dict_fp = "../data/sdg_v2.dict"
corpus_fp = "../data/sdg_v2.mm"
model_fp = "../data/sdg_v2.lda_model"
# loading dictionary, corpus, lda_model
dictionary = corpora.Dictionary.load(dict_fp)
corpus = corpora.MmCorpus(corpus_fp)
lda = models.ldamodel.LdaModel.load(model_fp)
cosSim = CosineSimilarity(dictionary=dictionary, corpus=corpus, model=lda)
cosSim.save_index("../data/sdg_v2.index")
# loading undaf data
undaf_fp = "../data/undaf_v2.json"
p = Preprocessor(undaf_fp)
p.loadjson()
undaf_docs = p.json
# loading metatopic data
sdg = Preprocessor(clean_data_fp)
sdg.loadjson()
sdg_metatopics = sdg.json[0].keys()
for topic in sdg_metatopics:
print topic
for k, v in p.json[0].items():
simScore = cosSim.calculate_sim(v, 10)
print "\n", k
{"y": 0.089070708, "x": 0, "doc": 2},
def preprocess(input_fp, output_fp):
p = Preprocessor(input_fp)
p.loadjson()
p.processjson()
p.savejson(output_fp)
dict_fp = '../data/sdg_v2.dict'
corpus_fp = '../data/sdg_v2.mm'
model_fp = '../data/sdg_v2.lda_model'
# loading dictionary, corpus, lda_model
dictionary = corpora.Dictionary.load(dict_fp)
corpus = corpora.MmCorpus(corpus_fp)
lda = models.ldamodel.LdaModel.load(model_fp)
cosSim = CosineSimilarity(dictionary=dictionary, corpus=corpus, model=lda)
cosSim.save_index('../data/sdg_v2.index')
# loading undaf data
undaf_fp = '../data/undaf_v2.json'
p = Preprocessor(undaf_fp)
p.loadjson()
undaf_docs = p.json
# loading metatopic data
sdg = Preprocessor(clean_data_fp)
sdg.loadjson()
sdg_metatopics = sdg.json[0].keys()
for topic in sdg_metatopics:
print topic
for k, v in p.json[0].items():
simScore = cosSim.calculate_sim(v, 10)
print '\n', k
{'y': 0.089070708, 'x': 0, 'doc': 2},
def preprocess(input_fp, output_fp):
p = Preprocessor(input_fp)
p.loadjson()
p.processjson()
p.savejson(output_fp)
