Open ManuBN786 opened 2 years ago
I used the mask_rcnn_swin-t-p4-w7_fpn_fp16_ms-crop-3x_coco.py
and i get this error: RuntimeError: Expected object of scalar type Float but got scalar type Half for argument #2 'mat2' in call to _th_bmm
my system configurations are:
sys.platform: linux Python: 3.8.9 (default, Apr 3 2021, 01:02:10) [GCC 5.4.0 20160609] CUDA available: True GPU 0: TITAN Xp GPU 1: GeForce GTX 1080 Ti CUDA_HOME: /usr/local/cuda-9.2 NVCC: Cuda compilation tools, release 9.2, V9.2.148 GCC: gcc (Ubuntu 5.4.0-6ubuntu1~16.04.12) 5.4.0 20160609 PyTorch: 1.5.1+cu92 PyTorch compiling details: PyTorch built with:
TorchVision: 0.6.1+cu92 OpenCV: 4.5.5 MMCV: 1.4.0 MMCV Compiler: GCC 5.4 MMCV CUDA Compiler: 9.2 MMDetection: 2.21.0+ff9bc39 .4.0 MMCV CompÁë
Need the full error report. Please follow the issue template to provide more information.
I used the mask_rcnn_swin-t-p4-w7_fpn_fp16_ms-crop-3x_coco.py
and i get this error: RuntimeError: Expected object of scalar type Float but got scalar type Half for argument #2 'mat2' in call to _th_bmm
my system configurations are:
2022-02-17 12:53:24,806 - mmdet - INFO - Environment info:
sys.platform: linux Python: 3.8.9 (default, Apr 3 2021, 01:02:10) [GCC 5.4.0 20160609] CUDA available: True GPU 0: TITAN Xp GPU 1: GeForce GTX 1080 Ti CUDA_HOME: /usr/local/cuda-9.2 NVCC: Cuda compilation tools, release 9.2, V9.2.148 GCC: gcc (Ubuntu 5.4.0-6ubuntu1~16.04.12) 5.4.0 20160609 PyTorch: 1.5.1+cu92 PyTorch compiling details: PyTorch built with:
TorchVision: 0.6.1+cu92 OpenCV: 4.5.5 MMCV: 1.4.0 MMCV Compiler: GCC 5.4 MMCV CUDA Compiler: 9.2 MMDetection: 2.21.0+ff9bc39 .4.0 MMCV CompÁë