工程实践_LFFD模型测试bug解决方案汇总

问题一

Traceback (most recent call last):
  File "predict.py", line 265, in <module>
    run_prediction_pickle()
  File "predict.py", line 254, in run_prediction_pickle
    bboxes = my_predictor.predict(im, resize_scale=1, score_threshold=0.5, top_k=10000, NMS_threshold=0.6)
  File "predict.py", line 163, in predict
    outputs.append(output.asnumpy())
  File "/root/work/mxnet/python/mxnet/ndarray/ndarray.py", line 2566, in asnumpy
    ctypes.c_size_t(data.size)))
  File "/root/work/mxnet/python/mxnet/base.py", line 246, in check_call
    raise get_last_ffi_error()
mxnet.base.MXNetError: Traceback (most recent call last):
  File "../include/mshadow/././././cuda/tensor_gpu-inl.cuh", line 110
Name: Check failed: err == cudaSuccess (48 vs. 0) : MapPlanKernel ErrStr:no kernel image is available for execution on the device

遇到上述问题，可能是GPU硬件问题，比如Tesla T4就会碰到这个问题，而是用Tesla P4就不会。还有就是如果是启用镜像的时候记得分配内存参数和使用nvidia运行命令。

[10:16:47] ../src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
: cannot connect to X server

问题二

[10:16:47] ../src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:97: Running performance tests to find the best convolution algorithm, this can take a while... (set the environment variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
: cannot connect to X server

一般在使用远程服务器时会遇到这个问题，因为远程服务器没办法提供图形用户界面的服务，而X server是Linux系统上提供图形用户界面的服务程序。

我们要将代码中的cv2.imshow函数注释掉，这个函数在windows下可以直接调出窗口显示图片，在命令行下使用Linux就没法显示了，注意不要这样用即可。

还有cv.waitKey函数也要注释掉。