def testChunksizeResultsTica(self):
chunk = 40
lag = 100
np.random.seed(0)
X = np.random.randn(23000, 3)
# un-chunked
d = DataInMemory(X)
tica = TICA(lag=lag, output_dimension=1)
tica.data_producer = d
tica.parametrize()
cov = tica.cov.copy()
mean = tica.mu.copy()
# ------- run again with new chunksize -------
d = DataInMemory(X)
d.chunksize = chunk
tica = TICA(lag=lag, output_dimension=1)
tica.data_producer = d
tica.parametrize()
np.testing.assert_allclose(tica.mu, mean)
np.testing.assert_allclose(tica.cov, cov)
def __init__(self, data, lag, units='frames'):
from pyemma.coordinates import tica
# data.dat.tolist() might be better?
self.data = data
if isinstance(data, Metric):
if units != 'frames':
raise RuntimeError(
'Cannot use delayed projection TICA with units other than frames for now. Report this to HTMD issues.'
)
metr = data
from pyemma.coordinates.transform.tica import TICA
self.tic = TICA(lag)
p = ProgressBar(len(metr.simulations))
for proj in _projectionGenerator(metr, _getNcpus()):
for pro in proj:
self.tic.partial_fit(pro[0])
p.progress(len(proj))
p.stop()
else:
lag = unitconvert(units, 'frames', lag, data.fstep)
if lag == 0:
raise RuntimeError(
'Lag time conversion resulted in 0 frames. Please use a larger lag-time for TICA.'
)
self.tic = tica(data.dat.tolist(), lag=lag)
def test(self):
np.random.seed(0)
tica = TICA(lag=50, output_dimension=1)
data = np.random.randn(100, 10)
ds = DataInMemory(data)
tica.data_producer = ds
tica.parametrize()
Y = tica.map(data)
def test_singular_zeros(self):
tica = TICA(lag=1, output_dimension=1)
# make some data that has one column of all zeros
X = np.random.randn(100, 2)
X = np.hstack((X, np.zeros((100, 1))))
d = DataInMemory(X)
tica.data_producer = d
tica.parametrize()
assert tica.eigenvectors.dtype == np.float64
assert tica.eigenvalues.dtype == np.float64
def test_duplicated_data(self):
tica = TICA(lag=1, output_dimension=1)
# make some data that has one column repeated twice
X = np.random.randn(100, 2)
X = np.hstack((X, X[:, 0, np.newaxis]))
d = DataInMemory(X)
tica.data_producer = d
tica.parametrize()
assert tica.eigenvectors.dtype == np.float64
assert tica.eigenvalues.dtype == np.float64
def __init__(self, data, lag, units='frames'):
from pyemma.coordinates import tica
# data.dat.tolist() might be better?
self.data = data
if isinstance(data, Metric):
from pyemma.coordinates.transform.tica import TICA
lag = unitconvert(units, 'frames', lag, data.fstep)
self.tic = TICA(lag)
p = ProgressBar(len(data.simulations))
for i in range(len(data.simulations)):
# Fix for pyemma bug. Remove eventually:
d, _, _ = data._projectSingle(i)
if d is None or d.shape[0] < lag:
continue
self.tic.partial_fit(d)
p.progress()
p.stop()
else:
self.tic = tica(data.dat.tolist(), lag=lag)
