def run_gradweight(self,
inputs_shape,
filters_shape,
dCdH_shape,
subsample=(1, 1, 1)):
inputs_shape = [inputs_shape[i] for i in (0, 4, 1, 2, 3)]
filters_shape = [filters_shape[i] for i in (0, 4, 1, 2, 3)]
dCdH_shape = [dCdH_shape[i] for i in (0, 4, 1, 2, 3)]
inputs_val = np.random.random(inputs_shape).astype(config.floatX)
dCdH_val = np.random.random(dCdH_shape).astype(config.floatX)
inputs = gpuarray_shared_constructor(inputs_val)
dCdH = gpuarray_shared_constructor(dCdH_val)
shape = gpuarray_shared_constructor(np.array(filters_shape[2:]))
if (subsample == (1, 1, 1)):
conv_ref = Corr3dMM_gradWeights(subsample=subsample)(
ref_cast(inputs), ref_cast(dCdH))
conv_gemm = GpuCorr3dMM_gradWeights(subsample=subsample)(inputs,
dCdH)
else:
conv_ref = Corr3dMM_gradWeights(subsample=subsample)(
ref_cast(inputs), ref_cast(dCdH), shape=shape)
conv_gemm = GpuCorr3dMM_gradWeights(subsample=subsample)(
inputs, dCdH, shape=shape)
f_ref = theano.function([], conv_ref, mode=mode_without_gpu)
f = theano.function([], conv_gemm, mode=mode_with_gpu)
res_ref = f_ref()
res = f()
utt.assert_allclose(res_ref, res)
def run_gradweight(self,
inputs_shape,
filters_shape,
dCdH_shape,
subsample=(1, 1, 1)):
inputs_val = numpy.random.random(inputs_shape).astype('float32')
dCdH_val = numpy.random.random(dCdH_shape).astype('float32')
inputs = shared(inputs_val)
dCdH = shared(dCdH_val)
img = gpu_contiguous(inputs.dimshuffle(0, 4, 1, 2, 3))
topgrad = gpu_contiguous(dCdH.dimshuffle(0, 4, 1, 2, 3))
if (subsample == (1, 1, 1)):
conv_ref = Corr3dMM_gradWeights(subsample=subsample)(img, topgrad)
conv_gemm = GpuCorr3dMM_gradWeights(subsample=subsample)(img,
topgrad)
else:
conv_ref = Corr3dMM_gradWeights(subsample=subsample)(
img, topgrad, shape=filters_shape[1:4])
conv_gemm = GpuCorr3dMM_gradWeights(subsample=subsample)(
img, topgrad, shape=filters_shape[1:4])
f_ref = theano.function([], conv_ref, mode='FAST_RUN')
f = theano.function([], conv_gemm, mode=mode_with_gpu)
res_ref = f_ref()
res = f()
utt.assert_allclose(res_ref, res)
def local_abstractconv3d_gradweight_gemm(node):
# If theano.config.blas.ldflags is empty, Theano will use
# a NumPy C implementation of [sd]gemm_.
if theano.config.cxx == "" or node.inputs[0].dtype == "float16":
return
if not isinstance(node.op, AbstractConv3d_gradWeights):
return None
img, topgrad, shape = node.inputs
if not isinstance(img.type, TensorType) or not isinstance(
topgrad.type, TensorType):
return None
rval = Corr3dMM_gradWeights(
border_mode=node.op.border_mode,
subsample=node.op.subsample,
filter_dilation=node.op.filter_dilation,
num_groups=node.op.num_groups,
)(img, topgrad, shape)
copy_stack_trace(node.outputs[0], rval)
# need to flip the kernel if necessary
if node.op.filter_flip:
rval = rval[:, :, ::-1, ::-1, ::-1]
rval = theano.tensor.patternbroadcast(rval, node.outputs[0].broadcastable)
copy_stack_trace(node.outputs[0], rval)
return [rval]
def local_abstractconv3d_gradweight_gemm(node):
if theano.config.cxx == "" or not theano.config.blas.ldflags:
return
if not isinstance(node.op, AbstractConv3d_gradWeights):
return None
img, topgrad, shape = node.inputs
if not isinstance(img.type, TensorType) or \
not isinstance(topgrad.type, TensorType):
return None
rval = Corr3dMM_gradWeights(border_mode=node.op.border_mode,
subsample=node.op.subsample,
filter_dilation=node.op.filter_dilation)(
img, topgrad, shape)
copy_stack_trace(node.outputs[0], rval)
# need to flip the kernel if necessary
if node.op.filter_flip:
rval = rval[:, :, ::-1, ::-1, ::-1]
rval = theano.tensor.patternbroadcast(rval, node.outputs[0].broadcastable)
copy_stack_trace(node.outputs[0], rval)
return [rval]