def confusion_matrix(log, plcf, col):
"""generate type confusion matrix"""
ins_lst = weka.log_parse(log)
classes = set()
places = dataset.DataItem()
with open(plcf) as fplc:
for line in fplc:
place = json.loads(line)
places[place[col]] = place
classes.add(place[col])
print classes
cmat = dict()
for ref in classes:
cmat[ref] = dict()
for prd in classes:
cmat[ref][prd] = list()
for ins in ins_lst:
ref = places[ins['refN']][col]
hyp = places[ins['prdN']][col]
cmat[ref][hyp].append(int(ins['id']))
return cmat
def thirgest(log):
"""The accuracy in first three"""
threshold = 3
ins_lst = weka.log_parse(log)
pos, cnt = 0, 0
for ins in ins_lst:
#print ins['score']
rnk = sorted(zip(ins['score'], range(1, len(ins['score']) + 1)), \
key=lambda x: x[0], reverse=True)
#print rnk
for i in range(threshold):
if rnk[i][1] == ins['ref']:
pos += 1
break
#print pos
#if cnt>10: return
cnt += 1
return pos/float(cnt)
