - mmdet - INFO - Distributed training: False 2022-03-02 22:06:41,744 - mmdet - INFO - MMDetection Version: 1.0.0+unknown 2022-03-02 22:06:41,744 - mmdet - INFO - Config: #...

I used my own dataset to train the instance segmentation task on K-Net and found that the seg mAP was always 0, and the number of training epochs still did...

RuntimeError: input image is smaller than kernel (shape_check at mmdet/ops/dcn/src/deform_conv_cuda.cpp:127)

6

当我训练您参数文件时，环境配置几乎与您的一样，但发现在deform conv卷积操作会发生报错，以下是我的环境配置 sys.platform: linux Python: 3.7.11 (default, Jul 27 2021, 14:32:16) [GCC 7.5.0] CUDA available: True GPU 0: NVIDIA TITAN RTX CUDA_HOME: /usr/local/cuda-10.1 NVCC: Cuda compilation tools, release 10.1, V10.1.105...