Hugging Face's logo

Miayan
/
project

Diffusers
Safetensors
Model card Files
xet
 Community
Miayan commited on
Commit
9dd62d9
·
verified ·
1 Parent(s): 7979d9e

Add checkpoint for train_ex6_10

Browse files
Files changed (11) hide show
  1. train_ex6_10/checkpoint-65000/controlnet/config.json +56 -0
  2. train_ex6_10/checkpoint-65000/controlnet/diffusion_pytorch_model.safetensors +3 -0
  3. train_ex6_10/checkpoint-65000/custom_encoder/config.json +8 -0
  4. train_ex6_10/checkpoint-65000/custom_encoder/diffusion_pytorch_model.safetensors +3 -0
  5. train_ex6_10/checkpoint-65000/custom_unet.pth +3 -0
  6. train_ex6_10/checkpoint-65000/custom_unet/config.json +73 -0
  7. train_ex6_10/checkpoint-65000/custom_unet/diffusion_pytorch_model.bin +3 -0
  8. train_ex6_10/checkpoint-65000/optimizer.bin +3 -0
  9. train_ex6_10/checkpoint-65000/random_states_0.pkl +3 -0
  10. train_ex6_10/checkpoint-65000/scheduler.bin +3 -0
  11. train_ex6_10/config.yaml +105 -0
train_ex6_10/checkpoint-65000/controlnet/config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "ControlNetModel",
3
+ "_diffusers_version": "0.33.1",
4
+ "act_fn": "silu",
5
+ "addition_embed_type": null,
6
+ "addition_embed_type_num_heads": 64,
7
+ "addition_time_embed_dim": null,
8
+ "attention_head_dim": [
9
+ 5,
10
+ 10,
11
+ 20,
12
+ 20
13
+ ],
14
+ "block_out_channels": [
15
+ 320,
16
+ 640,
17
+ 1280,
18
+ 1280
19
+ ],
20
+ "class_embed_type": null,
21
+ "conditioning_channels": 6,
22
+ "conditioning_embedding_out_channels": [
23
+ 16,
24
+ 32,
25
+ 96,
26
+ 256
27
+ ],
28
+ "controlnet_conditioning_channel_order": "rgb",
29
+ "cross_attention_dim": 1024,
30
+ "down_block_types": [
31
+ "CrossAttnDownBlock2D",
32
+ "CrossAttnDownBlock2D",
33
+ "CrossAttnDownBlock2D",
34
+ "DownBlock2D"
35
+ ],
36
+ "downsample_padding": 1,
37
+ "encoder_hid_dim": null,
38
+ "encoder_hid_dim_type": null,
39
+ "flip_sin_to_cos": true,
40
+ "freq_shift": 0,
41
+ "global_pool_conditions": false,
42
+ "in_channels": 4,
43
+ "layers_per_block": 2,
44
+ "mid_block_scale_factor": 1,
45
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
46
+ "norm_eps": 1e-05,
47
+ "norm_num_groups": 32,
48
+ "num_attention_heads": null,
49
+ "num_class_embeds": null,
50
+ "only_cross_attention": false,
51
+ "projection_class_embeddings_input_dim": null,
52
+ "resnet_time_scale_shift": "default",
53
+ "transformer_layers_per_block": 1,
54
+ "upcast_attention": true,
55
+ "use_linear_projection": true
56
+ }
train_ex6_10/checkpoint-65000/controlnet/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2f80685029d423ab771dacdf090c130094d92f32f8ef1ad157fe89d6a4cce96
3
+ size 1461729992
train_ex6_10/checkpoint-65000/custom_encoder/config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "Bc": 4,
3
+ "Bi": 4,
4
+ "K": 4,
5
+ "_class_name": "CustomEncoder",
6
+ "_diffusers_version": "0.33.1",
7
+ "cross_dim": 1024
8
+ }
train_ex6_10/checkpoint-65000/custom_encoder/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b32f57800d396749945f5d0c7c5c1ef4edc006d2508edcae8dd9a2d20bcd5fa
3
+ size 4531800
train_ex6_10/checkpoint-65000/custom_unet.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c3bdb313fb87b5f4e4c3288b7fa856b56679cac06d3fe7ddf0c6f1489a2feb87
3
+ size 3463973299
train_ex6_10/checkpoint-65000/custom_unet/config.json ADDED
@@ -0,0 +1,73 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "UNet2DConditionModel",
3
+ "_diffusers_version": "0.33.1",
4
+ "_name_or_path": "/mnt/HDD3/miayan/paper/envs/huggingface_cache/models--stabilityai--stable-diffusion-2-1/snapshots/5cae40e6a2745ae2b01ad92ae5043f95f23644d6",
5
+ "act_fn": "silu",
6
+ "addition_embed_type": null,
7
+ "addition_embed_type_num_heads": 64,
8
+ "addition_time_embed_dim": null,
9
+ "attention_head_dim": [
10
+ 5,
11
+ 10,
12
+ 20,
13
+ 20
14
+ ],
15
+ "attention_type": "default",
16
+ "block_out_channels": [
17
+ 320,
18
+ 640,
19
+ 1280,
20
+ 1280
21
+ ],
22
+ "center_input_sample": false,
23
+ "class_embed_type": null,
24
+ "class_embeddings_concat": false,
25
+ "conv_in_kernel": 3,
26
+ "conv_out_kernel": 3,
27
+ "cross_attention_dim": 1024,
28
+ "cross_attention_norm": null,
29
+ "down_block_types": [
30
+ "CrossAttnDownBlock2D",
31
+ "CrossAttnDownBlock2D",
32
+ "CrossAttnDownBlock2D",
33
+ "DownBlock2D"
34
+ ],
35
+ "downsample_padding": 1,
36
+ "dropout": 0.0,
37
+ "dual_cross_attention": false,
38
+ "encoder_hid_dim": null,
39
+ "encoder_hid_dim_type": null,
40
+ "flip_sin_to_cos": true,
41
+ "freq_shift": 0,
42
+ "in_channels": 8,
43
+ "layers_per_block": 2,
44
+ "mid_block_only_cross_attention": null,
45
+ "mid_block_scale_factor": 1,
46
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
47
+ "norm_eps": 1e-05,
48
+ "norm_num_groups": 32,
49
+ "num_attention_heads": null,
50
+ "num_class_embeds": null,
51
+ "only_cross_attention": false,
52
+ "out_channels": 4,
53
+ "projection_class_embeddings_input_dim": null,
54
+ "resnet_out_scale_factor": 1.0,
55
+ "resnet_skip_time_act": false,
56
+ "resnet_time_scale_shift": "default",
57
+ "reverse_transformer_layers_per_block": null,
58
+ "sample_size": 96,
59
+ "time_cond_proj_dim": null,
60
+ "time_embedding_act_fn": null,
61
+ "time_embedding_dim": null,
62
+ "time_embedding_type": "positional",
63
+ "timestep_post_act": null,
64
+ "transformer_layers_per_block": 1,
65
+ "up_block_types": [
66
+ "UpBlock2D",
67
+ "CrossAttnUpBlock2D",
68
+ "CrossAttnUpBlock2D",
69
+ "CrossAttnUpBlock2D"
70
+ ],
71
+ "upcast_attention": true,
72
+ "use_linear_projection": true
73
+ }
train_ex6_10/checkpoint-65000/custom_unet/diffusion_pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a83e718a4b2c45fb5535292d188bc9660ea726bef3aceb22cc36611d016f1a7
3
+ size 3463934051
train_ex6_10/checkpoint-65000/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:69cbf5e683033cbfe8a1e75206363b852c8bdbeb35c0d3a426eb01cba11a283b
3
+ size 5573559453
train_ex6_10/checkpoint-65000/random_states_0.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:caebf9492d274a1892b9eab130a82264e94e7587931be02ba602713ef1a2c974
3
+ size 14757
train_ex6_10/checkpoint-65000/scheduler.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:21fe3df4aac65d05da1f023f48dd5f8b1a1a28f85cd1cd6fb340a8b680361263
3
+ size 1401
train_ex6_10/config.yaml ADDED
@@ -0,0 +1,105 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ experiment:
2
+ project_name: train_controlnet_ex5
3
+ output_dir: train_ex6_10
4
+ logging_dir: logs
5
+ report_to: tensorboard
6
+ model:
7
+ pretrained_model_name_or_path: /mnt/HDD3/miayan/paper/envs/huggingface_cache/models--stabilityai--stable-diffusion-2-1/snapshots/5cae40e6a2745ae2b01ad92ae5043f95f23644d6
8
+ revision: null
9
+ variant: null
10
+ tokenizer_name: null
11
+ hf_repo_id: Miayan/project
12
+ hf_version: null
13
+ pretrain_unet_path: scribblelight_controlnet/checkpoint-10000
14
+ resume_from_checkpoint: latest
15
+ light_encoder:
16
+ cross_dim: 768
17
+ K: 4
18
+ Bi: 4
19
+ Bc: 4
20
+ data:
21
+ dataset_path: /mnt/HDD3/miayan/paper/relighting_datasets/
22
+ data_hf_repo_id: Miayan/physical-relighting-dataset
23
+ data_split: test
24
+ resolution: 256
25
+ dataloader_num_workers: 4
26
+ condition_mode: lightmap_normal
27
+ colors:
28
+ - - 255
29
+ - 255
30
+ - 255
31
+ - - 255
32
+ - 0
33
+ - 0
34
+ - - 0
35
+ - 255
36
+ - 0
37
+ - - 0
38
+ - 0
39
+ - 255
40
+ - - 255
41
+ - 255
42
+ - 0
43
+ - - 255
44
+ - 165
45
+ - 0
46
+ - - 128
47
+ - 0
48
+ - 128
49
+ - - 255
50
+ - 192
51
+ - 203
52
+ - - 0
53
+ - 255
54
+ - 255
55
+ - - 255
56
+ - 0
57
+ - 255
58
+ intensities:
59
+ - 0.0
60
+ - 0.1
61
+ - 0.2
62
+ - 0.4
63
+ - 0.7
64
+ - 1.0
65
+ training:
66
+ batch_size: 1
67
+ num_epochs: 20
68
+ max_train_steps: null
69
+ max_train_samples: null
70
+ seed: 42
71
+ mixed_precision: fp16
72
+ allow_tf32: true
73
+ gradient_accumulation_steps: 1
74
+ checkpointing_steps: 5000
75
+ checkpoints_total_limit: 5
76
+ learning_rate: 5.0e-06
77
+ scale_lr: false
78
+ lr_scheduler: constant
79
+ lr_warmup_steps: 500
80
+ adam_beta1: 0.9
81
+ adam_beta2: 0.999
82
+ adam_weight_decay: 0.01
83
+ adam_epsilon: 1.0e-08
84
+ max_grad_norm: 1.0
85
+ use_8bit_adam: false
86
+ set_grads_to_none: false
87
+ losses:
88
+ phys_loss:
89
+ enabled: true
90
+ weight: 1.0
91
+ latent_loss:
92
+ enabled: true
93
+ weight: 1.0
94
+ image_loss:
95
+ enabled: true
96
+ weight: 1.0
97
+ area_loss:
98
+ enabled: false
99
+ weight: 10.0
100
+ intensity_loss:
101
+ enabled: false
102
+ weight: 1.0
103
+ recon_loss:
104
+ enabled: false
105
+ weight: 1.0