在自定义对象检测上使用tensorflow-gpu时出错

Question

我正在尝试使用tensorflow-gpu进行自定义对象检测。我已经创建了conda环境，并使用我的tensorflow-gpu版本1.14下载并安装了CUDA和CuDNN 10.0。但是，我收到此错误

ResourceExhaustedError: 2 root error(s) found.
  (0) Resource exhausted: OOM when allocating tensor with shape[32,1792,1792,3,2] and type float on /job:localhost/replica:0/task:0/device:CPU:0 by allocator gpu_host_bfc
     [[{{node Tile_5}}]]
Hint: If you want to see a list of allocated tensors when OOM happens, add report_tensor_allocations_upon_oom to RunOptions for current allocation info.

  (1) Resource exhausted: OOM when allocating tensor with shape[32,1792,1792,3,2] and type float on /job:localhost/replica:0/task:0/device:CPU:0 by allocator gpu_host_bfc
     [[{{node Tile_5}}]]
Hint: If you want to see a list of allocated tensors when OOM happens, add report_tensor_allocations_upon_oom to RunOptions for current allocation info.

     [[training/Adam/cond_175/Switch_2/_6707]]
Hint: If you want to see a list of allocated tensors when OOM happens, add report_tensor_allocations_upon_oom to RunOptions for current allocation info.

0 successful operations.
0 derived errors ignored.

通过关注此网站Train Object Detection AI with 6 lines of code。