Tensorflow 版本 2.6.0，cuda 版本=10.2 未使用 GPU

Question

我使用的 tensorflow 被其他人意外更改后，现在没有使用 gpu。我记得我使用轮子安装了我的 tensorflow。目前Cuda版本是10.2，python版本是3.7，tensorflow版本是2.6.0。

以下代码的输出给了我：

import tensorflow as tf
print("Num GPUs Available: ", len(tf.config.experimental.list_physical_devices('GPU')))

tf.debugging.set_log_device_placement(True)

# Create some tensors
a = tf.constant([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])
b = tf.constant([[1.0, 2.0], [3.0, 4.0], [5.0, 6.0]])
c = tf.matmul(a, b)

print(c)

**Num GPUs Available:  0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:CPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:CPU:0
Executing op MatMul in device /job:localhost/replica:0/task:0/device:CPU:0
tf.Tensor(
[[22. 28.]
 [49. 64.]], shape=(2, 2), dtype=float32)**

我附上了我的 nvidia-smi 命令输出的图像

$ nvidia-smi
Sat Aug 28 14:05:15 2021
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.33.01    Driver Version: 440.33.01    CUDA Version: 10.2     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce RTX 2060    Off  | 00000000:21:00.0 Off |                  N/A |
| 32%   43C    P8     9W / 160W |     92MiB /  5934MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|    0     13060      C   /home/hemapriya/anaconda3/bin/python          81MiB |
+-----------------------------------------------------------------------------+

!conda list cudatoolkit
!conda list cudnn

Output:

# packages in environment at /home/hemapriya/anaconda3:
#
# Name                    Version                   Build  Channel
cudatoolkit               10.0.130                      0    anaconda
# packages in environment at /home/hemapriya/anaconda3:
#
# Name                    Version                   Build  Channel
cudnn                     7.6.5                cuda10.0_0    anaconda

from tensorflow.python.client import device_lib
print(device_lib.list_local_devices())

output:

[name: "/device:CPU:0"
device_type: "CPU"
memory_limit: 268435456
locality {
}
incarnation: 15409823727920383739
]

另外，在那次事件之后，我的 tf.keras.optimizers.Adam 现在不工作了。

能否请您告诉我，我应该如何纠正，以便我的 tensorflow 应该包含 GPU。

Answer 1

您正面临 Tensorflow 版本与 CUDA、cuDNN 驱动程序不兼容的问题。根据Tensorflow Tested build configurationTensorflow 2.6与CUDA 11.2和cuDNN 8.1兼容。

要解决您的问题，您可以安装CUDA

conda install -c anaconda cudatoolkit=11.2 cudnn=8.1

或将 TensorFlow 版本降级为 2.3。

import tensorflow as tf
print("Num GPUs Available: ", len(tf.config.experimental.list_physical_devices('GPU')))

tf.debugging.set_log_device_placement(True)

# Create some tensors
a = tf.constant([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])
b = tf.constant([[1.0, 2.0], [3.0, 4.0], [5.0, 6.0]])
c = tf.matmul(a, b)

print(c)

输出：

Num GPUs Available:  1
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op MatMul in device /job:localhost/replica:0/task:0/device:GPU:0
tf.Tensor(
[[22. 28.]
 [49. 64.]], shape=(2, 2), dtype=float32)

我可以运行tf.keras.optimizers.Adam，如下所示

opt = tf.keras.optimizers.Adam(learning_rate=0.1)
var1 = tf.Variable(10.0)
loss = lambda: (var1 ** 2)/2.0       
step_count = opt.minimize(loss, [var1]).numpy()
var1.numpy()

输出：

Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op VarHandleOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op AssignVariableOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op ReadVariableOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Pow in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op RealDiv in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Shape in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Shape in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op BroadcastGradientArgs in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op RealDiv in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Sum in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Reshape in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Neg in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op RealDiv in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op RealDiv in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Mul in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Sum in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Reshape in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Mul in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Sub in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Pow in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Mul in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Fill in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op VarHandleOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op AssignVariableOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op VarHandleOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op AssignVariableOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op VarHandleOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op AssignVariableOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op VarHandleOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op AssignVariableOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op VarHandleOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op AssignVariableOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Fill in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op VarHandleOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op AssignVariableOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Fill in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op VarHandleOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op AssignVariableOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op ReadVariableOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Identity in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op ReadVariableOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op AddV2 in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Cast in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op ReadVariableOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Identity in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op ReadVariableOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Identity in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Pow in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Pow in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Sub in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Sqrt in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Sub in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op RealDiv in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Mul in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Sub in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Sub in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op ResourceApplyAdam in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op _EagerConst in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op AssignAddVariableOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op ReadVariableOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op ReadVariableOp in device /job:localhost/replica:0/task:0/device:GPU:0
Executing op Identity in device /job:localhost/replica:0/task:0/device:GPU:0
9.899997