msmbuilder / msmbuilder-legacy

Legacy release of MSMBuilder
http://msmbuilder.org
GNU General Public License v2.0
25 stars 28 forks source link

Clustering by contacts fails with varying errors #441

Open Trayder opened 9 years ago

Trayder commented 9 years ago

I can cluster by RMSD fine. But clustering by contact fails on multiple data sets. I get all sorts of different errors depending on settings and clustering algorithm. Anyone know what's going on here? Thanks, -Trayder

In the below examples my test.dat file simply reads: 1 2 1 3

Cluster contact -c test.dat -C -1 -s closest hybrid -k 10

Traceback (most recent call last): File "/vcp2/student/trayder/anaconda/bin/Cluster", line 9, in load_entry_point('msmbuilder==2.8.2', 'console_scripts', 'Cluster')() File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/msmbuilder-2.8.2-py2.7.egg/msmbuilder/scripts/Cluster.py", line 287, in entry_point main(args, metric) File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/msmbuilder-2.8.2-py2.7.egg/msmbuilder/scripts/Cluster.py", line 248, in main raise Exception("something went wrong in loading step (%d v %d)" % (num_frames, len(which))) Exception: something went wrong in loading step (120 v 1260)`

Cluster contact -c test.dat -C 1 -s closest hybrid -k 10

Traceback (most recent call last): File "/vcp2/student/trayder/anaconda/bin/Cluster", line 9, in load_entry_point('msmbuilder==2.8.2', 'console_scripts', 'Cluster')() File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/msmbuilder-2.8.2-py2.7.egg/msmbuilder/scripts/Cluster.py", line 287, in entry_point main(args, metric) File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/msmbuilder-2.8.2-py2.7.egg/msmbuilder/scripts/Cluster.py", line 242, in main ptrajs, which = load_prep_trajectories(project, args.stride, atom_indices, metric) File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/msmbuilder-2.8.2-py2.7.egg/msmbuilder/scripts/Cluster.py", line 135, in load_prep_trajectories ptraj = np.concatenate(ptraj) ValueError: zero-dimensional arrays cannot be concatenated`

Cluster contact -c test.dat -C -1 -s closest hierarchical -m ward

16:14:07 - Loaded 60 trajs 16:14:07 - Preparing... 16:14:08 - Getting all to all pairwise distance matrix... Traceback (most recent call last): File "/vcp2/student/trayder/anaconda/bin/Cluster", line 9, in load_entry_point('msmbuilder==2.8.2', 'console_scripts', 'Cluster')() File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/msmbuilder-2.8.2-py2.7.egg/msmbuilder/scripts/Cluster.py", line 287, in entry_point main(args, metric) File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/msmbuilder-2.8.2-py2.7.egg/msmbuilder/scripts/Cluster.py", line 257, in main clusterer = cluster(metric, trajectories, ptrajs, args, extra_kwargs) File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/msmbuilder-2.8.2-py2.7.egg/msmbuilder/scripts/Cluster.py", line 189, in cluster method=args.hierarchical_method) File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/msmbuilder-2.8.2-py2.7.egg/msmbuilder/clustering.py", line 769, in init** dmat = metric.all_pairwise(pflat_trajectory) File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/msmbuilder-2.8.2-py2.7.egg/msmbuilder/metrics/baseclasses.py", line 343, in all_pairwise V=self.V, VI=self.VI) File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/scipy/spatial/distance.py", line 1178, in pdist [X] = _copy_arrays_if_base_present([_convert_to_double(X)]) File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/scipy/spatial/distance.py", line 113, in _convert_to_double X = X.astype(np.double) ValueError: setting an array element with a sequence.`

Cluster contact -c test.dat -C 1 -s closest hierarchical -m ward

16:15:15 - Loaded 60 trajs 16:15:15 - Preparing... 16:15:17 - Getting all to all pairwise distance matrix... Traceback (most recent call last): File "/vcp2/student/trayder/anaconda/bin/Cluster", line 9, in load_entry_point('msmbuilder==2.8.2', 'console_scripts', 'Cluster')() File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/msmbuilder-2.8.2-py2.7.egg/msmbuilder/scripts/Cluster.py", line 287, in entry_point main(args, metric) File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/msmbuilder-2.8.2-py2.7.egg/msmbuilder/scripts/Cluster.py", line 257, in main clusterer = cluster(metric, trajectories, ptrajs, args, extra_kwargs) File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/msmbuilder-2.8.2-py2.7.egg/msmbuilder/scripts/Cluster.py", line 189, in cluster method=args.hierarchical_method) File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/msmbuilder-2.8.2-py2.7.egg/msmbuilder/clustering.py", line 769, in init** dmat = metric.all_pairwise(pflat_trajectory) File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/msmbuilder-2.8.2-py2.7.egg/msmbuilder/metrics/baseclasses.py", line 343, in all_pairwise V=self.V, VI=self.VI) File "/vcp2/student/trayder/anaconda/lib/python2.7/site-packages/scipy/spatial/distance.py", line 1182, in pdist raise ValueError('A 2-dimensional array must be passed.') ValueError: A 2-dimensional array must be passed.`