testMovies.py 文件源码

def readData(self,filename):
        file = open(filename, 'r')
        line = file.readline()
        print line
        gson = json.loads(line)
        file.close()
        vector = []
        input = []
        labels = []
        titles = []
        #indice = 0
        for elem in gson:
            #titles.append(gson.get(elem).get("title"))
            actors = gson.get(elem).get("actors")
            directors = gson.get(elem).get("director")
            writers = gson.get(elem).get("writer")
            input.append([actors,directors,writers])
            #imdbRating = float(gson.get(elem).get("imdbRating"))
            mediaAct, mediaDir, mediaWri = self.calcolaMedie(actors, directors, writers)
            vect = [1,mediaAct, mediaDir, mediaWri]
            vector.append(vect)
            #labels.append(int(imdbRating))  ## CAST PER CLASSI DISCRETE ##
        data = np.array(vector)
        #labels = np.array(labels)
        #train_data,test_data,train_labels,test_labels = train_test_split(data,labels, train_size= 0.5)
        #return train_data, train_labels,test_data,test_labels
        print "lettura terminata"
        return data,input