我的cupy和chainer版本如下
Chainer:4.1.0 NumPy:1.15.4 CuPy:CuPy 版本:4.1.0
CUDA 根目录:/usr/local/cuda-9.0 CUDA 构建版本:9000 CUDA 驱动程序版本:9020 CUDA 运行时版本:9000
cuDNN 构建版本:7104 cuDNN版本:7104 NCCL 构建版本:2104
我正在尝试按照此链接运行测试脚本,我使用的脚本如下
python image_sheeping.py figure_skating/models/resnet_50_augmentation_no_noise_75_100/Resnet50SheepLocalizer_97305.npz \log -i figure_skating/evaluation_dataset/test_images/22.png -g 0 -o validation_data/images/analyzed
我收到的错误消息是
回溯(最后一次调用):文件“image_sheeping.py”,第 50 行,在 bboxes 中,分数 = localizer.localize(processed_image)[:2] 文件“/home/rahul/Desktop/Thesis/code/loans/loans/ sheep/unsupervised_sheep_localizer.py”,第 43 行,在本地化 bbox、rois、scores、visual_backprop = self.model.predict([processed_image], return_visual_backprop=return_visual_backprop) 文件“/home/rahul/Desktop/Thesis/code/loans/loans /figure_skating/models/resnet_50_augmentation_no_noise_75_100/localizer.py”,第 102 行,在预测 rois 中,bboxes = self(images) 文件“/home/rahul/Desktop/Thesis/code/loans/loans/figure_skating/models/resnet_50_augmentation_no_noise_75_100/localizer。 py",第 144 行,通话中 h = self.feature_extractor(input_images, layers=['res5', 'pool5']) 文件 "/home/rahul/.virtualenvs/loans/lib/python3.6/site-packages/chainer/links/model/vision/ resnet.py”,第 198 行,调用 h = func(h) 文件“/home/rahul/.virtualenvs/loans/lib/python3.6/site-packages/chainer/links/connection/convolution_2d.py”,行175,待命 groups=self.groups)文件“/home/rahul/.virtualenvs/loans/lib/python3.6/site-packages/chainer/functions/connection/convolution_2d.py”,第 582 行,convolution_2d y,= fnode.apply (args) 文件“/home/rahul/.virtualenvs/loans/lib/python3.6/site-packages/chainer/function_node.py”,第 258 行,应用输出 = self.forward(in_data) 文件“/home/ rahul/.virtualenvs/loans/lib/python3.6/site-packages/chainer/function_node.py”,第 367 行,向前返回 self.forward_gpu(输入) 文件“/home/rahul/.virtualenvs/loans/lib/python3.6/site-packages/chainer/functions/connection/convolution_2d.py”,第 161 行,在 forward_gpu 中返回 self._forward_cudnn(x, W, b, y ) 文件“/home/rahul/.virtualenvs/loans/lib/python3.6/site-packages/chainer/functions/connection/convolution_2d.py”,第 234 行,在 _forward_cudnn auto_tune=auto_tune, tensor_core=tensor_core) 文件“cupy /cudnn.pyx”,第 598 行,在 cupy.cudnn.convolution_forward 文件“cupy/cudnn.pyx”,第 33 行,在 cupy.cudnn.get_handle 文件“cupy/cuda/cudnn.pyx”,第 473 行,在 cupy 中。 cuda.cudnn.create 文件“cupy/cuda/cudnn.pyx”,第 446 行,在 cupy.cuda.cudnn.check_status cupy.cuda.cudnn.CuDNNError: CUDNN_STATUS_INTERNAL_ERROR 异常被忽略:del of 0%| | 0/1 [00:17 Traceback(最近一次通话最后一次):文件“/home/rahul/.virtualenvs/loans/lib/python3.6/site-packages/tqdm/_tqdm.py”,第 931 行,在del self .close() 文件“/home/rahul/.virtualenvs/loans/lib/python3.6/site-packages/tqdm/_tqdm.py”,第 1133 行,关闭 self._decr_instances(self) 文件“/home/rahul /.virtualenvs/loans/lib/python3.6/site-packages/tqdm/_tqdm.py”,第 496 行,在 _decr_instances cls.monitor.exit() 文件中“/home/rahul/.virtualenvs/loans/lib/python3 .6/site-packages/tqdm/_monitor.py”,第 52 行,退出 self.join() 文件“/usr/lib/python3.6/threading.py”,第 1053 行,加入引发 RuntimeError("
谁能帮我解决错误?