def main(readcsv=read_csv, method='defaultDense'):
# read data from file
file = "./data/batch/covcormoments_dense.csv"
# Configure a low order moments object for streaming
algo = d4p.low_order_moments(streaming=True)
chunk_size = 55
lines_read = 0
# read and feed chunk by chunk
while True:
# Read data in chunks
try:
data = readcsv(file, range(10), lines_read, chunk_size)
except:
break
# Now feed chunk
algo.compute(data)
lines_read += data.shape[0]
# All files are done, now finalize the computation
result = algo.finalize()
# result provides minimum, maximum, sum, sumSquares, sumSquaresCentered,
# mean, secondOrderRawMoment, variance, standardDeviation, variation
return result
def _daal_mean_var(X):
fpt = getFPType(X)
try:
alg = daal4py.low_order_moments(fptype=fpt,
method='defaultDense',
estimatesToCompute='estimatesAll')
except AttributeError:
return np.var(X, axis=0).mean()
ssc = alg.compute(X).sumSquaresCentered
ssc = ssc.reshape((-1, 1))
alg = daal4py.low_order_moments(fptype=fpt,
method='defaultDense',
estimatesToCompute='estimatesAll')
ssc_total_res = alg.compute(ssc)
mean_var = ssc_total_res.sum / X.size
return mean_var[0, 0]
def main():
infile = "./data/batch/covcormoments_dense.csv"
# We know the number of lines in the file
# and use this to separate data between processes
skiprows, nrows = get_chunk_params(lines_count=200,
chunks_count=d4p.num_procs(),
chunk_number=d4p.my_procid())
# Each process reads its chunk of the file
data = read_csv(infile, sr=skiprows, nr=nrows)
# Create algorithm with distributed mode
alg = d4p.low_order_moments(method='defaultDense', distributed=True)
# Perform computation
res = alg.compute(data)
# result provides minimum, maximum, sum, sumSquares, sumSquaresCentered,
# mean, secondOrderRawMoment, variance, standardDeviation, variation
assert (all(
getattr(res, name).shape == (1, data.shape[1]) for name in [
'minimum', 'maximum', 'sum', 'sumSquares', 'sumSquaresCentered',
'mean', 'secondOrderRawMoment', 'variance', 'standardDeviation',
'variation'
]))
return res
def _daal_var(X):
"""DAAL-based threaded computation of X.std()"""
fpt = getFPType(X)
try:
alg = daal4py.low_order_moments(fptype=fpt, method='defaultDense', estimatesToCompute='estimatesMeanVariance')
except AttributeError:
return np.var(X)
ssc = alg.compute(X.reshape(-1,1)).sumSquaresCentered
return ssc[0, 0] / X.size
def main(readcsv=read_csv, method="defaultDense"):
# read data from file
file = "./data/batch/covcormoments_dense.csv"
data = readcsv(file, range(10))
# compute
alg = d4p.low_order_moments(method=method)
res = alg.compute(data)
# result provides minimum, maximum, sum, sumSquares, sumSquaresCentered,
# mean, secondOrderRawMoment, variance, standardDeviation, variation
assert(all(getattr(res, name).shape==(1, data.shape[1]) for name in
['minimum', 'maximum', 'sum', 'sumSquares', 'sumSquaresCentered', 'mean',
'secondOrderRawMoment', 'variance', 'standardDeviation', 'variation']))
return res
def main():
# read data from file
file = "./data/batch/covcormoments_dense.csv"
data = read_csv(file, range(10))
# compute
alg = d4p.low_order_moments()
res = alg.compute(data)
# result provides minimum, maximum, sum, sumSquares, sumSquaresCentered,
# mean, secondOrderRawMoment, variance, standardDeviation, variation
assert res.minimum.shape == (1, data.shape[1])
assert res.maximum.shape == (1, data.shape[1])
assert res.sum.shape == (1, data.shape[1])
assert res.sumSquares.shape == (1, data.shape[1])
assert res.sumSquaresCentered.shape == (1, data.shape[1])
assert res.mean.shape == (1, data.shape[1])
assert res.secondOrderRawMoment.shape == (1, data.shape[1])
assert res.variance.shape == (1, data.shape[1])
assert res.standardDeviation.shape == (1, data.shape[1])
assert res.variation.shape == (1, data.shape[1])
return res
def compute(data, method):
alg = d4p.low_order_moments(method=method)
return alg.compute(data)
def main(readcsv=None, method='defaultDense'):
# read data from file
infile = os.path.join('..', 'data', 'batch', 'covcormoments_dense.csv')
# Using of the classic way (computations on CPU)
# Configure a low order moments object for streaming
algo = d4p.low_order_moments(streaming=True)
# get the generator (defined in stream.py)...
rn = read_next(infile, 55, readcsv)
# ... and iterate through chunks/stream
for chunk in rn:
algo.compute(chunk)
# finalize computation
result_classic = algo.finalize()
# It is possible to specify to make the computations on GPU
try:
with sycl_context('gpu'):
# Configure a low order moments object for streaming
algo = d4p.low_order_moments(streaming=True)
# get the generator (defined in stream.py)...
rn = read_next(infile, 55, readcsv)
# ... and iterate through chunks/stream
for chunk in rn:
sycl_chunk = sycl_buffer(to_numpy(chunk))
algo.compute(sycl_chunk)
# finalize computation
result_gpu = algo.finalize()
for name in [
'minimum', 'maximum', 'sum', 'sumSquares',
'sumSquaresCentered', 'mean', 'secondOrderRawMoment',
'variance', 'standardDeviation', 'variation'
]:
assert np.allclose(getattr(result_classic, name),
getattr(result_gpu, name))
except:
pass
# It is possible to specify to make the computations on CPU
with sycl_context('cpu'):
# Configure a low order moments object for streaming
algo = d4p.low_order_moments(streaming=True)
# get the generator (defined in stream.py)...
rn = read_next(infile, 55, readcsv)
# ... and iterate through chunks/stream
for chunk in rn:
sycl_chunk = sycl_buffer(to_numpy(chunk))
algo.compute(sycl_chunk)
# finalize computation
result_cpu = algo.finalize()
# result provides minimum, maximum, sum, sumSquares, sumSquaresCentered,
# mean, secondOrderRawMoment, variance, standardDeviation, variation
for name in [
'minimum', 'maximum', 'sum', 'sumSquares', 'sumSquaresCentered',
'mean', 'secondOrderRawMoment', 'variance', 'standardDeviation',
'variation'
]:
assert np.allclose(getattr(result_classic, name),
getattr(result_cpu, name))
return result_classic
def compute(data, method):
alg = d4p.low_order_moments(method=method, fptype='float')
return alg.compute(data)