Skip to content

Instantly share code, notes, and snippets.

@hughperkins

hughperkins/Training working???

Created October 23, 2016 23:25
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save hughperkins/bd58940677c2d391dcbc2efb9c098654 to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save hughperkins/bd58940677c2d391dcbc2efb9c098654 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
Training working???
(env3cl) ubuntu@peach:~/git/tensorflow-blas$ python tensorflow/stream_executor/cl/test/test_gradients.py
CLBlas::initialize_cublas()
Using NVIDIA Corporation , OpenCL platform: NVIDIA CUDA
Using OpenCL device: GeForce 940M
X [[ 1. 0. 1. 0.]
[ 1. 0. 0. 1.]
[ 0. 1. 1. 0.]
[ 0. 1. 0. 1.]]
y [[ 1. 0.]
[ 0. 1.]
[ 0. 1.]
[ 0. 1.]]
cl_driver.cc cuInit()
I tensorflow/core/common_runtime/gpu/gpu_device.cc:983] Found device 0 with properties:
name: an opencl device
major: -1 minor: -1 memoryClockRate (GHz) 900
pciBusID 0000.0000
Total memory: 1.00MiB
Free memory: 1.00MiB
I tensorflow/core/common_runtime/gpu/gpu_device.cc:871] cannot enable peer access from device ordinal 0 to device ordinal 0
I tensorflow/core/common_runtime/gpu/gpu_device.cc:1005] DMA: 0
I tensorflow/core/common_runtime/gpu/gpu_device.cc:1015] 0: N
I tensorflow/core/common_runtime/gpu/gpu_device.cc:1077] Creating TensorFlow device (/gpu:0) -> (device: 0, name: an opencl device, pci bus id: 0000.0000)
grid(1, 1, 1)
block(128, 1, 1)
configureKernel (name=_ZN5Eigen8internal15EigenMetaKernelINS_15TensorEvaluatorIKNS_14TensorAssignOpINS_9TensorMapINS_6TensorIfLi2ELi1EiEELi16EEEKNS_19TensorCwiseBinaryOpINS0_13scalar_sum_opIffEEKNS4_INS5_IKfLi2ELi1EiEELi16EEEKNS_20TensorBroadcastingOpIKNS_5arrayIlLm2EEESE_EEEEEENS_9GpuDeviceEEEiEEvT_T0_
setKernelArgStruct structsize=52
setKernelArgCharStar 0x880
memory 0x41fb3d0 clmem 0x43b7710 offset=2048
setKernelArgCharStar 0x780
memory 0x41fb3d0 clmem 0x43b7710 offset=1792
setKernelArgCharStar 0x680
memory 0x41fb3d0 clmem 0x43b7710 offset=1536
setKernelArgInt32 8
.. kernel queued
epoch 0
loss [[ 0.80437583 1.21300113]
[ 4.65767813 0.13183187]
[ 2.03704906 0.13435066]
[ 2.85119271 0.39471319]]
[0 0 0 0]
grid(1, 1, 1)
block(128, 1, 1)
configureKernel (name=_ZN5Eigen8internal15EigenMetaKernelINS_15TensorEvaluatorIKNS_14TensorAssignOpINS_9TensorMapINS_6TensorIfLi2ELi1EiEELi16EEEKNS_19TensorCwiseBinaryOpINS0_13scalar_sum_opIffEEKNS4_INS5_IKfLi2ELi1EiEELi16EEEKNS_20TensorBroadcastingOpIKNS_5arrayIlLm2EEESE_EEEEEENS_9GpuDeviceEEEiEEvT_T0_
setKernelArgStruct structsize=52
setKernelArgCharStar 0x880
memory 0x41fb3d0 clmem 0x43b7710 offset=2048
setKernelArgCharStar 0x780
memory 0x41fb3d0 clmem 0x43b7710 offset=1792
setKernelArgCharStar 0x680
memory 0x41fb3d0 clmem 0x43b7710 offset=1536
setKernelArgInt32 8
.. kernel queued
epoch 1
loss [[ 1.99693286e+00 5.30970516e-04]
[ 2.08256766e-01 3.84304821e-01]
[ 8.00844789e-01 3.81741047e-01]
[ 8.80073190e-01 6.80260286e-02]]
[1 1 1 1]
grid(1, 1, 1)
block(128, 1, 1)
configureKernel (name=_ZN5Eigen8internal15EigenMetaKernelINS_15TensorEvaluatorIKNS_14TensorAssignOpINS_9TensorMapINS_6TensorIfLi2ELi1EiEELi16EEEKNS_19TensorCwiseBinaryOpINS0_13scalar_sum_opIffEEKNS4_INS5_IKfLi2ELi1EiEELi16EEEKNS_20TensorBroadcastingOpIKNS_5arrayIlLm2EEESE_EEEEEENS_9GpuDeviceEEEiEEvT_T0_
setKernelArgStruct structsize=52
setKernelArgCharStar 0x880
memory 0x41fb3d0 clmem 0x43b7710 offset=2048
setKernelArgCharStar 0x780
memory 0x41fb3d0 clmem 0x43b7710 offset=1792
setKernelArgCharStar 0x680
memory 0x41fb3d0 clmem 0x43b7710 offset=1536
setKernelArgInt32 8
.. kernel queued
epoch 2
loss [[ 2.65277009e-02 3.09681982e-01]
[ 8.77855897e-01 8.57838953e-04]
[ 4.54021275e-01 7.86582124e-04]
[ 2.00594068e-01 1.49130613e-01]]
[0 1 1 1]
grid(1, 1, 1)
block(128, 1, 1)
configureKernel (name=_ZN5Eigen8internal15EigenMetaKernelINS_15TensorEvaluatorIKNS_14TensorAssignOpINS_9TensorMapINS_6TensorIfLi2ELi1EiEELi16EEEKNS_19TensorCwiseBinaryOpINS0_13scalar_sum_opIffEEKNS4_INS5_IKfLi2ELi1EiEELi16EEEKNS_20TensorBroadcastingOpIKNS_5arrayIlLm2EEESE_EEEEEENS_9GpuDeviceEEEiEEvT_T0_
setKernelArgStruct structsize=52
setKernelArgCharStar 0x880
memory 0x41fb3d0 clmem 0x43b7710 offset=2048
setKernelArgCharStar 0x780
memory 0x41fb3d0 clmem 0x43b7710 offset=1792
setKernelArgCharStar 0x680
memory 0x41fb3d0 clmem 0x43b7710 offset=1536
setKernelArgInt32 8
.. kernel queued
epoch 3
loss [[ 4.47194904e-01 2.83233561e-02]
[ 1.83748307e-05 1.46821082e-01]
[ 2.62971055e-02 1.46250159e-01]
[ 2.47729212e-01 4.36995085e-03]]
[0 1 1 1]
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment