RuntimeError: clBuildProgram failed: BUILD_PROGRAM_FAILURE -

[assadRasheed]

i run the following code from Hunse repository t end of this code

import nengo
import nengo_ocl
import numpy as np

os.environ['PYOPENCL_COMPILER_OUTPUT'] = '1'

from nengo_extras.data import load_cifar10
from nengo_extras.cuda_convnet import CudaConvnetNetwork, load_model_pickle
from nengo_extras.gui import image_display_function



(X_train, y_train), (X_test, y_test), label_names = load_cifar10(label_names=True)
X_train = X_train.reshape(-1, 3, 32, 32).astype('float32')
X_test = X_test.reshape(-1, 3, 32, 32).astype('float32')
n_classes = len(label_names)

# crop data
X_train = X_train[:, :, 4:-4, 4:-4]
X_test = X_test[:, :, 4:-4, 4:-4]

# subtract mean
data_mean = X_train.mean(axis=0)
X_train -= data_mean
X_test -= data_mean

# retrieve from https://figshare.com/s/49741f9e2d0d29f68871
cc_model = load_model_pickle('cifar10-lif-1628.pkl')


# --- Run model in Nengo
presentation_time = 0.2

model = nengo.Network()
with model:
    u = nengo.Node(nengo.processes.PresentInput(X_test, presentation_time))
    ccnet = CudaConvnetNetwork(cc_model, synapse=nengo.synapses.Alpha(0.005))
    nengo.Connection(u, ccnet.input, synapse=None)

    input_p = nengo.Probe(u)
    output_p = nengo.Probe(ccnet.output)

    # --- image display
    image_shape = X_test.shape[1:]
    display_f = image_display_function(image_shape, scale=1, offset=data_mean)
    display_node = nengo.Node(display_f, size_in=u.size_out)
    nengo.Connection(u, display_node, synapse=None)

    # --- output spa display
    vocab_names = [s.upper() for s in label_names]
    print(vocab_names)
    
    vocab_vectors = np.eye(len(vocab_names))

    vocab = nengo.spa.Vocabulary(len(vocab_names))
    
    
    for name, vector in zip(vocab_names, vocab_vectors):
        
        vocab.add(name.decode('ascii'), vector)

    config = nengo.Config(nengo.Ensemble)
    config[nengo.Ensemble].neuron_type = nengo.Direct()
    with config:
        output = nengo.spa.State(len(vocab_names), subdimensions=10, vocab=vocab)
    nengo.Connection(ccnet.output, output.input)


n_presentations = 100

if 0:
    # run ANN in Theano
    os.environ['THEANO_FLAGS'] = 'device=gpu,floatX=float32'
    Q = ccnet.theano_compute(X_test[:n_presentations])
    Z = np.argmax(Q, axis=-1) == y_test[:n_presentations]
    print("ANN accuracy (%d examples): %0.4f" % (n_presentations, Z.mean()))


with nengo_ocl.Simulator(model) as sim:
    sim.run(n_presentations * presentation_time)

nt = int(presentation_time / sim.dt)
blocks = sim.data[output_p].reshape(n_presentations, nt, n_classes)
choices = np.argmax(blocks[:, -20:, :].mean(axis=1), axis=1)
accuracy = (choices == y_test[:n_presentations]).mean()
print('Spiking accuracy (%d examples): %0.3f' % (n_presentations, accuracy))

The errors messages are given below which i dont know how to rectify please suggest any solution

[b'AIRPLANE', b'AUTOMOBILE', b'BIRD', b'CAT', b'DEER', b'DOG', b'FROG', b'HORSE', b'SHIP', b'TRUCK']
C:\Users\zaid-pc\Anaconda3.0\lib\site-packages\pyopencl\cffi_cl.py:1516: CompilerWarning: Built kernel retrieved from cache. Original from-source build had warnings:
Build on <pyopencl.Device 'Intel(R) Core(TM) i3-3110M CPU @ 2.40GHz' on 'Intel(R) OpenCL' at 0x86f86c4> succeeded, but said:

Compilation started
Compilation done
Linking started
Linking done
Kernel <reset> was not vectorized
Done.
  warn(text, CompilerWarning)
C:\Users\zaid-pc\Anaconda3.0\lib\site-packages\pyopencl\cffi_cl.py:1516: CompilerWarning: From-binary build succeeded, but resulted in non-empty logs:
Build on <pyopencl.Device 'Intel(R) Core(TM) i3-3110M CPU @ 2.40GHz' on 'Intel(R) OpenCL' at 0x86f86c4> succeeded, but said:

Linking started
Linking done
Kernel <reset> was not vectorized
Done.
  warn(text, CompilerWarning)
C:\Users\zaid-pc\Anaconda3.0\lib\site-packages\pyopencl\cffi_cl.py:1516: CompilerWarning: Built kernel retrieved from cache. Original from-source build had warnings:
Build on <pyopencl.Device 'Intel(R) Core(TM) i3-3110M CPU @ 2.40GHz' on 'Intel(R) OpenCL' at 0x86f86c4> succeeded, but said:

Compilation started
Compilation done
Linking started
Linking done
Kernel <copy> was not vectorized
Done.
  warn(text, CompilerWarning)
C:\Users\zaid-pc\Anaconda3.0\lib\site-packages\pyopencl\cffi_cl.py:1516: CompilerWarning: From-binary build succeeded, but resulted in non-empty logs:
Build on <pyopencl.Device 'Intel(R) Core(TM) i3-3110M CPU @ 2.40GHz' on 'Intel(R) OpenCL' at 0x86f86c4> succeeded, but said:

Linking started
Linking done
Kernel <copy> was not vectorized
Done.
  warn(text, CompilerWarning)
C:\Users\zaid-pc\Anaconda3.0\lib\site-packages\pyopencl\cffi_cl.py:1516: CompilerWarning: Built kernel retrieved from cache. Original from-source build had warnings:
Build on <pyopencl.Device 'Intel(R) Core(TM) i3-3110M CPU @ 2.40GHz' on 'Intel(R) OpenCL' at 0x86f86c4> succeeded, but said:

Compilation started
Compilation done
Linking started
Linking done
Kernel <fn> was successfully vectorized
Done.
  warn(text, CompilerWarning)
C:\Users\zaid-pc\Anaconda3.0\lib\site-packages\pyopencl\cffi_cl.py:1516: CompilerWarning: From-binary build succeeded, but resulted in non-empty logs:
Build on <pyopencl.Device 'Intel(R) Core(TM) i3-3110M CPU @ 2.40GHz' on 'Intel(R) OpenCL' at 0x86f86c4> succeeded, but said:

Linking started
Linking done
Kernel <fn> was successfully vectorized
Done.
  warn(text, CompilerWarning)
C:\Users\zaid-pc\Anaconda3.0\lib\site-packages\pyopencl\cffi_cl.py:1516: CompilerWarning: Built kernel retrieved from cache. Original from-source build had warnings:
Build on <pyopencl.Device 'Intel(R) Core(TM) i3-3110M CPU @ 2.40GHz' on 'Intel(R) OpenCL' at 0x86f86c4> succeeded, but said:

Compilation started
Compilation done
Linking started
Linking done
Kernel <reduce> was not vectorized
Done.
  warn(text, CompilerWarning)
C:\Users\zaid-pc\Anaconda3.0\lib\site-packages\pyopencl\cffi_cl.py:1516: CompilerWarning: From-binary build succeeded, but resulted in non-empty logs:
Build on <pyopencl.Device 'Intel(R) Core(TM) i3-3110M CPU @ 2.40GHz' on 'Intel(R) OpenCL' at 0x86f86c4> succeeded, but said:

Linking started
Linking done
Kernel <reduce> was not vectorized
Done.
  warn(text, CompilerWarning)
C:\Users\zaid-pc\Anaconda3.0\lib\site-packages\pyopencl\cffi_cl.py:1516: CompilerWarning: Built kernel retrieved from cache. Original from-source build had warnings:
Build on <pyopencl.Device 'Intel(R) Core(TM) i3-3110M CPU @ 2.40GHz' on 'Intel(R) OpenCL' at 0x86f86c4> succeeded, but said:

Compilation started
Compilation done
Linking started
Linking done
Kernel <elementwise_inc> was not vectorized
Done.
  warn(text, CompilerWarning)
C:\Users\zaid-pc\Anaconda3.0\lib\site-packages\pyopencl\cffi_cl.py:1516: CompilerWarning: From-binary build succeeded, but resulted in non-empty logs:
Build on <pyopencl.Device 'Intel(R) Core(TM) i3-3110M CPU @ 2.40GHz' on 'Intel(R) OpenCL' at 0x86f86c4> succeeded, but said:

Linking started
Linking done
Kernel <elementwise_inc> was not vectorized
Done.
  warn(text, CompilerWarning)
C:\Users\zaid-pc\Anaconda3.0\lib\site-packages\pyopencl\cffi_cl.py:1516: CompilerWarning: Built kernel retrieved from cache. Original from-source build had warnings:
Build on <pyopencl.Device 'Intel(R) Core(TM) i3-3110M CPU @ 2.40GHz' on 'Intel(R) OpenCL' at 0x86f86c4> succeeded, but said:

Compilation started
Compilation done
Linking started
Linking done
Kernel <timeupdate> was successfully vectorized
Done.
  warn(text, CompilerWarning)
C:\Users\zaid-pc\Anaconda3.0\lib\site-packages\pyopencl\cffi_cl.py:1516: CompilerWarning: From-binary build succeeded, but resulted in non-empty logs:
Build on <pyopencl.Device 'Intel(R) Core(TM) i3-3110M CPU @ 2.40GHz' on 'Intel(R) OpenCL' at 0x86f86c4> succeeded, but said:

Linking started
Linking done
Kernel <timeupdate> was successfully vectorized
Done.
  warn(text, CompilerWarning)
---------------------------------------------------------------------------
RuntimeError                              Traceback (most recent call last)
<ipython-input-2-ebe41d061ca0> in <module>()
     81 
     82 
---> 83 with nengo_ocl.Simulator(model) as sim:
     84     sim.run(n_presentations * presentation_time)
     85 

~\Anaconda3.0\lib\site-packages\nengo_ocl\simulator.py in __init__(self, network, dt, seed, model, context, n_prealloc_probes, profiling, if_python_code, planner, progress_bar)
    306         with Timer() as plans_timer:
    307             for op_type, op_list in op_groups:
--> 308                 plans.extend(self.plan_op_group(op_type, op_list))
    309             plans.extend(self.plan_probes())
    310 

~\Anaconda3.0\lib\site-packages\nengo_ocl\simulator.py in plan_op_group(self, op_type, ops)
    600 
    601     def plan_op_group(self, op_type, ops):
--> 602         return getattr(self, 'plan_' + op_type.__name__)(ops)
    603 
    604     def plan_PreserveValue(self, ops):  # LEGACY

~\Anaconda3.0\lib\site-packages\nengo_ocl\simulator.py in plan_SimProcess(self, all_ops)
    930                 python_ops.extend(ops)
    931 
--> 932         process_plans = [p for attr, ops in iteritems(plan_groups)
    933                          for p in getattr(self, attr)(ops)]
    934         python_plans = [p for op in python_ops

~\Anaconda3.0\lib\site-packages\nengo_ocl\simulator.py in <listcomp>(.0)
    931 
    932         process_plans = [p for attr, ops in iteritems(plan_groups)
--> 933                          for p in getattr(self, attr)(ops)]
    934         python_plans = [p for op in python_ops
    935                         for p in self._plan_python_process(op)]

~\Anaconda3.0\lib\site-packages\nengo_ocl\simulator.py in _plan_Conv2d(self, ops)
   1024             plans.append(plan_conv2d(
   1025                 self.queue, X, Y, F, B, p.shape_in, p.shape_out,
-> 1026                 kernel_shape, conv, p.padding, p.strides))
   1027 
   1028         return plans

~\Anaconda3.0\lib\site-packages\nengo_ocl\clra_nonlinearities.py in plan_conv2d(queue, X, Y, filters, biases, shape_in, shape_out, kernel_shape, conv, padding, strides, tag, transposed)
   1642 
   1643     full_args = (X.base_data, filters.data, biases.data, Y.base_data)
-> 1644     _fn = cl.Program(queue.context, text).build().conv2d
   1645     _fn.set_args(*full_args)
   1646 

~\Anaconda3.0\lib\site-packages\pyopencl\__init__.py in build(self, options, devices, cache_dir)
    460                         self._context, self._source, options_bytes, devices,
    461                         cache_dir=cache_dir, include_path=include_path),
--> 462                     options_bytes=options_bytes, source=self._source)
    463 
    464             if was_cached:

~\Anaconda3.0\lib\site-packages\pyopencl\__init__.py in _build_and_catch_errors(self, build_func, options_bytes, source)
    504         # Python 3.2 outputs the whole list of currently active exceptions
    505         # This serves to remove one (redundant) level from that nesting.
--> 506         raise err
    507 
    508     # }}}

RuntimeError: clBuildProgram failed: BUILD_PROGRAM_FAILURE -

_Build on <pyopencl.Device 'Intel(R) Core(TM) i3-3110M CPU @ 2.40GHz' on 'Intel(R) OpenCL' at 0x86f86c4>:

Compilation started
:24:11: error: invalid operands to binary expression ('const __global float *' and 'double')
:25:11: error: invalid operands to binary expression ('__global float *' and 'double')
Compilation failed

(options: -I "c:\users\zaid-pc\anaconda3.0\lib\site-packages\pyopencl\cl")
(source saved as C:\Users\zaid-pc\AppData\Local\Temp\tmpx3vqw_gi.cl)_

[hunse]

You're building for a CPU. Nengo OCL has not been well tested for CPUs, since there are so many different kinds, and the performance increase over the reference Nengo simulator is minimal.

If you want to run on your CPU, I'd just use nengo.Simulator instead of nengo_ocl.Simulator. If you want to run things faster, then look into using your GPU with PyOpenCL.

[jgosmann]

performance increase over the reference Nengo simulator is minimal

I actually foud that there can be quite an performance improve running with OCL on the CPU in this paper.