TRACE/tnc_for_hyper_param_optimization.py at main · Addison-Weatherhead/TRACE · 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
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
import random
import argparse
import torch
from tnc.tnc import main

device = 'cuda' if torch.cuda.is_available() else 'cpu'


if __name__ == '__main__':
    print("STARTED RUNNING", flush=True)
    from datetime import datetime
    print("Started running on ", datetime.now())
    random.seed(1234)
    parser = argparse.ArgumentParser(description='Run TNC')
    parser.add_argument('--train', action='store_true')
    parser.add_argument('--cont', action='store_true')
    parser.add_argument('--ID', type=str)
    parser.add_argument('--plot_embeddings', action='store_true')
    parser.add_argument('--encoder_type', type=str)
    parser.add_argument('--DEBUG', action='store_true')

    # Transformer Hyperparameters
    parser.add_argument('--Transformer_fc_dropout', type=float)
    parser.add_argument('--Transformer_act', type=str)
    parser.add_argument('--Transformer_res_dropout', type=float)
    parser.add_argument('--Transformer_d_ff', type=int)
    parser.add_argument('--Transformer_d_v', type=int)
    parser.add_argument('--Transformer_d_qk', type=int)
    parser.add_argument('--Transformer_n_heads', type=int)
    parser.add_argument('--Transformer_hidden_size', type=int)
    parser.add_argument('--Transformer_n_layers', type=int)
    parser.add_argument('--Transformer_encoding_size', type=int)
    parser.add_argument('--Transformer_num_features', type=int)

    # CNN RNN Hyperparameters
    parser.add_argument('--CNN_RNN_latent_size', type=int)
    parser.add_argument('--CNN_RNN_encoding_size', type=int)
    parser.add_argument('--CNN_RNN_in_channel', type=int)

    # GRUD Hyperparameters
    parser.add_argument('--GRUD_num_features', type=int)
    parser.add_argument('--GRUD_hidden_size', type=int)
    parser.add_argument('--GRUD_num_layers', type=int)
    parser.add_argument('--GRUD_encoding_size', type=int)
    parser.add_argument('--GRUD_extra_layer_types', type=str)
    parser.add_argument('--GRUD_dropout', type=float)

    # RNN Hyperparameters
    parser.add_argument('--RNN_hidden_size', type=int)
    parser.add_argument('--RNN_in_channel', type=int)
    parser.add_argument('--RNN_encoding_size', type=int)

    # CNN_Transformer Hyperparameters
    parser.add_argument('--CNN_Transformer_latent_size', type=int)
    parser.add_argument('--CNN_Transformer_encoding_size', type=int)
    parser.add_argument('--CNN_Transformer_in_channel', type=int)
    parser.add_argument('--CNN_Transformer_transformer_n_layers', type=int)
    parser.add_argument('--CNN_Transformer_transformer_hidden_size', type=int)
    parser.add_argument('--CNN_Transformer_transformer_n_heads', type=int)
    parser.add_argument('--CNN_Transformer_transformer_d_ff', type=int)
    parser.add_argument('--CNN_Transformer_transformer_res_dropout', type=float)
    parser.add_argument('--CNN_Transformer_transformer_act', type=str)
    parser.add_argument('--CNN_Transformer_transformer_fc_dropout', type=float)

    # CausalCNNEncoder Hyperparameters
    parser.add_argument('--CausalCNNEncoder_in_channels', type=int)
    parser.add_argument('--CausalCNNEncoder_channels', type=int)
    parser.add_argument('--CausalCNNEncoder_depth', type=int)
    parser.add_argument('--CausalCNNEncoder_reduced_size', type=int)
    parser.add_argument('--CausalCNNEncoder_encoding_size', type=int)
    parser.add_argument('--CausalCNNEncoder_kernel_size', type=int)
    parser.add_argument('--CausalCNNEncoder_window_size', type=int)


    # Learn encoder hyperparams
    parser.add_argument('--window_size', type=int)
    parser.add_argument('--w', type=float)
    parser.add_argument('--batch_size', type=int)
    parser.add_argument('--lr', type=float)
    parser.add_argument('--decay', type=float)
    parser.add_argument('--mc_sample_size', type=int)
    parser.add_argument('--n_epochs', type=int)
    parser.add_argument('--data_type', type=str)
    parser.add_argument('--n_cross_val_encoder', type=int)
    parser.add_argument('--ETA', type=int)
    parser.add_argument('--ADF', action='store_true')
    parser.add_argument('--ACF', action='store_true')
    parser.add_argument('--ACF_PLUS', action='store_true')
    parser.add_argument('--ACF_nghd_Threshold', type=float)
    parser.add_argument('--ACF_out_nghd_Threshold', type=float)

    # Classifier hyper params
    parser.add_argument('--n_cross_val_classification', type=int)

    args = parser.parse_args()

    if args.encoder_type == 'Transformer':
        encoder_hyper_params = {'verbose': False,
                                'y_range': None,
                                'fc_dropout': args.Transformer_fc_dropout,
                                'act': args.Transformer_act,
                                'res_dropout': args.Transformer_res_dropout,
                                'd_ff': args.Transformer_d_ff,
                                'd_v': args.Transformer_d_v,
                                'd_qk': args.Transformer_d_qk,
                                'n_heads': args.Transformer_n_heads,
                                'hidden_size': args.Transformer_hidden_size,
                                'n_layers': args.Transformer_n_layers,
                                'max_seq_len': None,
                                'seq_len': args.window_size,
                                'encoding_size': args.Transformer_encoding_size,
                                'num_features': args.Transformer_num_features}

    elif args.encoder_type == 'CNN_RNN':
        encoder_hyper_params = {'latent_size': args.CNN_RNN_latent_size,
                                'encoding_size': args.CNN_RNN_encoding_size,
                                'in_channel': 2}

    elif args.encoder_type == 'GRUD':
        encoder_hyper_params = {'num_features': args.GRUD_num_features,
                                'hidden_size': args.GRUD_hidden_size,
                                'num_layers': args.GRUD_num_layers,
                                'encoding_size': args.GRUD_encoding_size,
                                'extra_layer_types': args.GRUD_extra_layer_types,
                                'dropout': args.GRUD_dropout}

    elif args.encoder_type == 'RNN':
        encoder_hyper_params = {'hidden_size': args.RNN_hidden_size,
                                'in_channel': args.RNN_in_channel,
                                'encoding_size': args.RNN_encoding_size}

    elif args.encoder_type == 'CNN_Transformer':
        encoder_hyper_params = {'latent_size': args.CNN_Transformer_latent_size,
                                'encoding_size': args.CNN_Transformer_encoding_size,
                                'in_channel': args.CNN_Transformer_in_channel,
                                'transformer_n_layers': args.CNN_Transformer_transformer_n_layers,
                                'transformer_hidden_size': args.CNN_Transformer_transformer_hidden_size,
                                'transformer_n_heads': args.CNN_Transformer_transformer_n_heads,
                                'transformer_d_ff': args.CNN_Transformer_transformer_d_ff,
                                'transformer_res_dropout': args.CNN_Transformer_transformer_res_dropout,
                                'transformer_act': args.CNN_Transformer_transformer_act,
                                'transformer_fc_dropout': args.CNN_Transformer_transformer_fc_dropout}

    elif args.encoder_type == 'CausalCNNEncoder':
        encoder_hyper_params = {'in_channels': args.CausalCNNEncoder_in_channels,
                                'channels': args.CausalCNNEncoder_channels,
                                'depth': args.CausalCNNEncoder_depth,
                                'reduced_size': args.CausalCNNEncoder_reduced_size,
                                'encoding_size': args.CausalCNNEncoder_encoding_size,
                                'kernel_size': args.CausalCNNEncoder_kernel_size,
                                'window_size': args.CausalCNNEncoder_window_size}


    learn_encoder_hyper_params = {'window_size': args.window_size,
                                    'w': args.w,
                                    'batch_size': args.batch_size,
                                    'lr': args.lr,
                                    'decay': args.decay,
                                    'mc_sample_size': args.mc_sample_size,
                                    'n_epochs': args.n_epochs,
                                    'data_type': args.data_type,
                                    'n_cross_val_encoder': args.n_cross_val_encoder,
                                    'cont': True,
                                    'ETA': args.ETA,
                                    'ADF': args.ADF,
                                    'ACF': args.ACF,
                                    'ACF_PLUS': args.ACF_PLUS,
                                    'ACF_nghd_Threshold': args.ACF_nghd_Threshold,
                                    'ACF_out_nghd_Threshold': args.ACF_out_nghd_Threshold}

    classification_hyper_params = {'n_cross_val_classification': args.n_cross_val_classification}


    UNIQUE_ID = args.ID


    UNIQUE_NAME = UNIQUE_ID + '_' + args.encoder_type + '_' + args.data_type

    print('UNIQUE_NAME: ', UNIQUE_NAME)


    if 'device' not in encoder_hyper_params:
        encoder_hyper_params['device'] = device

    if 'device' not in learn_encoder_hyper_params:
        learn_encoder_hyper_params['device'] = device

    pretrain_hyper_params = {}

    #################################################################
    print("ENCODER HYPER PARAMETERS")
    for key in encoder_hyper_params:
        print(key)
        print(encoder_hyper_params[key])
        print()
    print("LEARN ENCODER HYPER PARAMETERS")
    for key in learn_encoder_hyper_params:
        print(key)
        print(learn_encoder_hyper_params[key])
        print()

    main(args.train, args.data_type, args.encoder_type, encoder_hyper_params, learn_encoder_hyper_params, classification_hyper_params, args.cont, pretrain_hyper_params, args.plot_embeddings, UNIQUE_ID, UNIQUE_NAME)
    print("Finished running on ", datetime.now())