[error] Could not create cudnn handle: CUDNN_STATUS_INTERNAL_ERROR

[zg9uagfv]

os:ubuntu18.04
cuda:10.0
cudnn:7.6.5

run command:
python3 test_segmentation_deeplab.py -i sample_data/input

got some error as below:

2020-10-10 09:49:56.226141: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:1005] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero 2020-10-10 09:49:56.226345: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1326] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 4919 MB memory) -> physical GPU (device: 0, name: GeForce GTX 1660 Ti, pci bus id: 0000:01:00.0, compute capability: 7.5) model loaded successfully! 2020-10-10 09:49:57.495158: I tensorflow/stream_executor/platform/default/dso_loader.cc:42] Successfully opened dynamic library libcudnn.so.7 2020-10-10 09:49:57.850118: E tensorflow/stream_executor/cuda/cuda_dnn.cc:329] Could not create cudnn handle: CUDNN_STATUS_INTERNAL_ERROR 2020-10-10 09:49:57.853908: E tensorflow/stream_executor/cuda/cuda_dnn.cc:329] Could not create cudnn handle: CUDNN_STATUS_INTERNAL_ERROR

[cxgincsu]

add this code in test_segmentation_deeplab.py. It works for me

############# fix cudnn error : CUDNN_STATUS_INTERNAL_ERROR ##############
from tensorflow.compat.v1 import ConfigProto
from tensorflow.compat.v1 import InteractiveSession
config = ConfigProto()
config.gpu_options.allow_growth = True
session = InteractiveSession(config=config)
##########################################################################