I'm not a beginner in machine learning and GPU usage. I had a fine work environment, but recently the Windows was updated and there were problems. I have next env:
- OS: Win10 64
- Python 3.5.4
- Tensorflow-gpu 1.7.0
- Nvidia driver
- CUDA Toolkit - 9.0
- cuDNN - 7.1 for CUDA 9.0
Next code work fine
>>> import tensorflow as tf
>>> hello = tf.constant('Hello, TensorFlow!')
>>> sess = tf.Session()
>>> print(sess.run(hello))
But when i try use GPU i give next trouble:
tensorflow.python.framework.errors_impl.InvalidArgumentError:
Cannot assign a device for operation 'random_uniform/sub':
Operation was explicitly assigned to /device:GPU:0 but available devices are [ /job:localhost/replica:0/task:0/device:CPU:0 ].
Make sure the device specification refers to a valid device.
[[Node: random_uniform/sub = Sub[T=DT_FLOAT, _device="/device:GPU:0"](random_uniform/max, random_uniform/min)]]
I think that is because i don't have Microsoft Visual C++ 2015 Redistributable Update 3. But when i try install it i get message that "another version of this product is already installed". I tried to find answer in google a few hours but nothing. Why this problem may be occurs?
