CODEWITHSUNDEEP
tensorflowgpu
user297850
user297850

Reputation: 8015

Regarding the message of ignoring visible gpu devices

When running a tensorflow program leveraging GPU devices, I got the following message. What does the statement of Ignoring visible gpu device (device: 0, name: Quadro 5000, pci bus id: 0000:05:00.0) with Cuda compute capability 2.0. mean? What is to be ignored?

2017-12-04 16:06:17.784599: I 
tensorflow/core/common_runtime/gpu/gpu_device.cc:887] Found device 0 with 
properties:
name: Quadro 5000
major: 2 minor: 0 memoryClockRate (GHz) 1.026
pciBusID 0000:05:00.0
Total memory: 2.50GiB
Free memory: 2.22GiB
2017-12-04 16:06:17.784636: I 
tensorflow/core/common_runtime/gpu/gpu_device.cc:908] DMA: 0
2017-12-04 16:06:17.784645: I 
tensorflow/core/common_runtime/gpu/gpu_device.cc:918] 0:   Y
2017-12-04 16:06:17.784658: I 
tensorflow/core/common_runtime/gpu/gpu_device.cc:950] Ignoring visible gpu 
device (device: 0, name: Quadro 5000, pci bus id: 0000:05:00.0) with Cuda 
compute capability 2.0. The minimum required Cuda capability is 3.0.

Upvotes: 2

Views: 814

Answers (1)

mrry
mrry

Reputation: 126184

TensorFlow prints this message when you have a GPU installed in the local machine but its compute capability is too low for TensorFlow to use that GPU for accelerated computation. As the error message states:

The minimum required Cuda capability is 3.0.

This means that you must use a CUDA-capable GPU with compute capability 3.0 or greater to use TensorFlow for GPU.

Upvotes: 2

Related Questions