graphe of merge clusters

diff --git a/mergeclustergraph.py b/mergeclustergraph.py
new file mode 100644 (file)
index 0000000..7ca85b6
--- /dev/null
+++ b/mergeclustergraph.py
@@ -0,0 +1,98 @@
+import os
+import sys
+#path = '/home/pierre/workspace/iramuteq'
+#sys.path.append(path)
+#import iracmd
+from functions import DoConf, read_chd, ReadProfileAsDico
+from export_html import DoHTML
+from chemins import ConstructPathOut, ChdTxtPathOut, FFF, ffr, PathOut, StatTxtPathOut, simipath
+from layout import SimiLayout
+from textsimi import *
+from analyse_merge import AnalyseMerge
+
+#corpusin = '/home/pierre/fac/lerass/formation_iramuteq/giletsjaunes-2018-12-01/giletsjaunes-2018-12-01_corpus_1/Corpus.cira'
+#corpusin = "/home/pierre/fac/lerass/giletsjaunes/corpus_vrdb_vi_ss_et_corpus_1/Corpus.cira"
+#corpusin ="/home/pierre/fac/lerass/giletsjaunes/corpus_vrdb_vi_ss_et_pascal/Corpus.cira"
+#corpusin = '/home/pierre/fac/lerass/granddebat/20190321-LE-GRAND-DEBAT_corpus_1/Corpus.cira'
+#corpusin ='/home/pierre/fac/etudiant/grumet/M2/Corpus_focus_GRUMET_corpus_2/Corpus.cira'
+
+#cmd = iracmd.CmdLine(args=['-r',corpusin], AppliPath = path)
+
+#analysein = "/home/pierre/fac/lerass/formation_iramuteq/giletsjaunes-2018-12-01/giletsjaunes-2018-12-01_corpus_1/giletsjaunes-2018-12-01_alceste_2/Analyse.ira"
+#analysein = "/home/pierre/fac/lerass/giletsjaunes/corpus_vrdb_vi_ss_et_corpus_1/corpus_vrdb_vi_ss_et_alceste_1/Analyse.ira"
+#analysein ="/home/pierre/fac/lerass/giletsjaunes/corpus_vrdb_vi_ss_et_pascal/corpus_vrdb_vi_ss_et_alceste_1/Analyse.ira"
+#analysein = "/home/pierre/fac/lerass/granddebat/20190321-LE-GRAND-DEBAT_corpus_1/20190321-LE-GRAND-DEBAT_alceste_1/Analyse.ira"
+#analysein = "/home/pierre/fac/etudiant/grumet/M2/Corpus_focus_GRUMET_corpus_2/Corpus_focus_GRUMET[5682]_alceste_3/Analyse.ira"
+#parametres = DoConf(analysein).getoptions()
+
+#actives = cmd.corpus.make_actives_nb(10000, 1)[0]
+#actives = dict(zip(actives,actives))
+#print len(actives)
+#print cmd.ConfigPath
+
+
+class MergeClusterGraph :
+    def __init__(self, ira, corpus, parametres) :
+        self.ira = ira
+        self.corpus = corpus
+        self.parametres = parametres
+        self.pathout = PathOut(parametres['ira'])
+        self.pathout.basefiles(ChdTxtPathOut)
+        self.corpus.make_ucecl_from_R(self.pathout['uce'])
+        self.encoding = self.parametres['encoding']
+        self.clnb = parametres['clnb']
+        dictprofile = ReadProfileAsDico(self.pathout['PROFILE_OUT'], True, self.encoding)
+        self.dograph(dictprofile)
+
+    def dograph(self, dictprofile) :
+        tomerge = []
+        #OUTCH!!First cluster removed !!
+        #print 'ELMINATION CLUSTER 1'
+        for i in range(0, self.clnb):
+            self.pathout = PathOut(self.parametres['ira'])
+            simiparam = DoConf(self.ira.ConfigPath['simitxt']).getoptions()
+            simiparam['coeff'] = 3
+            simiparam['cexfromchi'] = True
+            profclasse = dictprofile[`i+1`]
+            line1 = profclasse.pop(0)
+            classen = [line for line in profclasse if line[0] != '*' and line[0] != '*****']
+            try :
+                self.lenact = profclasse.index([u'*****', u'*', u'*', u'*', u'*', u'*', '', ''])
+                profclasse.pop(self.lenact)
+            except ValueError:
+                try :
+                    self.lenact = profclasse.index([u'*', u'*', u'*', u'*', u'*', u'*', '', ''])
+                    profclasse.pop(self.lenact)
+                except ValueError:
+                    self.lenact = len(profclasse)
+            try :
+                self.lensup = profclasse.index([u'*', u'*', u'*', u'*', u'*', u'*', '', ''])
+                self.lensup = self.lensup - self.lenact
+                profclasse.pop(self.lensup)
+            except ValueError:
+                self.lensup = len(profclasse) - self.lenact
+            self.lenet = len(profclasse) - (self.lenact + self.lensup)
+#            print self.lenact, self.lensup, self.lenet
+            for l,  line in enumerate(classen) :
+                line[0] = l
+            dictdata = dict(zip([l for l in range(0,len(classen))], classen))
+
+            if self.lenact != 0 :
+                self.la = [dictdata[l][6] for l in range(0, self.lenact)]
+                self.lchi = [dictdata[l][4] for l in range(0, self.lenact)]
+                self.lfreq = [dictdata[l][1] for l in range(0, self.lenact)]
+            else :
+                self.la = []
+                self.lchi = []
+                self.lfreq = []
+            print 'cluster : ', i
+            simi = SimiFromCluster(self.ira, self.corpus, self.la, self.lfreq,
+                            self.lchi, i, parametres = simiparam, limit=100)
+            tomerge.append(simi.parametres['ira'])
+            print tomerge
+        newparam = {'type': 'merge', 'fileout' : '/tmp/test.txt'}
+        newparam['graphs'] = tomerge
+        AnalyseMerge(self.ira, newparam, dlg=None)
+
+
+#MergeClusterGraph(cmd, cmd.corpus, parametres)