def averageSimilarityMatrix(dictOfClassesLocal, dictOfWeightsLocal,title="Cluster similarity matrix",savePlot=False):
global bar, progressCount, dictOfClasses, dictOfWeights, arguments, distances
dictOfClasses = dictOfClassesLocal
dictOfWeights = dictOfWeightsLocal
print "Computing similarity matrix..."
bar = progressbar.ProgressBar(maxval=len(dictOfClasses.keys())**2, widgets=[progressbar.Bar('=', '[', ']'), ' ', progressbar.Percentage()])
bar.start()
progressCount = 0
# Initialise task matrix
arguments = []
distances = []
for i,k in zip(range(len(dictOfClasses.keys())),reversed(sorted(dictOfClasses.keys()))):
arguments.append([])
distances.append([])
for j in sorted(dictOfClasses.keys()):
arguments[i].append((dictOfClasses[k],dictOfClasses[j],dictOfWeights[k],dictOfWeights[j]))
distances[i].append([])
# Distribute tasks
poo = Pool()
for i in range(len(dictOfClasses.keys())**2):
poo.apply_async(interClassDistance,(i,),callback=updateResultMatrix)
poo.close()
poo.join()
bar.finish()
plot.plotSimilarityMatrix(distances,sorted(dictOfClasses.keys()),title,savePlot)
def dtwDirectoryAgainstItself(dir_name):
inputFiles = glob.glob("%s/*.csv"%dir_name)
distances = []
for i in range(len(inputFiles)):
distances.append([])
for j in range(len(inputFiles)):
distances[i].append(dtw.getDTWdist(inputFiles[i],inputFiles[j]))
plot.plotSimilarityMatrix(distances,[f.split('/')[-1].split('.')[0] for f in inputFiles])
data = pandas.DataFrame(np.array(distances), columns=[f.split('/')[-1].split('.')[0] for f in inputFiles])
square_table(data)
plt.show()
def similarityMatrixWeighted(segments,segmentNames,weights,title,savePlot=False):
print "Constructing similarity matrix"
bar = progressbar.ProgressBar(maxval=len(segments)**2, widgets=[progressbar.Bar('=', '[', ']'), ' ', progressbar.Percentage()])
bar.start()
progressCount = 0
distances = []
for i,k in zip(range(len(segments)),reversed(range(len(segments)))):
distances.append([])
for j in range(len(segments)):
distances[i].append(dtw.getDTWdist2DweightedSum(segments[k],segments[j],weights))
progressCount+=1
bar.update(progressCount)
bar.finish()
plot.plotSimilarityMatrix(distances,segmentNames,title,savePlot)
def averageSimilarityMatrix(dictOfClasses, dictOfWeights=None,title,savePlot=False):
print "Constructing similarity matrix"
if sorted(dictOfClasses.keys()) != sorted(dictOfWeights.keys()):
import sys
sys.exit("Mismatching keys between weights and classes to compute average similarity matrix")
bar = progressbar.ProgressBar(maxval=len(dictOfClasses.keys())**2, widgets=[progressbar.Bar('=', '[', ']'), ' ', progressbar.Percentage()])
bar.start()
progressCount = 0
distances = []
for i,k in zip(range(len(dictOfClasses.keys())),reversed(sorted(dictOfClasses.keys()))):
distances.append([])
for j in sorted(dictOfClasses.keys()):
if dictOfWeights == None:
distances[i].append(interClassDistance(dictOfClasses[k],dictOfClasses[j]))
else:
distances[i].append(interClassDistance(dictOfClasses[k],dictOfClasses[j],dictOfWeights[k],dictOfWeights[j]))
progressCount+=1
bar.update(progressCount)
bar.finish()
plot.plotSimilarityMatrix(distances,sorted(dictOfClasses.keys()),title,savePlot)
