예제 #1
0
def main():
  args = parse_options()

  client = Regression('127.0.0.1', 9199, '')

  # train
  num = 0
  if args.traindata:
    with open(args.traindata, 'r') as traindata:
      for data in traindata:

        # skip comments
        if not len(data) or data.startswith('#'):
          continue
        num += 1

        rent, distance, space, age, stair, aspect = map(str.strip, data.strip().split(','))
        d = Datum({
            'aspect': aspect,
            'distance': float(distance),
            'space': float(space),
            'age': float(age),
            'stair': float(stair) })
        train_data = [[float(rent), d]]

        # train
        client.train(train_data)

    # print train number
    print 'train ...', num

  # anaylze
  with open(args.analyzedata, 'r') as analyzedata:
    for data in analyzedata:
      if not len(data) or data.startswith('#'):
        continue
      distance, space, age, stair, aspect = map(str.strip, data.strip().split(','))

      d = Datum({
        'aspect': str(aspect),
        'distance': float(distance),
        'space': float(space),
        'age': float(age),
        'stair': float(stair)
      })

      analyze_data = [d]
      result = client.estimate(analyze_data)

      print 'rent ....', round(result[0], 1)
예제 #2
0
def main():
    cl = Regression("127.0.0.1", 9199, "test")
    
    d = Datum()
    for i in xrange(10):
        d.add_number('x', 1)
        d.add_number('y', 4)
        cl.train([[10.0, d]])

    d = Datum()
    d.add_number('x', 1)
    d.add_number('y', 4)
    result = cl.estimate([d])
    print("{0:30.30f}".format(result[0]))
예제 #3
0
class RegressionTest(unittest.TestCase):
    def setUp(self):
        self.config = {
            "method": "PA",
            "converter": {
                "string_filter_types": {},
                "string_filter_rules": [],
                "num_filter_types": {},
                "num_filter_rules": [],
                "string_types": {},
                "string_rules": [{"key": "*", "type": "str", "sample_weight": "bin", "global_weight": "bin"}],
                "num_types": {},
                "num_rules": [{"key": "*", "type": "num"}]
                },
            "parameter": {
                "sensitivity" : 0.1,
                "regularization_weight" : 3.402823e+38
                }
            }

        TestUtil.write_file('config_regression.json', json.dumps(self.config))
        self.srv = TestUtil.fork_process('regression', port, 'config_regression.json')
        try:
            self.cli = Regression(host, port, "name")
        except:
            TestUtil.kill_process(self.srv)
            raise

    def tearDown(self):
        if self.cli:
            self.cli.get_client().close()
        TestUtil.kill_process(self.srv)

    def test_get_client(self):
        self.assertTrue(isinstance(self.cli.get_client(), msgpackrpc.client.Client))

    def test_get_config(self):
        config = self.cli.get_config()
        self.assertEqual(json.dumps(json.loads(config), sort_keys=True), json.dumps(self.config, sort_keys=True))

    def test_train(self):
        d = Datum({"skey1": "val1", "skey2": "val2", "nkey1": 1.0, "nkey2": 2.0})
        data = [[1.0, d]]
        self.assertEqual(self.cli.train(data), 1)

    def test_estimate(self):
        d = Datum({"skey1": "val1", "skey2": "val2", "nkey1": 1.0, "nkey2": 2.0})
        data = [d]
        result = self.cli.estimate(data)

    def test_save(self):
        self.assertEqual(len(self.cli.save("regression.save_test.model")), 1)

    def test_load(self):
        model_name = "regression.load_test.model"
        self.cli.save(model_name)
        self.assertEqual(self.cli.load(model_name), True)

    def test_get_status(self):
        self.cli.get_status()
예제 #4
0
def main():
  args = parse_options()

  client = Regression('127.0.0.1', 9199, '')

  # train
  num = 0
  if args.traindata:
    with open(args.traindata, 'r') as traindata:
      for data in traindata:

        # skip comments
        if not len(data) or data.startswith('#'):
          continue
        num += 1

        rent, distance, space, age, stair, aspect = map(str.strip, data.strip().split(','))
        d = Datum({
            'aspect': aspect,
            'distance': float(distance),
            'space': float(space),
            'age': float(age),
            'stair': float(stair) })
        train_data = [[float(rent), d]]

        # train
        client.train(train_data)

    # print train number
    print 'train ...', num

  # anaylze
  with open(args.analyzedata, 'r') as analyzedata:
    myhome = yaml.load(analyzedata)
    d = Datum({
        'aspect': str(myhome['aspect']),
        'distance': float(myhome['distance']),
        'space': float(myhome['space']),
        'age': float(myhome['age']),
        'stair': float(myhome['stair'])
        })
    analyze_data = [d]
    result = client.estimate(analyze_data)

    print 'rent ....', round(result[0], 1)
예제 #5
0
def train(training_data):
    client = Regression("127.0.0.1", 9199, "boston")
    client.clear()
    for data in training_data:
        score, d = make_datum(data)
        client.train([[score, d]])
if __name__ == '__main__':
    options, remainder = parse_args()

    client = Regression(options.server_ip, options.server_port, '')

    print(client.get_config())
    print(client.get_status())

    train_list = cross_validation_python()
    data_train, data_test, label_train, label_test = train_test_split(train_list[0], train_list[1])

    for label, dat in izip(label_train, data_train):
        print(dat[0])
        data_dict = json.loads(dat[0])
        datum = Datum(data_dict)
        client.train([[float(label), datum]])

    print(client.get_status())

    print(client.save("tutorial_regression"))

    print(client.load("tutorial_regression"))

    print(client.get_config())

    count_ok = 0
    count_ng = 0
    for label, dat in izip(label_test, data_test):
        data_dict = json.loads(dat[0])
        datum = Datum(data_dict)
        ans = client.estimate([datum])
예제 #7
0
class RegressionTest(unittest.TestCase):
    def setUp(self):
        self.config = {
            "method": "PA1",
            "converter": {
                "string_filter_types": {},
                "string_filter_rules": [],
                "num_filter_types": {},
                "num_filter_rules": [],
                "string_types": {},
                "string_rules": [{
                    "key": "*",
                    "type": "str",
                    "sample_weight": "bin",
                    "global_weight": "bin"
                }],
                "num_types": {},
                "num_rules": [{
                    "key": "*",
                    "type": "num"
                }]
            },
            "parameter": {
                "sensitivity": 0.1,
                "regularization_weight": 3.402823e+38
            }
        }

        TestUtil.write_file('config_regression.json', json.dumps(self.config))
        self.srv = TestUtil.fork_process('regression', port,
                                         'config_regression.json')
        try:
            self.cli = Regression(host, port, "name")
        except:
            TestUtil.kill_process(self.srv)
            raise

    def tearDown(self):
        if self.cli:
            self.cli.get_client().close()
        TestUtil.kill_process(self.srv)

    def test_get_client(self):
        self.assertTrue(
            isinstance(self.cli.get_client(), msgpackrpc.client.Client))

    def test_get_config(self):
        config = self.cli.get_config()
        self.assertEqual(json.dumps(json.loads(config), sort_keys=True),
                         json.dumps(self.config, sort_keys=True))

    def test_train(self):
        d = Datum({
            "skey1": "val1",
            "skey2": "val2",
            "nkey1": 1.0,
            "nkey2": 2.0
        })
        data = [[1.0, d]]
        self.assertEqual(self.cli.train(data), 1)

    def test_estimate(self):
        d = Datum({
            "skey1": "val1",
            "skey2": "val2",
            "nkey1": 1.0,
            "nkey2": 2.0
        })
        data = [d]
        result = self.cli.estimate(data)

    def test_save(self):
        self.assertEqual(len(self.cli.save("regression.save_test.model")), 1)

    def test_load(self):
        model_name = "regression.load_test.model"
        self.cli.save(model_name)
        self.assertEqual(self.cli.load(model_name), True)

    def test_get_status(self):
        self.cli.get_status()