[FIX] Group asymetrics and graph parameters: note, too many specific nodes at this stage.

analysis/cooccurrences.py

@@ -57,7 +57,7 @@ def cooc(corpus=None
     doc_id = cache.NodeType['Document'].id
 
     cooc_query = (session.query(NodeNgramX.ngram_id, NodeNgramY.ngram_id, 
-                    func.sqrt(func.sum(NodeNgram.weight) * func.sum(NodeNgraY.wei)))
+                    func.sqrt(func.sum(NodeNgramX.weight) * func.sum(NodeNgramY.weight)))
              .join(Node, Node.id == NodeNgramX.node_id)
              .join(NodeNgramY, NodeNgramY.node_id == Node.id)
              .filter(Node.parent_id==corpus.id, Node.type_id==doc_id)

analysis/functions.py

@@ -229,10 +229,10 @@ def get_cooc(request=None, corpus=None, cooc_id=None, type='node_link', size=siz
     #print(n)
     #print(m)
 
-    nodes_included = 300 #int(round(size/20,0))
+    nodes_included = 333 #int(round(size/20,0))
     #nodes_excluded = int(round(size/10,0))
 
-    nodes_specific = 300 #int(round(size/10,0))
+    nodes_specific = 333 #int(round(size/10,0))
     #nodes_generic = int(round(size/10,0))
 
     # TODO user the included score for the node size

ngram/group.py

@@ -137,9 +137,10 @@ def compute_groups(corpus, limit_inf=None, limit_sup=None, how='Stem'):
         miam_to_insert.add((miam_node.id, n[0],1))
         #print([n for n in group])
         for g in group:
+            if (miam_node.id, g[0],1) not in miam_to_insert:
 #list_to_check.remove(g)
-            group_to_insert.append((node_group.id, n[0], g[0], 1))
-            print(n[1], "=", g[1])
+                group_to_insert.append((node_group.id, n[0], g[0], 1))
+                print(n[1], "=", g[1])
 
 # Deleting previous groups
     session.query(NodeNgramNgram).filter(NodeNgramNgram.node_id == node_group.id).delete()

ngram/workflow.py

@@ -26,7 +26,7 @@ def ngram_workflow(corpus, n=5000):
 #    compute_stop(corpus)
     compute_groups(corpus,limit_inf=part, limit_sup=n)
     
-    compute_miam(corpus,limit=part) # size
+#    compute_miam(corpus,limit=part) # size
     
     compute_tfidf(corpus)