In [28]:
%reload_ext autoreload
%autoreload 2
%matplotlib inline
from salicon.salicon import SALICON
from saliconeval.eval import SALICONEval
import matplotlib.pyplot as plt
import as io
import pylab
pylab.rcParams['figure.figsize'] = (10.0, 8.0)
In [29]:
import json
from json import encoder
encoder.FLOAT_REPR = lambda o: format(o, '.3f')
In [30]:
algName = 'fake'
subtypes=['results', 'evalImgs', 'eval']
[resFile, evalImgsFile, evalFile]= ['%s/results/salmaps_%s_%s_%s.json'%(dataDir,dataType,algName,subtype) for subtype in subtypes]
In [31]:
# create coco object and cocoRes object
salicon = SALICON(annFile)
saliconRes = salicon.loadRes(resFile)
loading annotations into memory...
creating index...
index created!
Loading and preparing results...     
DONE (t=0.39s)
creating index...
index created!
In [33]:
# create cocoEval object by taking coco and cocoRes
saliconEval = SALICONEval(salicon, saliconRes)

# evaluate on a subset of images by setting
# cocoEval.params['image_id'] = cocoRes.getImgIds()
saliconEval.params['image_id'] = saliconRes.getImgIds()

# evaluate results

# print output evaluation scores
print "Final Result for each Metric:"
for metric, score in saliconEval.eval.items():
    print '%s: %.3f'%(metric, score)
setting up scorers...
computing NSS score...
NSS: 0.011
Final Result for each Metric:
NSS: 0.011
In [ ]: