caffe-windows10 安装问题 Check failed: error == cudaSuccess (8 vs.0) invalid device function
2016-10-16 03:31
615 查看
按【2】中的windows10 gpu配置后 出现了 caffe.exe has stopped working的问题，发现实际是cuda architecture与gpu不匹配【1】，所以加对应cuda architecure在下图所示未知
然后这个问题就解决了。。花了我两个下午。。。。
reference
【1】 https://github.com/rbgirshick/py-faster-rcnn/issues/118
【2】http://blog.csdn.net/zb1165048017/article/details/51549105
然后这个问题就解决了。。花了我两个下午。。。。
reference
【1】 https://github.com/rbgirshick/py-faster-rcnn/issues/118
【2】http://blog.csdn.net/zb1165048017/article/details/51549105
内容来自用户分享和网络整理，不保证内容的准确性，如有侵权内容，可联系管理员处理 
相关文章推荐
- 配置SSD-caffe测试时出现“Check failed: error == cudaSuccess (10 vs. 0) invalid device ordinal”解决
- Check failed: error == cudaSuccess (8 vs. 0) invalid device function
- Check failed: error == cudaSuccess (8 vs. 0) invalid device function
- Check failed: error == cudaSuccess (8 vs. 0) invalid device function
- Check failed: error == cudaSuccess (8 vs. 0) invalid device function
- 配置SSD-caffe测试时出现“Check failed: error == cudaSuccess (10 vs. 0) invalid device ordinal”解决
- caffe：Check failed: error == cudaSuccess (10 vs. 0) invalid device ordinal
- Check failed: error == cudaSuccess (8 vs. 0) invalid device function
- caffe运行错误： im2col.cu:61] Check failed: error == cudaSuccess (8 vs. 0) invalid device function
- Check failed: error == cudaSuccess (8 vs. 0) invalid device function
- Caffe 分类问题 Check failed: error == cudaSuccess (2 vs. 0) out of memory
- 运行caffe时报错：Check failed: error == cudaSuccess
- 【caffe】 Check failed: error == cudaSuccess (30 vs. 0) unknown error
- 【caffe跑试验遇到错误：Check failed: error == cudaSuccess (2 vs. 0) out of memory】
- Check failed: error == cudaSuccess (8 vs. 0) invalid device function
- Caffe | Check failed: error == cudaSuccess (2 vs. 0) out of memory
- Check failed: error == cudaSuccess (8 vs. 0) invalid device function
- 【caffe跑试验遇到错误：Check failed: error == cudaSuccess (2 vs. 0) out of memory】
- CUDA/caffe ERROR:cudaGetDeviceCount returned 30/35,Check failed: error == cudaSuccess (30/35 vs. 0)
- caffe训练过程中显示Check failed：error == cudaSuccess(2 vs. 0) out of memory