pytorch报错：RuntimeError: CUDA error: CUBLAS_STATUS_NOT_INITIALIZED when calling `cublasCreate(handle)

最新推荐文章于 2024-10-12 09:35:46 发布

蓝海渔夫

最新推荐文章于 2024-10-12 09:35:46 发布

阅读量672

点赞数 4

分类专栏： python及pytorch异常错误处理文章标签： pytorch 人工智能 python

本文链接：https://blog.csdn.net/wwimhere/article/details/136546927

版权

python及pytorch异常错误处理专栏收录该内容

5 篇文章 0 订阅

订阅专栏

用户在使用PyTorch进行图像分类时遇到CUDA错误，源于CUBLAS_STATUS_NOT_INITIALIZED。问题在于代码中num_classes设置为6，而实际CIFAR10数据集应有10个类别。

摘要由CSDN通过智能技术生成

pytorch做图像分类时报错如下：

RuntimeError: CUDA error: CUBLAS_STATUS_NOT_INITIALIZED when calling `cublasCreate(handle)`
../aten/src/ATen/native/cuda/Loss.cu:257: nll_loss_forward_reduce_cuda_kernel_2d: block: [0,0,0], thread: [0,0,0] Assertion `t >= 0 && t < n_classes` failed.
../aten/src/ATen/native/cuda/Loss.cu:257: nll_loss_forward_reduce_cuda_kernel_2d: block: [0,0,0], thread: [1,0,0] Assertion `t >= 0 && t < n_classes` failed.
../aten/src/ATen/native/cuda/Loss.cu:257: nll_loss_forward_reduce_cuda_kernel_2d: block: [0,0,0], thread: [2,0,0] Assertion `t >= 0 && t < n_classes` failed.
../aten/src/ATen/native/cuda/Loss.cu:257: nll_loss_forward_reduce_cuda_kernel_2d: block: [0,0,0], thread: [7,0,0] Assertion `t >= 0 && t < n_classes` failed.
../aten/src/ATen/native/cuda/Loss.cu:257: nll_loss_forward_reduce_cuda_kernel_2d: block: [0,0,0], thread: [8,0,0] Assertion `t >= 0 && t < n_classes` failed.
../aten/src/ATen/native/cuda/Loss.cu:257: nll_loss_forward_reduce_cuda_kernel_2d: block: [0,0,0], thread: [11,0,0] Assertion `t >= 0 && t < n_classes` failed.
../aten/src/ATen/native/cuda/Loss.cu:257: nll_loss_forward_reduce_cuda_kernel_2d: block: [0,0,0], thread: [12,0,0] Assertion `t >= 0 && t < n_classes` failed.
../aten/src/ATen/native/cuda/Loss.cu:257: nll_loss_forward_reduce_cuda_kernel_2d: block: [0,0,0], thread: [14,0,0] Assertion `t >= 0 && t < n_classes` failed.
../aten/src/ATen/native/cuda/Loss.cu:257: nll_loss_forward_reduce_cuda_kernel_2d: block: [0,0,0], thread: [15,0,0] Assertion `t >= 0 && t < n_classes` failed.
../aten/src/ATen/native/cuda/Loss.cu:257: nll_loss_forward_reduce_cuda_kernel_2d: block: [0,0,0], thread: [16,0,0] Assertion `t >= 0 && t < n_classes` failed.
../aten/src/ATen/native/cuda/Loss.cu:257: nll_loss_forward_reduce_cuda_kernel_2d: block: [0,0,0], thread: [19,0,0] Assertion `t >= 0 && t < n_classes` failed.
../aten/src/ATen/native/cuda/Loss.cu:257: nll_loss_forward_reduce_cuda_kernel_2d: block: [0,0,0], thread: [22,0,0] Assertion `t >= 0 && t < n_classes` failed.
../aten/src/ATen/native/cuda/Loss.cu:257: nll_loss_forward_reduce_cuda_kernel_2d: block: [0,0,0], thread: [23,0,0] Assertion `t >= 0 && t < n_classes` failed.
../aten/src/ATen/native/cuda/Loss.cu:257: nll_loss_forward_reduce_cuda_kernel_2d: block: [0,0,0], thread: [25,0,0] Assertion `t >= 0 && t < n_classes` failed.
../aten/src/ATen/native/cuda/Loss.cu:257: nll_loss_forward_reduce_cuda_kernel_2d: block: [0,0,0], thread: [31,0,0] Assertion `t >= 0 && t < n_classes` failed.

这个报错看似是CUDA的问题，其实并不是。

错误信息中的“Assertion `t >= 0 && t < n_classes` failed.”这句提示显示错误可能和number of classes有关。

检查代码发现，我的代码设置的分类类别错了。

我在跑cifar10类别时，创建网络部分把num_classes设置成了6

#创建网络
vit=ViT(
        image_size = 256,
        patch_size = 32,
        num_classes = 6,
        dim = 1024,
        depth = 6,
        heads = 16,
        mlp_dim = 2048,
        dropout = 0.1,
        emb_dropout = 0.1
    )