Add python installation script.

Author: mdimura

CMakeLists.txt

@@ -22,7 +22,10 @@ target_link_libraries(sparsehc-dm ${LIBS})
 add_definitions(-std=c++11)
 
 file(GLOB SHARED_SOURCES src/averagecluster.cpp src/cluster.cpp src/common.cpp src/completecluster.cpp src/dendrogram.cpp src/inmatrix.cpp src/matrix.cpp src/singlecluster.cpp)
+add_custom_target(python_init COMMAND touch __init__.py)
 ADD_LIBRARY(sparsehc_dm SHARED python/sparsehc-dm_python.cpp ${SHARED_SOURCES})
+add_dependencies(sparsehc_dm python_init)
 target_link_libraries(sparsehc_dm ${LIBS})
 set_target_properties(sparsehc_dm PROPERTIES PREFIX "" SUFFIX ".so")
 
+install(CODE "execute_process(COMMAND python ../python/setup.py install --prefix=${CMAKE_INSTALL_PREFIX} WORKING_DIRECTORY ${CMAKE_CURRENT_BINARY_DIR})")

examples/aMD-148l-all_1.pdb

@@ -7,8 +7,10 @@
 import mdtraj as md
 import numpy as np
 
-N=int(sys.argv[1])
-traj_filename='aMD-148l-first{}k.nc'.format(N)
+Nk=1
+if len(sys.argv)>1:
+  Nk=int(sys.argv[1])
+traj_filename='aMD-148l-first{}k.nc'.format(Nk)
 top_filename='aMD-148l-all_1.pdb'
 
 first_frame = md.load_frame(traj_filename, 0,top=top_filename)
@@ -17,15 +19,15 @@
 start=time.time()
 traj=md.load(traj_filename,top=top_filename, atom_indices=atoms_to_keep)
 finishedLoad=time.time()
-print ("finished loading ({}k): {}".format(N,finishedLoad-start))
+print ("finished loading ({}k): {}".format(Nk,finishedLoad-start))
 
-N=traj.n_frames
-D=np.empty((N*(N-1)/2,))
+Nframes=traj.n_frames
+D=np.empty((Nframes*(Nframes-1)/2,))
 j=0
-for i in range(0,N-1):
+for i in range(0,Nframes-1):
   rmsds=md.rmsd(traj, traj, i)[i+1:]
-  D[j:j+N-i-1]=rmsds
-  j+=N-1-i
+  D[j:j+Nframes-i-1]=rmsds
+  j+=Nframes-1-i
 
 #D = np.empty((traj.n_frames, traj.n_frames))
 #for i in range(traj.n_frames):
@@ -34,15 +36,10 @@
 
 finishedRMSD=time.time()
 print ("finished rmsd: {}".format(finishedRMSD-finishedLoad))
+#Z-matrix contains the linkage history, see more at http://docs.scipy.org/doc/scipy/reference/generated/scipy.cluster.hierarchy.linkage.html#scipy.cluster.hierarchy.linkage
 Z=linkage(D, method='complete',preserve_input=False)
 finishedClust=time.time()
-print ("finished clustering: {}".format(finishedClust-finishedRMSD))
 
-#for r in Z:
-  #print("{}\t{}\t{}\t{}".format(int(r[0]),int(r[1]),r[2],int(r[3])))
-
-#plt.figure(101)
-#plt.subplot(1, 2, 1)
-#plt.title("ascending")
-#cluster.hierarchy.dendrogram(Z)
-#plt.show()
+#One would probably like to save the Z-matrix for the future, so that there is no need to redo the clustering
+open('z_sparsehc-dm_load_{}k.json'.format(Nk), 'w').write(json.dumps(Z))
+print ("finished clustering: {}".format(finishedClust-finishedRMSD))

examples/aMD-148l-first1k.nc

@@ -1,11 +1,13 @@
-import sparsehc_dm
+from sparsehc_dm import sparsehc_dm
 import random
 import time
 import numpy as np
 import mdtraj as md
 
-N=10000
-traj_filename='aMD-148l-first10k.nc'
+N=1
+if len(sys.argv)>1:
+  N=int(sys.argv[1])
+traj_filename='aMD-148l-first{}k.nc'.format(N)
 top_filename='aMD-148l-all_1.pdb'
 
 first_frame = md.load_frame(traj_filename, 0,top=top_filename)

examples/fastcluster_load.py

@@ -1,13 +1,16 @@
 import sys
 import gc
-import sparsehc_dm
+from sparsehc_dm import sparsehc_dm
 import random
 import time
 import mdtraj as md
 import numpy
+import json
 
-N=int(sys.argv[1])
-traj_filename='aMD-148l-first{}k.nc'.format(N)
+Nk=1
+if len(sys.argv)>1:
+  Nk=int(sys.argv[1])
+traj_filename='aMD-148l-first{}k.nc'.format(Nk)
 top_filename='aMD-148l-all_1.pdb'
 
 first_frame = md.load_frame(traj_filename, 0,top=top_filename)
@@ -18,28 +21,27 @@
 
 m=sparsehc_dm.InMatrix()
 finishedLoad=time.time()
-print ("finished loading ({}k): {}".format(N,finishedLoad-start))
+print ("finished loading ({}k): {}".format(Nk,finishedLoad-start))
 
-N=traj.n_frames
+Nframes=traj.n_frames
 rmsds=list()
-for i in range(0,N-1):
+for i in range(0,Nframes-1):
   rmsds=md.rmsd(traj, traj, i)[i+1:].tolist()
   sparsehc_dm.push(m,rmsds,i)
-  #if( (i+1)%20==0):
-    #sparsehc_dm.push(m,rmsds,i-19,N)
-    #rmsds=list()
 
 
-#for i in range(0,N-1):
+#for i in range(0,Nframes-1):
   #rmsds=md.rmsd(traj, traj, i)
-  #for j in range(i+1,N):
+  #for j in range(i+1,Nframes):
     #m.push(i,j,float(rmsds[j]))
 
 finishedRMSD=time.time()
 print ("finished rmsd: {}".format(finishedRMSD-finishedLoad))
 
+#Z-matrix contains the linkage history, see more at http://docs.scipy.org/doc/scipy/reference/generated/scipy.cluster.hierarchy.linkage.html#scipy.cluster.hierarchy.linkage
 Z=sparsehc_dm.linkage(m,"complete")
 finishedClust=time.time()
-print ("finished clustering: {}".format(finishedClust-finishedRMSD))
-#for r in Z:
-  #print("{}\t{}\t{}\t{}".format(r[0],r[1],r[2],r[3]))
+
+#One would probably like to save the Z-matrix for the future, so that there is no need to redo the clustering
+open('z_sparsehc-dm_load_{}.json'.format(Nk), 'w').write(json.dumps(Z))
+print ("finished clustering: {}".format(finishedClust-finishedRMSD))

examples/sparsehc-dm_interload.py

@@ -1,11 +1,9 @@
-from distutils.core import setup
-from distutils.extension import Extension
-setup(name="sparsehc-dm",    
-      ext_modules=[
-                    Extension("sparsehc-dm",["sparsehc-dm_python.cpp",],
-                    library_dirs=["/usr/local/lib",],
-                    libraries=["",],
-                    include_dirs=[".","../src"],
-                    depends=[]),
-                    ]
-     )
+from distutils.core import setup, Extension
+import os
+setup (name = 'sparsehc_dm',
+       version = '',
+       description = 'Python wrapper for sparsehc_dm with STXXL sorting integrated',
+       packages=['sparsehc_dm'],
+       package_dir={'sparsehc_dm':''},
+       package_data={'sparsehc_dm':['sparsehc_dm.so',]},
+       )