-
Notifications
You must be signed in to change notification settings - Fork 2
/
directories_names.py
203 lines (117 loc) · 6.81 KB
/
directories_names.py
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
import numpy as np
# ------------------------------------------ main figures saving dirs ------------------------------------------
def get_dir_aux_results_dir(dataset_name, transformation, epoch, is_real, seed, tau, dropout=0., hs='[64, 64, 64]',
vae_loss='KL',
vae_z_dim=3, vae_mode='CVAE', method_name='directional', kl_mult=0.01, **kwargs):
if is_real:
datatype = 'real'
else:
datatype = 'syn'
if 'identity' in str(transformation):
transformation_name = str(transformation)
else:
transformation_name = f"{str(transformation)}_{get_vae_save_name_by_params(vae_loss, vae_z_dim, vae_mode, kl_mult)}"
model_params = f"method={method_name}_dropout={np.round(dropout, 2)}_hs={str(hs)}"
base_save_dir = f"{datatype}/{dataset_name}/{transformation_name}_{model_params}_tau={tau}"
return base_save_dir
def get_save_final_results_dir(dataset_name, **kwargs):
base_save_dir = f"final_results/{get_dir_aux_results_dir(dataset_name=dataset_name, **kwargs)}"
return base_save_dir
def get_save_final_figure_results_dir(dataset_name, **kwargs):
base_save_dir = f"results/full_figures/{get_dir_aux_results_dir(dataset_name=dataset_name, **kwargs)}"
return base_save_dir
def get_model_summary_save_dir(dataset_name, **kwargs):
save_dir = f"results/summary_figures/{get_dir_aux_results_dir(dataset_name=dataset_name, **kwargs)}"
return save_dir
# ------------------------------------------ VAE figures saving dirs ------------------------------------------
def get_vae_save_name_by_params(vae_loss, vae_z_dim, vae_mode, kl_mult=0.01):
return f"mode={vae_mode}_loss={vae_loss}_z_dim={vae_z_dim}_kl={kl_mult}"
def get_encoder_base_save_dir_aux(dataset_name, epoch, is_real, seed, z_dim, loss, mode, kl_mult=0.01):
if is_real:
datatype = 'real'
else:
datatype = 'syn'
base_save_dir = f"{datatype}/{dataset_name}/vae/seed={seed}/{get_vae_save_name_by_params(loss, z_dim, mode, kl_mult)}"
return base_save_dir
def get_vae_results_save_dir(dataset_name, epoch, is_real, seed, z_dim, loss, mode, kl_mult=0.01, **kwargs):
if is_real:
datatype = 'real'
else:
datatype = 'syn'
partial_save_dir = f"{datatype}/{dataset_name}/{get_vae_save_name_by_params(loss, z_dim, mode, kl_mult)}"
return f'final_results/vae/{partial_save_dir}'
def get_vae_results_save_path(dataset_name, epoch, is_real, seed, z_dim, loss, mode, kl_mult=0.01, **kwargs):
return f'{get_vae_results_save_dir(dataset_name, epoch, is_real, seed, z_dim, loss, mode, kl_mult=kl_mult)}/seed={seed}.csv'
def get_encoder_base_save_dir(**kwargs):
base_save_dir = f"results/full_figures/{get_encoder_base_save_dir_aux(**kwargs)}"
return base_save_dir
def get_encoder_summary_save_dir(**kwargs):
base_save_dir = f"results/summary_figures/{get_encoder_base_save_dir_aux(**kwargs)}"
return base_save_dir
# ------------------------------------------ VAE model saving dirs ------------------------------------------
def get_vae_model_save_path(dataset_name, seed, vae_loss, vae_z_dim, vae_mode, kl_mult=0.01):
return f'models_save/{dataset_name}/seed={seed}/{get_vae_save_name_by_params(vae_loss, vae_z_dim, vae_mode, kl_mult)}'
def get_vae_model_save_name(dataset_name, seed, vae_loss, vae_z_dim, vae_mode, kl_mult=0.01):
return f'{get_vae_model_save_path(dataset_name, seed, vae_loss, vae_z_dim, vae_mode, kl_mult)}/vae'
def get_cvae_model_save_name(dataset_name, seed, vae_loss, vae_z_dim, vae_mode, kl_mult=0.01):
return f'{get_vae_model_save_path(dataset_name, seed, vae_loss, vae_z_dim, vae_mode, kl_mult)}/cvae'
# ------------------------------------------ Area plotting saving dir ------------------------------------------
def get_area_vs_d_figure_save_dir(is_linear, x_dim, tau):
return f'results/area_figures/is_linear={is_linear}_x_dim={x_dim}_tau={tau}'
def get_area_vs_d_figure_save_path(is_linear, x_dim, scale, tau):
return f'{get_area_vs_d_figure_save_dir(is_linear, x_dim, tau)}/scale={scale}.png'
# ------------------------------------------ VAE MSE losses plotting saving dir ------------------------------------------
def get_mse_vs_r_figure_save_dir():
return 'results/vae_figures'
def get_mse_vs_r_figure_save_path():
return f'{get_mse_vs_r_figure_save_dir()}/mse_vs_r.png'
# ------------------------------------------ Calibration example saving dir ------------------------------------------
def get_cal_example_save_dir():
return f'results/cal_example'
def get_too_large_qr_figure_save_path():
return f'{get_cal_example_save_dir()}/large_gamma.png'
def get_too_small_qr_figure_save_path():
return f'{get_cal_example_save_dir()}/small_gamma.png'
def get_qr_complement_figure_save_path():
return f'{get_cal_example_save_dir()}/out_of_qr.png'
def get_best_qr_figure_save_path():
return f'{get_cal_example_save_dir()}/best_qr.png'
def get_case_1_too_large_qr_figure_save_path():
return f'{get_cal_example_save_dir()}/case_1_large_gamma.png'
def get_case_1_too_small_qr_figure_save_path():
return f'{get_cal_example_save_dir()}/case_1_small_gamma.png'
def get_case_1_best_gamma_figure_save_path():
return f'{get_cal_example_save_dir()}/case_1_best_gamma.png'
def get_cov_vs_gamma__graph_figure_save_path():
return f'{get_cal_example_save_dir()}/cov_vs_gamma_graph.png'
def get_shrinking_qr1_figure_save_path():
return f'{get_cal_example_save_dir()}/shrinking_qr1.png'
def get_shrinking_qr2_figure_save_path():
return f'{get_cal_example_save_dir()}/shrinking_qr2.png'
# ------------------------------------------ Grid size saving dir ------------------------------------------
def get_grid_info_save_path(dataset_name, seed):
return f'{get_grid_info_save_dir(dataset_name, seed)}/grid_info.csv'
def get_grid_info_save_dir(dataset_name, seed):
return f'grid_info/{dataset_name}/{seed}'
def get_encoder_transformed_y_fig_name():
return 'transformed_y'
def get_original_and_reconstructed_y_fig_name():
return 'original_and_reconstructed_y'
def get_save_halfspace_figure_sub_path():
return 'directional_region'
def get_transformed_region_figure_name():
return 'quantile_region.png'
def get_inverse_transformed_region_figure_name():
return 'inverse_transformed_quantile_region.png'
def get_z_space_region_figure_name():
return 'transformed_space_quantile_region.png'
def get_transformed_contour_figure_name():
return 'quantile_contour.png'
def get_untransformed_region_figure_name():
return 'untransformed_quantile_region.png'
def get_naive_quantile_contours_figure_name():
return 'quantile_contour.png'
def get_vqr_results_dir(dataset_name, is_real, seed):
ds_type = 'real_data' if is_real else 'syn_data'
results_dir = f'VQR/results/{ds_type}/{dataset_name}/seed={seed}'
return results_dir