Hugging Face's logo

Datasets:
datnguyentien204
/
CRNetAblation

Dataset card Files
xet
 Community
1
datnguyentien204 commited on
Commit
951beb7
·
verified ·
1 Parent(s): 31cc1cb

Upload 16 files

Browse files
Files changed (16) hide show
  1. ast_rafael_v2_sharpening_ablation_v1/20_net_D2.pth +3 -0
  2. ast_rafael_v2_sharpening_ablation_v1/20_net_G.pth +3 -0
  3. ast_rafael_v2_sharpening_ablation_v1/iter.txt +2 -0
  4. ast_rafael_v2_sharpening_ablation_v1/latest_net_D2.pth +3 -0
  5. ast_rafael_v2_sharpening_ablation_v1/latest_net_G.pth +3 -0
  6. ast_rafael_v2_sharpening_ablation_v1/loss_log.txt +0 -0
  7. ast_rafael_v2_sharpening_ablation_v1/opt.pkl +3 -0
  8. ast_rafael_v2_sharpening_ablation_v1/opt.txt +119 -0
  9. ast_rafael_v2_sharpening_ablation_v3/20_net_D2.pth +3 -0
  10. ast_rafael_v2_sharpening_ablation_v3/20_net_G.pth +3 -0
  11. ast_rafael_v2_sharpening_ablation_v3/iter.txt +2 -0
  12. ast_rafael_v2_sharpening_ablation_v3/latest_net_D2.pth +3 -0
  13. ast_rafael_v2_sharpening_ablation_v3/latest_net_G.pth +3 -0
  14. ast_rafael_v2_sharpening_ablation_v3/loss_log.txt +0 -0
  15. ast_rafael_v2_sharpening_ablation_v3/opt.pkl +3 -0
  16. ast_rafael_v2_sharpening_ablation_v3/opt.txt +119 -0
ast_rafael_v2_sharpening_ablation_v1/20_net_D2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c838ebd3d360c63ebe130aaee7630c7217c6927be7ed37a11f475472ec1ef4e9
3
+ size 133474379
ast_rafael_v2_sharpening_ablation_v1/20_net_G.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d6b67c46f13b5593b5fb5cb0a662ec873e648f595f9baf1c9e887528aba4ac6
3
+ size 365458163
ast_rafael_v2_sharpening_ablation_v1/iter.txt ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ 21
2
+ 0
ast_rafael_v2_sharpening_ablation_v1/latest_net_D2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da6ecbd4821199663187eae252e208bdffac164420a53f23a39e8fe24a8eb143
3
+ size 133475351
ast_rafael_v2_sharpening_ablation_v1/latest_net_G.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb8cc868c395aea4bd5be7d0bf40a40f89f500e3ea05a4885619b4ced17623a8
3
+ size 365459027
ast_rafael_v2_sharpening_ablation_v1/loss_log.txt ADDED
The diff for this file is too large to render. See raw diff
 
ast_rafael_v2_sharpening_ablation_v1/opt.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6cffa08f532b550e53fb12deeabbd7b2ef7b4d9b987bc0c4e4433d297500f503
3
+ size 2661
ast_rafael_v2_sharpening_ablation_v1/opt.txt ADDED
@@ -0,0 +1,119 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ D_steps_per_G: 1
2
+ T: 0.07
3
+ add_phi_to_latent: True [default: False]
4
+ aspect_ratio: 1.0
5
+ attn_layers: 4,7,9
6
+ batchSize: 2 [default: 1]
7
+ beta1: 0.5
8
+ beta2: 0.999
9
+ blackout_threshold: -0.9
10
+ cache_filelist_read: False
11
+ cache_filelist_write: False
12
+ checkpoints_dir: ./checkpoints_v2 [default: ./checkpoints]
13
+ cond_enc_input_nc: 3
14
+ cond_enc_ngf: 32 [default: 64]
15
+ cond_enc_num_downs: 2 [default: 3]
16
+ contain_dontcare_label: False
17
+ continue_train: True [default: False]
18
+ croot: ./datasets/summer2winter_yosemite/
19
+ crop_size: 256
20
+ d_cross_attn_num_heads: 4
21
+ dataroot: /workspace/RAFAEL_Recon/rafael_ds_ver2 [default: None]
22
+ dataset_mode: unaligned_day_night [default: summer2winteryosemite]
23
+ daylight_curve_steepness: 1.5 [default: 0.75]
24
+ debug: False
25
+ display_freq: 100
26
+ display_winsize: 256
27
+ gan_mode: hinge
28
+ gpu_ids: 0
29
+ init_type: xavier
30
+ init_variance: 0.02
31
+ isTrain: True [default: None]
32
+ label_nc: 3
33
+ lambda_G: 1.0
34
+ lambda_content_focus: 5.0
35
+ lambda_continuity: 20.0 [default: 10.0]
36
+ lambda_feat: 15.0 [default: 10.0]
37
+ lambda_identity: 10.0
38
+ lambda_kld: 0.05
39
+ lambda_laplacian: 25.0 [default: 5.0]
40
+ lambda_latent_diversity: 1.0
41
+ lambda_night_content: 0.0
42
+ lambda_patch_vgg: 10.0
43
+ lambda_ppl: 0.5 [default: 2.0]
44
+ lambda_r1: 10.0
45
+ lambda_spatial: 5.0 [default: 10.0]
46
+ lambda_ssim: 1.0
47
+ lambda_style: 0.0 [default: 30.0]
48
+ lambda_style_recon: 15.0
49
+ lambda_thresholded_blackout: 20.0
50
+ lambda_tv: 8.0 [default: 5.0]
51
+ lambda_vgg: 10.0
52
+ lambda_vgg_interp: 100.0 [default: 15.0]
53
+ lambda_z_reg: 5.0 [default: 0.0]
54
+ learned_attn: False
55
+ light_map_channels: 1
56
+ load_from_opt_file: False
57
+ load_size: 286
58
+ loss_mode: cos
59
+ lr: 0.0001 [default: 0.0002]
60
+ max_dataset_size: 9223372036854775807
61
+ model: pix2pix
62
+ nThreads: 4 [default: 0]
63
+ name: ast_rafael_v2_ablation_v1 [default: ast_s2wat_default_experiment]
64
+ ndf: 64
65
+ netD: swin_transformer_conditional_discriminator [default: None]
66
+ netG: RafaelGenerator
67
+ ngf: 64
68
+ night_loss_warmup_iters: 10000
69
+ niter: 15 [default: 50]
70
+ niter_decay: 5 [default: 0]
71
+ no_TTUR: False
72
+ no_flip: False
73
+ no_ganFeat_loss: False
74
+ no_html: False
75
+ no_instance: True [default: False]
76
+ no_vgg_loss: False
77
+ norm_D: spectralinstance
78
+ norm_E: spectralinstance
79
+ norm_G: spectralinstance
80
+ norm_S: spectralinstance
81
+ optimizer: adam
82
+ output_nc: 3
83
+ patch_nums: 128.0
84
+ patch_size: 32
85
+ patch_vgg_num_patches: 64
86
+ patch_vgg_size: 64
87
+ phase: train
88
+ phi: 0.0
89
+ ppl_reg_every: 8 [default: 4]
90
+ preprocess_mode: resize_and_crop
91
+ print_freq: 100
92
+ r1_reg_every: 16
93
+ rafael_depths: 2,2,2
94
+ rafael_embed_dim: 192
95
+ rafael_img_size: 256
96
+ rafael_nheads: 3,6,12
97
+ rafael_patch_size: 2
98
+ rafael_strip_widths: 2,4,7
99
+ rafael_use_checkpoint: True [default: False]
100
+ save_epoch_freq: 2 [default: 10]
101
+ save_latest_freq: 2000 [default: 5000]
102
+ semantic_nc: 3 [default: None]
103
+ serial_batches: False
104
+ sroot: ./datasets/summer2winter_yosemite/
105
+ style_enc_latent_dim: 256
106
+ style_mixing_prob: 0.9
107
+ swin_d_model_name: swinv2_tiny_window8_256.ms_in1k
108
+ swin_d_num_conv_head: 3 [default: 2]
109
+ swin_d_output_style: patch
110
+ swin_d_pretrained: True [default: False]
111
+ swin_d_use_sigmoid: False
112
+ synth_num_res_blocks: 4
113
+ task: AST
114
+ tf_log: False
115
+ use_norm: False
116
+ use_patch_vgg_loss: True [default: False]
117
+ use_thresholded_blackout: False
118
+ use_vae: False
119
+ which_epoch: latest
ast_rafael_v2_sharpening_ablation_v3/20_net_D2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4cbb9a7221da71951b7e43f63bddc65b9f6a84f7f7b4ed77635b94e84a95936b
3
+ size 133474379
ast_rafael_v2_sharpening_ablation_v3/20_net_G.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa72a5394fd2d6250a2ed6cf0ff37e2ab484d8bab8ab36d0d94a38216b0905f6
3
+ size 365458163
ast_rafael_v2_sharpening_ablation_v3/iter.txt ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ 21
2
+ 0
ast_rafael_v2_sharpening_ablation_v3/latest_net_D2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:539ee53330a0736901bd2b378d233f5712a40f0534fb79f8541643c1150eacfc
3
+ size 133475351
ast_rafael_v2_sharpening_ablation_v3/latest_net_G.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d92aaec98f8affb8f067980df6cca22f9a57618f83e07fab1cdd9c34aad872b
3
+ size 365459027
ast_rafael_v2_sharpening_ablation_v3/loss_log.txt ADDED
The diff for this file is too large to render. See raw diff
 
ast_rafael_v2_sharpening_ablation_v3/opt.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:74ddc8307106425838a24787e6a9d5afc8c8134f670328e69ca8b349923c7db4
3
+ size 2661
ast_rafael_v2_sharpening_ablation_v3/opt.txt ADDED
@@ -0,0 +1,119 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ D_steps_per_G: 1
2
+ T: 0.07
3
+ add_phi_to_latent: True [default: False]
4
+ aspect_ratio: 1.0
5
+ attn_layers: 4,7,9
6
+ batchSize: 2 [default: 1]
7
+ beta1: 0.5
8
+ beta2: 0.999
9
+ blackout_threshold: -0.9
10
+ cache_filelist_read: False
11
+ cache_filelist_write: False
12
+ checkpoints_dir: ./checkpoints_v2 [default: ./checkpoints]
13
+ cond_enc_input_nc: 3
14
+ cond_enc_ngf: 32 [default: 64]
15
+ cond_enc_num_downs: 2 [default: 3]
16
+ contain_dontcare_label: False
17
+ continue_train: True [default: False]
18
+ croot: ./datasets/summer2winter_yosemite/
19
+ crop_size: 256
20
+ d_cross_attn_num_heads: 4
21
+ dataroot: /workspace/RAFAEL_Recon/rafael_ds_ver2 [default: None]
22
+ dataset_mode: unaligned_day_night [default: summer2winteryosemite]
23
+ daylight_curve_steepness: 1.5 [default: 0.75]
24
+ debug: False
25
+ display_freq: 100
26
+ display_winsize: 256
27
+ gan_mode: hinge
28
+ gpu_ids: 0
29
+ init_type: xavier
30
+ init_variance: 0.02
31
+ isTrain: True [default: None]
32
+ label_nc: 3
33
+ lambda_G: 1.0
34
+ lambda_content_focus: 5.0
35
+ lambda_continuity: 20.0 [default: 10.0]
36
+ lambda_feat: 15.0 [default: 10.0]
37
+ lambda_identity: 10.0
38
+ lambda_kld: 0.05
39
+ lambda_laplacian: 0.0 [default: 5.0]
40
+ lambda_latent_diversity: 1.0
41
+ lambda_night_content: 0.0
42
+ lambda_patch_vgg: 10.0
43
+ lambda_ppl: 0.5 [default: 2.0]
44
+ lambda_r1: 10.0
45
+ lambda_spatial: 5.0 [default: 10.0]
46
+ lambda_ssim: 1.0
47
+ lambda_style: 0.0 [default: 30.0]
48
+ lambda_style_recon: 15.0
49
+ lambda_thresholded_blackout: 0.0 [default: 20.0]
50
+ lambda_tv: 0.0 [default: 5.0]
51
+ lambda_vgg: 10.0
52
+ lambda_vgg_interp: 100.0 [default: 15.0]
53
+ lambda_z_reg: 5.0 [default: 0.0]
54
+ learned_attn: False
55
+ light_map_channels: 1
56
+ load_from_opt_file: False
57
+ load_size: 286
58
+ loss_mode: cos
59
+ lr: 0.0001 [default: 0.0002]
60
+ max_dataset_size: 9223372036854775807
61
+ model: pix2pix
62
+ nThreads: 4 [default: 0]
63
+ name: ast_rafael_v2_ablation_v3 [default: ast_s2wat_default_experiment]
64
+ ndf: 64
65
+ netD: swin_transformer_conditional_discriminator [default: None]
66
+ netG: RafaelGenerator
67
+ ngf: 64
68
+ night_loss_warmup_iters: 10000
69
+ niter: 15 [default: 50]
70
+ niter_decay: 5 [default: 0]
71
+ no_TTUR: False
72
+ no_flip: False
73
+ no_ganFeat_loss: False
74
+ no_html: False
75
+ no_instance: True [default: False]
76
+ no_vgg_loss: False
77
+ norm_D: spectralinstance
78
+ norm_E: spectralinstance
79
+ norm_G: spectralinstance
80
+ norm_S: spectralinstance
81
+ optimizer: adam
82
+ output_nc: 3
83
+ patch_nums: 128.0
84
+ patch_size: 32
85
+ patch_vgg_num_patches: 64
86
+ patch_vgg_size: 64
87
+ phase: train
88
+ phi: 0.0
89
+ ppl_reg_every: 8 [default: 4]
90
+ preprocess_mode: resize_and_crop
91
+ print_freq: 100
92
+ r1_reg_every: 16
93
+ rafael_depths: 2,2,2
94
+ rafael_embed_dim: 192
95
+ rafael_img_size: 256
96
+ rafael_nheads: 3,6,12
97
+ rafael_patch_size: 2
98
+ rafael_strip_widths: 2,4,7
99
+ rafael_use_checkpoint: True [default: False]
100
+ save_epoch_freq: 2 [default: 10]
101
+ save_latest_freq: 2000 [default: 5000]
102
+ semantic_nc: 3 [default: None]
103
+ serial_batches: False
104
+ sroot: ./datasets/summer2winter_yosemite/
105
+ style_enc_latent_dim: 256
106
+ style_mixing_prob: 0.9
107
+ swin_d_model_name: swinv2_tiny_window8_256.ms_in1k
108
+ swin_d_num_conv_head: 3 [default: 2]
109
+ swin_d_output_style: patch
110
+ swin_d_pretrained: True [default: False]
111
+ swin_d_use_sigmoid: False
112
+ synth_num_res_blocks: 4
113
+ task: AST
114
+ tf_log: False
115
+ use_norm: False
116
+ use_patch_vgg_loss: True [default: False]
117
+ use_thresholded_blackout: False
118
+ use_vae: False
119
+ which_epoch: latest