BoxE/ModelOptions.py at master · ralphabb/BoxE · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
import Cnst


class ModelOptions:
    def __init__(self):

        self.embedding_dim = 64

        self.neg_sampling_opt = Cnst.SELFADV

        self.adversarial_temp = 1.0
        self.nb_neg_examples_per_pos = 25
        self.loss_margin = 3.0
        self.loss_fct = Cnst.POLY_LOSS
        self.obj_fct = Cnst.NEG_SAMP
        self.batch_size = 1024
        self.loss_norm_ord = 2
        self.regularisation_lambda = 0
        self.regularisation_points = 0
        self.total_log_box_size = -5
        self.hard_total_size = False
        self.hard_code_size = False


        self.learning_rate = 10 ** -2
        self.ranking_function = Cnst.DIST_RANK
        self.stop_gradient = Cnst.NO_STOPS


        self.restricted_training = False
        self.restriction = 1024

        self.stop_gradient_negated = False
        self.use_bumps = True


        self.shared_shape = False
        self.learnable_shape = True
        self.fixed_width = False


        self.bounded_pt_space = True
        self.bounded_box_space = True
        self.space_bound = 1.0

        self.dim_dropout_prob = 0.0
        self.use_tensorboard = True
        self.gradient_clip = -1.0

        self.bounded_norm = False
        self.learning_rate_decay = 0
        self.decay_period = 100

        self.rule_dir = False
        self.normed_bumps = False

        self.augment_inv = False