Commit f233de53 authored by delanoe's avatar delanoe

[FIX] grouped ngrams not in the mapList.

parent ec78c646
...@@ -60,17 +60,27 @@ def compute_mapList(corpus,limit=500,n=1): ...@@ -60,17 +60,27 @@ def compute_mapList(corpus,limit=500,n=1):
.order_by(desc(Spec.score)) .order_by(desc(Spec.score))
.limit(multigrams_limit) .limit(multigrams_limit)
) )
stop_ngrams = (session.query(NodeNgram.ngram_id)
.filter(NodeNgram.node_id == node_stop.id)
.all()
)
grouped_ngrams = (session.query(NodeNgramNgram.ngramy_id)
.filter(NodeNgramNgram.node_id == node_group.id)
.all()
)
#print([t for t in top_ngrams]) #print([t for t in top_ngrams])
node_mapList = get_or_create_node(nodetype='MapList', corpus=corpus) node_mapList = get_or_create_node(nodetype='MapList', corpus=corpus)
session.query(NodeNgram).filter(NodeNgram.node_id==node_mapList.id).delete() session.query(NodeNgram).filter(NodeNgram.node_id==node_mapList.id).delete()
session.commit() session.commit()
data = zip( data = zip(
[node_mapList.id for i in range(1,limit)] [node_mapList.id for i in range(1,limit)]
, [n[0] for n in list(top_multigrams) + list(top_monograms)] , [n[0] for n in list(top_multigrams) + list(top_monograms)
if (n[0],) not in list(stop_ngrams) + list(grouped_ngrams)
]
, [1 for i in range(1,limit)] , [1 for i in range(1,limit)]
) )
#print([d for d in data]) #print([d for d in data])
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment