Closed tgs266 closed 7 years ago
change to a different directory :-) You cant run python from the source tree.
The problem is that I am not in the source directory. I am in the home directory, running "python" in the terminal, then trying to import tensorflow. Even if I try to go into another folder like Documents or Downloads it doesn't work
Ah, hmmm, right. Seems like it cant import the version of libstdc++ that its expecting. What answers did you give to the questions about python paths, when you did ./configure
in tensorflow?
I never did ./configure. Nothing about that came up during installation. All I did was run the wheel file
I am trying to install from source right now to see if it changes anything
Ah. Installing from source sounds like a good plan. Or using a binary that was built against the exact same version of anaconda you are using.
Ok I think I found one of the problems... I am not running OpenCL 1.2+ I am on 1.1. I have been googling for about and hour now with no luck. Do you know how to update opencl on ubuntu 16.04?
What do you mean by 'running OpenCL 1.2'? Normally, your system doesnt realy have an opencl version as such, I think. The GPU driver is what has an OpenCL version. What is the output of running:
sudo apt-get install clinfo
clinfo
?
Here is the output:
Number of platforms 1
Platform Name Clover
Platform Vendor Mesa
Platform Version OpenCL 1.1 MESA 11.2.0
Platform Profile FULL_PROFILE
Platform Extensions cl_khr_icd
Platform Extensions function suffix MESA
Platform Name Clover
Number of devices 1
Device Name AMD PITCAIRN (DRM 2.43.0, LLVM 3.8.0)
Device Vendor AMD
Device Vendor ID 0x1002
Device Version OpenCL 1.1 MESA 11.2.0
Driver Version 11.2.0
Device OpenCL C Version OpenCL C 1.1
Device Type GPU
Device Profile FULL_PROFILE
Max compute units 20
Max clock frequency 945MHz
Max work item dimensions 3
Max work item sizes 256x256x256
Max work group size 256
Preferred work group size multiple 64
Preferred / native vector sizes
char 16 / 16
short 8 / 8
int 4 / 4
long 2 / 2
half 0 / 0 (n/a)
float 4 / 4
double 2 / 2 (cl_khr_fp64)
Half-precision Floating-point support (n/a)
Single-precision Floating-point support (core)
Denormals No
Infinity and NANs Yes
Round to nearest Yes
Round to zero No
Round to infinity No
IEEE754-2008 fused multiply-add No
Support is emulated in software No
Correctly-rounded divide and sqrt operations No
Double-precision Floating-point support (cl_khr_fp64)
Denormals Yes
Infinity and NANs Yes
Round to nearest Yes
Round to zero Yes
Round to infinity Yes
IEEE754-2008 fused multiply-add Yes
Support is emulated in software No
Correctly-rounded divide and sqrt operations No
Address bits 32, Little-Endian
Global memory size 1073741824 (1024MiB)
Error Correction support No
Max memory allocation 268435456 (256MiB)
Unified memory for Host and Device Yes
Minimum alignment for any data type 128 bytes
Alignment of base address 1024 bits (128 bytes)
Global Memory cache type None
Image support No
Local memory type Local
Local memory size 32768 (32KiB)
Max constant buffer size 268435456 (256MiB)
Max number of constant args 16
Max size of kernel argument 1024
Queue properties
Out-of-order execution No
Profiling Yes
Profiling timer resolution 0ns
Execution capabilities
Run OpenCL kernels Yes
Run native kernels No
Device Available Yes
Compiler Available Yes
Device Extensions cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_byte_addressable_store cl_khr_fp64
NULL platform behavior
clGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) Clover
clGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) Success [MESA]
clCreateContext(NULL, ...) [default] Success [MESA]
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) Success (1)
Platform Name Clover
Device Name AMD PITCAIRN (DRM 2.43.0, LLVM 3.8.0)
clCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) No devices found in platform
clCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) Success (1)
Platform Name Clover
Device Name AMD PITCAIRN (DRM 2.43.0, LLVM 3.8.0)
ICD loader properties
ICD loader Name OpenCL ICD Loader
ICD loader Vendor OCL Icd free software
ICD loader Version 2.2.8
ICD loader Profile OpenCL 1.2
NOTE: your OpenCL library declares to support OpenCL 1.2,
but it seems to support up to OpenCL 2.1 too.
Also my gpu is r9 270x
Oh, Clover...
Soo....
Clover is an opensource driver, but it's OpenCL 1.1 as you see. Except it's not, because it doesnt support basic maths like tanh
.
You can choose to try using Clover, but I dont support 1.1 for now: you'd need to hack around in the code, adding shims and stuff. This is hard.
On the whole, if you prefer 'easy' to 'hard', you might consider installing the proprietary AMD drivers.
(Note that the AMD driver for 270x supports 1.2: https://en.wikipedia.org/wiki/List_of_AMD_graphics_processing_units )
Ok. I am going to have to look around a little for a driver that is compatible with both the 270x and 16.04... There is a chance I am just going to have to roll back ubuntu to 14.04 or 15 which is going to be a pain to do. Thanks for your help
Hello Hugh, I installed tensorflow-cl as it said, but now I can't run any of the tests... I cant even import tensorflow in any file. Everytime I get the same error:
How do I fix this? Thanks in advance