TensorFlow GPU : is cudnn optional?无法打开CUDA库libcudnn.so

Question

我安装了tensorflow-0.8.0 GPU版本，tensorflow-0.8.0-cp27-none-linux_x86_64.whl。它说它需要CUDA工具包7.5和CuDNN v4。

# Ubuntu/Linux 64-bit, GPU enabled. Requires CUDA toolkit 7.5 and CuDNN v4.  For
# other versions, see "Install from sources" below.

但是，我不小心忘记安装CuDNN v4，但是除了错误消息“无法打开CUDA库libcudnn.so”之外，它还可以正常工作。但是它起作用了，并说:“正在创建TensorFlow设备(/gpu:0)”。

没有CuDNN的消息

I tensorflow/stream_executor/dso_loader.cc:105] successfully opened CUDA library libcublas.so locally
I tensorflow/stream_executor/dso_loader.cc:99] Couldn't open CUDA library libcudnn.so. LD_LIBRARY_PATH: /usr/local/cuda/lib64:
I tensorflow/stream_executor/cuda/cuda_dnn.cc:1562] Unable to load cuDNN DSO
I tensorflow/stream_executor/dso_loader.cc:105] successfully opened CUDA library libcufft.so locally
I tensorflow/stream_executor/dso_loader.cc:105] successfully opened CUDA library libcuda.so.1 locally
I tensorflow/stream_executor/dso_loader.cc:105] successfully opened CUDA library libcurand.so locally
('Extracting', 'MNIST_data/train-images-idx3-ubyte.gz')
/usr/lib/python2.7/gzip.py:268: VisibleDeprecationWarning: converting an array with ndim > 0 to an index will result in an error in the future
  chunk = self.extrabuf[offset: offset + size]
/home/ubuntu/TensorFlow-Tutorials/input_data.py:42: VisibleDeprecationWarning: converting an array with ndim > 0 to an index will result in an error in the future
  data = data.reshape(num_images, rows, cols, 1)
('Extracting', 'MNIST_data/train-labels-idx1-ubyte.gz')
('Extracting', 'MNIST_data/t10k-images-idx3-ubyte.gz')
('Extracting', 'MNIST_data/t10k-labels-idx1-ubyte.gz')
I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:900] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
I tensorflow/core/common_runtime/gpu/gpu_init.cc:102] Found device 0 with properties:
name: GRID K520
major: 3 minor: 0 memoryClockRate (GHz) 0.797
pciBusID 0000:00:03.0
Total memory: 4.00GiB
Free memory: 3.95GiB
I tensorflow/core/common_runtime/gpu/gpu_init.cc:126] DMA: 0
I tensorflow/core/common_runtime/gpu/gpu_init.cc:136] 0:   Y
I tensorflow/core/common_runtime/gpu/gpu_device.cc:755] Creating TensorFlow device (/gpu:0) -> (device: 0, name: GRID K520, pci bus id: 0000:00:03.0)
I tensorflow/core/common_runtime/gpu/pool_allocator.cc:244] PoolAllocator: After 1704 get requests, put_count=1321 evicted_count=1000 eviction_rate=0.757002 and unsatisfied allocation rate=0.870305
I tensorflow/core/common_runtime/gpu/pool_allocator.cc:256] Raising pool_size_limit_ from 100 to 110
I tensorflow/core/common_runtime/gpu/pool_allocator.cc:244] PoolAllocator: After 1704 get requests, put_count=1812 evicted_count=1000 eviction_rate=0.551876 and unsatisfied allocation rate=0.536972
I tensorflow/core/common_runtime/gpu/pool_allocator.cc:256] Raising pool_size_limit_ from 256 to 281

后来，我安装了CuDNN，但看不到差异。

消息与CuDNN

I tensorflow/stream_executor/dso_loader.cc:105] successfully opened CUDA library libcublas.so locally
I tensorflow/stream_executor/dso_loader.cc:105] successfully opened CUDA library libcudnn.so locally
I tensorflow/stream_executor/dso_loader.cc:105] successfully opened CUDA library libcufft.so locally
I tensorflow/stream_executor/dso_loader.cc:105] successfully opened CUDA library libcuda.so.1 locally
I tensorflow/stream_executor/dso_loader.cc:105] successfully opened CUDA library libcurand.so locally
('Extracting', 'MNIST_data/train-images-idx3-ubyte.gz')
/usr/lib/python2.7/gzip.py:268: VisibleDeprecationWarning: converting an array with ndim > 0 to an index will result in an error in the future
  chunk = self.extrabuf[offset: offset + size]
/home/ubuntu/TensorFlow-Tutorials/input_data.py:42: VisibleDeprecationWarning: converting an array with ndim > 0 to an index will result in an error in the future
  data = data.reshape(num_images, rows, cols, 1)
('Extracting', 'MNIST_data/train-labels-idx1-ubyte.gz')
('Extracting', 'MNIST_data/t10k-images-idx3-ubyte.gz')
('Extracting', 'MNIST_data/t10k-labels-idx1-ubyte.gz')
I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:900] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
I tensorflow/core/common_runtime/gpu/gpu_init.cc:102] Found device 0 with properties:
name: GRID K520
major: 3 minor: 0 memoryClockRate (GHz) 0.797
pciBusID 0000:00:03.0
Total memory: 4.00GiB
Free memory: 3.95GiB
I tensorflow/core/common_runtime/gpu/gpu_init.cc:126] DMA: 0
I tensorflow/core/common_runtime/gpu/gpu_init.cc:136] 0:   Y
I tensorflow/core/common_runtime/gpu/gpu_device.cc:755] Creating TensorFlow device (/gpu:0) -> (device: 0, name: GRID K520, pci bus id: 0000:00:03.0)
I tensorflow/core/common_runtime/gpu/pool_allocator.cc:244] PoolAllocator: After 1704 get requests, put_count=1321 evicted_count=1000 eviction_rate=0.757002 and unsatisfied allocation rate=0.870305
I tensorflow/core/common_runtime/gpu/pool_allocator.cc:256] Raising pool_size_limit_ from 100 to 110
I tensorflow/core/common_runtime/gpu/pool_allocator.cc:244] PoolAllocator: After 1704 get requests, put_count=1811 evicted_count=1000 eviction_rate=0.552181 and unsatisfied allocation rate=0.537559
I tensorflow/core/common_runtime/gpu/pool_allocator.cc:256] Raising pool_size_limit_ from 256 to 281

那么，使用/不使用CuDNN的区别是什么？

Answer 1

cuDNN用于加速一些TensorFlow操作，例如卷积。我在您的日志文件中注意到您正在针对MNIST数据集进行训练。 TensorFlow随附的引用MNIST模型围绕2个完全连接的层和一个softmax构建。因此，在训练此模型时，TensorFlow不会尝试调用cuDNN。

我不确定cuDNN不可用时TensorFlow是否会自动回退到较慢的卷积算法。如果不是，您可以通过在运行TensorFlow之前将TF_USE_CUDNN环境变量设置为0来始终禁用cuDNN。