def get_predictions_from_vector(lang, concepts, cat, count):
count = 100
tensor = get_tensor(lang)
svd = get_svd_results(lang)
items = svd.v_dotproducts_with(cat).top_items(count)
print len(items)
for feature, score in items:
for concept in concepts:
print concept, feature, score
# Exclude items that are already in the database.
# FIXME: check tensor format
if (concept, feature) in tensor: continue
print 'not in tensor'
if not isinstance(feature, tuple): continue
f = Feature.from_tuple(feature)
if f.relation == u'InheritsFrom': continue
print f
prop = f.fill_in(concept)
print prop
# Exclude self-relations.
if prop.concept1 == prop.concept2: continue
yield prop, score
def start_lattice(request, concept_name, lr, relation_name, fconcept_name):
from csc.conceptnet4.models import Concept, Feature
concept = Concept.get(concept_name, 'en')
# Build the concept tree
edges = concept.get_tree()
session_data = dict(
concept=concept,
feature=Feature.from_tuple((lr, relation_name, fconcept_name)),
edges=edges,
)
# Store the new session data, overwriting anything that was there before.
request.session['lattice_learning'] = session_data
# Render the graphviz.
src = as_graphviz(edges)
# Go to the initial view.
g = GVGraph(src, engine='dot')
return HttpResponse('''
<html><head><title>Test</title></head><body>
<img src="%(src)s" usemap="#%(map_name)s">%(map)s
</body></html>''' % dict(src='/lattice/graph_images/%s/' % g.key,
map_name=g.map_name,
map=g.map))
def feature_to_dict(feature_tup, score):
feature = Feature.from_tuple(feature_tup)
return dict(
raw = feature_tup,
logical = str(feature),
text = feature.nl_statement('__'),
score = score
)
def predictions_for_concept(lang, concept, count):
svd = get_svd_results(lang)
if concept not in svd.u.label_list(0): return
tensor = get_tensor(lang)
predictions = predict_features(svd, concept).top_items(count*5)
for feature, score in predictions:
if (concept, feature) in tensor: continue
if not isinstance(feature, tuple): continue
f = Feature.from_tuple(feature)
if f.relation.name == u'InheritsFrom': continue
prop = f.fill_in(concept)
if prop.concept1 == prop.concept2: continue
yield prop, score
def get_predictions(lang, concepts):
count = 100
tensor = get_tensor(lang)
svd = get_svd_results(lang)
cat = make_category_failsoft(svd, concepts, [], concepts)
items = svd.v_distances_to(cat).top_items(count)
for feature, score in items:
for concept in concepts:
# Exclude items that are already in the database.
# FIXME: check tensor format
if (concept, feature) in tensor: continue
f = Feature.from_tuple(feature)
prop = f.fill_in(concept)
# Exclude self-relations.
if prop.concept1 == prop.concept2: continue
yield prop, score