如何在python中保存词袋数据？

Question

Here is my code :

sift=cv2.xfeatures2d.SIFT_create()
descriptors_unclustered=[]
dictionarysize=800
BOW=cv2.BOWKmeansTrainer(dictionarysize)
for p in training-paths :
    kp,dsc=sift.detectAndCompute(image,None)
    BOW.add(dsc)

dictionary=BOW.cluster()
bowdiction=cv2.BOWImgDescriptorExtractor(sift, cv2.BFMatcher(cv2.NORM_L2))
bowdiction.setvocabulary(dictionary)

I want to save this bowdiction data to use it later. I dont want to wait every time for these calculations so how can I save this data ?

Answer 1

Use pickle for this

1. Save BOW to pickle:

    import pickle sift=cv2.xfeatures2d.SIFT_create() descriptors_unclustered=[] dictionarysize=800 BOW=cv2.BOWKmeansTrainer(dictionarysize) for p in training-paths : kp,dsc=sift.detectAndCompute(image,None) BOW.add(dsc) with open('bow_pickle.pickle', 'wb') as f: pickle.dump(f)

2. Return the data from pickle:

    import pickle with open('bow_pickle.pickle', 'rb') as f: BOW = pickle.load(f) dictionary=BOW.cluster() bowdiction=cv2.BOWImgDescriptorExtractor(sift,cv2.BFMatcher(cv2.NORM_L2)) bowdiction.setvocabulary(dictionary)

Answer 2

I know this is old. But as I came here and saw no answer, I tried a few things and this worked:

dictionary=BOW.cluster() is what takes most time. So you only need to save dictionary. Which is just an ndarray:

np.savetxt('bow_dict.txt', dictionary)

loaded_dictionary = np.loadtxt('bow_dict.txt')

And then proceed with

bowdiction=cv2.BOWImgDescriptorExtractor(sift,cv2.BFMatcher(cv2.NORM_L2))
bowdiction.setvocabulary(loaded_dictionary)