python - Tensorflow 不显示 "Successfully opened so & so CUDA libraries locally"

Question

我将 tensorflow 配置为在我的 GPU (GeForce 840M) 上使用 CUDA 支持，但与我之前使用的 CPU 相比，这些程序运行起来相当慢。此外，当我运行该程序时，我没有收到任何类型的消息，表明某某 CUDA 库已成功打开。相反，这是我在运行任何 tensorflow 程序时在日志中得到的:

python Neuralnet.py 
Successfully downloaded train-images-idx3-ubyte.gz 9912422 bytes.
Extracting /tmp/data/train-images-idx3-ubyte.gz
Successfully downloaded train-labels-idx1-ubyte.gz 28881 bytes.
Extracting /tmp/data/train-labels-idx1-ubyte.gz
Successfully downloaded t10k-images-idx3-ubyte.gz 1648877 bytes.
Extracting /tmp/data/t10k-images-idx3-ubyte.gz
Successfully downloaded t10k-labels-idx1-ubyte.gz 4542 bytes.
Extracting /tmp/data/t10k-labels-idx1-ubyte.gz
2017-03-28 07:53:57.979382: W tensorflow/core/platform/cpu_feature_guard.cc:45]
    The TensorFlow library wasn't compiled to use SSE4.1 instructions, 
    but these are available on your machine and could speed up CPU computations.
2017-03-28 07:53:57.979413: W tensorflow/core/platform/cpu_feature_guard.cc:45]
    The TensorFlow library wasn't compiled to use SSE4.2 instructions, 
    but these are available on your machine and could speed up CPU computations.
2017-03-28 07:53:57.979431: W tensorflow/core/platform/cpu_feature_guard.cc:45] 
    The TensorFlow library wasn't compiled to use AVX instructions, 
    but these are available on your machine and could speed up CPU computations.
2017-03-28 07:53:57.979438: W tensorflow/core/platform/cpu_feature_guard.cc:45] 
    The TensorFlow library wasn't compiled to use AVX2 instructions, 
    but these are available on your machine and could speed up CPU computations.
2017-03-28 07:53:57.979447: W tensorflow/core/platform/cpu_feature_guard.cc:45] 
    The TensorFlow library wasn't compiled to use FMA instructions, 
    but these are available on your machine and could speed up CPU computations.
2017-03-28 07:53:58.233876: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:901] 
    successful NUMA node read from SysFS had negative value (-1),
    but there must be at least one NUMA node, so returning NUMA node zero
2017-03-28 07:53:58.234333: I tensorflow/core/common_runtime/gpu/gpu_device.cc:887] 
Found device 0 with properties: 
name: GeForce 840M
major: 5 minor: 0 memoryClockRate (GHz) 1.124
pciBusID 0000:08:00.0
Total memory: 1.96GiB
Free memory: 1.75GiB
2017-03-28 07:53:58.234362: I tensorflow/core/common_runtime/gpu/gpu_device.cc:908] DMA: 0 
2017-03-28 07:53:58.234372: I tensorflow/core/common_runtime/gpu/gpu_device.cc:918] 0:   Y 
2017-03-28 07:53:58.234388: I tensorflow/core/common_runtime/gpu/gpu_device.cc:977] 
Creating TensorFlow device (/gpu:0) -> (device: 0, name: GeForce 840M, pci bus id: 0000:08:00.0)
('Epoch', 0, 'completed out of', 15, 'loss:', 115374329.04653475)

然后程序开始运行，但并没有按照我的预期运行得更快。我从官方文档安装了 CUDA，但我没有重置 git master head，因为它产生了问题，我使用了提供的相同优化标志 bazel build -c opt --config=cuda//tensorflow/tools/pip_package :build_pip_package 当通过 bazel 构建时。

Answer 1

您是否使用 nvidia-smi 来判断您是否安装了正确的 cuda 驱动程序以及您的 gpu 是否对系统可见？

在 TF 中你可以设置 log_device_placement了解是否有任何操作被分配给 GPU 的选项。