def __init__(self):
self.conf = config_parser.ParseConfig()
self.data_set_loc = self.conf.config_section_mapper("filePath").get("data_set_loc")
self.data_extractor = data_extractor.DataExtractor(self.data_set_loc)
self.mlmovies = self.data_extractor.get_mlmovies_data()
self.mltags = self.data_extractor.get_mltags_data()
self.mlmovies = self.data_extractor.get_mlmovies_data()
self.mlratings = self.data_extractor.get_mlratings_data()
self.combined_data = self.get_combined_data()
self.util = Util()
self.reshuffle = False
def __init__(self, user_id):
self.user_id = user_id
self.conf = config_parser.ParseConfig()
self.data_set_loc = self.conf.config_section_mapper("filePath").get(
"data_set_loc")
self.data_extractor = data_extractor.DataExtractor(self.data_set_loc)
self.feedback_data = self.get_feedback_data()
self.util = Util()
self.movies_dict = {}
self.tag_dict = {}
self.movie_tag_matrix = self.get_movie_tag_matrix()
self.feedback_metadata_dict = {}
def __init__(self):
self.conf = config_parser.ParseConfig()
self.data_set_loc = self.conf.config_section_mapper("filePath").get(
"data_set_loc")
self.data_extractor = data_extractor.DataExtractor(self.data_set_loc)
self.util = Util()
self.movies_dict = {}
self.movie_tag_matrix = self.get_movie_tag_matrix()
self.task_3_input = self.data_extractor.get_lsh_details()
self.movieLSH = MovieLSH(self.task_3_input["num_layers"],
self.task_3_input["num_hashs"])
(self.query_df, self.query_vector) = self.fetch_query_vector_from_csv()
self.movieLSH.create_index_structure(self.task_3_input["movie_list"])
SENTENCES_PER_SPEAKER = 10
FREQ_ELEMENTS = 128
MAX_SPEAKERS = 5
WITH_SPLIT = True
SPEAKER_LIST = '../../data/speaker_lists/speakers_5_clustering_vs_reynolds_v3.txt'
#OUTPUT_1 = '../../data/training/TIMIT_extracted/train_speakers_5_clustering_vs_reynolds_v3.pickle'
#OUTPUT_2 = '../../data/training/TIMIT_extracted/test_speakers_5_clustering_vs_reynolds_v3.pickle'
OUTPUT_1 = '../../data/training/TIMIT_extracted/train_speakers_5_clustering_vs_reynolds.pickle'
OUTPUT_2 = '../../data/training/TIMIT_extracted/test_speakers_5_clustering_vs_reynolds.pickle'
ONE_SEC = 100 # array elements corresponding to one sec
STEP_SIZE = 100
SPECT_DIMENSION = (FREQ_ELEMENTS, ONE_SEC)
MAX_AUDIO_LENGTH = 100
extractor = de.DataExtractor(MAX_SPEAKERS, ONE_SEC, STEP_SIZE, SPECT_DIMENSION)
#X = np.zeros((MAX_SPEAKERS*1000, 1, SPECT_DIMENSION[0], SPECT_DIMENSION[1]), dtype=np.float32)
X = np.zeros((MAX_SPEAKERS * 20, 1, SPECT_DIMENSION[0], MAX_AUDIO_LENGTH),
dtype=np.float32)
y = np.zeros(MAX_SPEAKERS * 20, dtype=np.int32)
valid_speakers = []
with open(SPEAKER_LIST, 'rb') as f:
for line in f:
valid_speakers.append(line.replace('\n', ''))
X, y, speaker_names = extractor.traverse_TIMIT_data(
'../../data/training/TIMIT/', X, y, valid_speakers)
print X.shape
import data_extractor as data_extractor
import data_transformer as data_transformer
import data_loader as data_loader
import config
# This component is the main program for the requested task
# It uses a configuration file (config.py) for input, output and transformation configurations
# defining multiple configurations is possible in config.py.
# runs the transfer operation for each configuration.
for configNo in range(0, config.CONFIGURATIONS_COUNT):
# get the proper implementations of extractor, transformer and loader based on the configurations
extractor = data_extractor.DataExtractor(
config.INPUT_TYPE[configNo], config.INPUT_DATA_CONNECTION[configNo])
transformer = data_transformer.DataTransformer(
config.INPUT_TYPE[configNo], config.OUTPUT_TYPE[configNo],
config.INPUT_TO_OUTPUT_MAPPING[configNo])
loader = data_loader.DataLoader(config.OUTPUT_TYPE[configNo],
config.OUTPUT_DATA_CONNECTION[configNo],
config.OUTPUT_DB[configNo],
config.OUTPUT_COLLECTION[configNo])
counter = 1
# loop through the data extracted from the input source
for item in extractor.extract_data():
# transform the item into desired format
out_item = transformer.transform_data(item)
# insert the transformed item into output destination
insert_id = loader.load_data(out_item)
print("Created item {0} with id {1}".format(counter, insert_id))
counter += 1
import argparse
import operator
from collections import Counter
import config_parser
import data_extractor
import numpy
from scipy.spatial import distance as dist
from sklearn.svm import SVC
from util import *
util = Util()
conf = config_parser.ParseConfig()
data_set_loc = conf.config_section_mapper("filePath").get("data_set_loc")
data_extractor_obj = data_extractor.DataExtractor(data_set_loc)
movie_tag_frame = util.get_movie_tag_matrix()
movie_tag_matrix_value = movie_tag_frame.values
(U, s, Vh) = util.SVD(movie_tag_matrix_value)
movie_latent_matrix = U[:, :10]
movies = list(movie_tag_frame.index.values)
tags = list(movie_tag_frame)
label_movies_json_data = data_extractor_obj.get_json()
class Classifier(object):
def __init__(self, r=0):
self.util = util
self.r = r
self.movie_latent_matrix = movie_latent_matrix