def test_4():
data = [np.random.randn(10, 1), np.random.randn(100, 1)]
print(cluster.KMeans(n_clusters=3).fit_predict(data))
print(cluster.MiniBatchKMeans(n_clusters=3).fit_predict(data))
print(cluster.AffinityPropagation().fit_predict(data))
print(cluster.MeanShift().fit_predict(data))
print(cluster.SpectralClustering(n_clusters=2).fit_predict(data))
print(cluster.Ward(n_clusters=2).fit_predict(data))
def test_fit_on_many_clusterings():
data = [np.random.randn(10, 1), np.random.randn(100, 1)]
print(cluster.KMeans(n_clusters=3).fit_predict(data))
print(cluster.MiniBatchKMeans(n_clusters=3).fit_predict(data))
print(cluster.AffinityPropagation().fit_predict(data))
print(cluster.MeanShift().fit_predict(data))
print(cluster.SpectralClustering(n_clusters=2).fit_predict(data))
print(cluster.AgglomerativeClustering(n_clusters=2).fit_predict(data))
def test_sample_1():
# Test that the code actually runs and gives something non-crazy
# Make an ergodic dataset with two gaussian centers offset by 25 units.
chunk = np.random.normal(size=(20000, 3))
data = [np.vstack((chunk, chunk + 25)), np.vstack((chunk + 25, chunk))]
clusterer = cluster.KMeans(n_clusters=2)
msm = MarkovStateModel()
pipeline = sklearn.pipeline.Pipeline([("clusterer", clusterer),
("msm", msm)])
pipeline.fit(data)
trimmed_assignments = pipeline.transform(data)
# Now let's make make the output assignments start with
# zero at the first position.
i0 = trimmed_assignments[0][0]
if i0 == 1:
for m in trimmed_assignments:
m *= -1
m += 1
pairs = msm.draw_samples(trimmed_assignments, 2000)
samples = map_drawn_samples(pairs, data)
mu = np.mean(samples, axis=1)
eq(mu, np.array([[0., 0., 0.0], [25., 25., 25.]]), decimal=1)
# We should make sure we can sample from Trajectory objects too...
# Create a fake topology with 1 atom to match our input dataset
top = md.Topology.from_dataframe(pd.DataFrame({
"serial": [0],
"name": ["HN"],
"element": ["H"],
"resSeq": [1],
"resName": "RES",
"chainID": [0]
}),
bonds=np.zeros(shape=(0, 2), dtype='int'))
# np.newaxis reshapes the data to have a 40000 frames, 1 atom, 3 xyz
trajectories = [md.Trajectory(x[:, np.newaxis], top) for x in data]
trj_samples = map_drawn_samples(pairs, trajectories)
mu = np.array([t.xyz.mean(0)[0] for t in trj_samples])
eq(mu, np.array([[0., 0., 0.0], [25., 25., 25.]]), decimal=1)
"""
from msmbuilder import example_datasets, cluster, msm, featurizer, lumping, utils, dataset, decomposition
from sklearn.pipeline import make_pipeline
dih = dataset.NumpyDirDataset("./dihedrals/")
X = dataset.dataset("./tica.h5")
Xf = np.concatenate(X)
tica_model = utils.load("./tica.pkl")
dih_model = utils.load("./dihedrals/model.pkl")
n_first = 17
n_clusters = 9
slicer = featurizer.FirstSlicer(n_first)
clusterer = cluster.KMeans(n_clusters=n_clusters)
msm_model = msm.MarkovStateModel()
pipeline = make_pipeline(slicer, clusterer, msm_model)
s = pipeline.fit_transform(X)
p0 = make_pipeline(dih_model, tica_model, slicer)
trajectories = dataset.MDTrajDataset("./trajectories/*.h5")
selected_pairs_by_state = msm_model.draw_samples(s, 5)
samples = utils.map_drawn_samples(selected_pairs_by_state, trajectories)
for k, t in enumerate(samples):
t.save("./pdbs/state%d.pdb" % k)
Y = p0.transform(samples)