camenduru commited on
Commit
ec4170b
1 Parent(s): 2e01957

thanks to lllyasviel ❤

Browse files
README.md ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ ---
2
+ license: openrail
3
+ ---
4
+
5
+ This is the model files for [ControlNet 1.1](https://github.com/lllyasviel/ControlNet-v1-1-nightly).
6
+ This model card will be filled in a more detailed way after 1.1 is officially merged into ControlNet.
control_v11e_sd15_ip2p.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a8d77429b111b25ffad4adb4c686a19c73c905d638dd8acb1e0b1d1da300192c
3
+ size 1445234339
control_v11e_sd15_ip2p.yaml ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model:
2
+ target: cldm.cldm.ControlLDM
3
+ params:
4
+ linear_start: 0.00085
5
+ linear_end: 0.0120
6
+ num_timesteps_cond: 1
7
+ log_every_t: 200
8
+ timesteps: 1000
9
+ first_stage_key: "jpg"
10
+ cond_stage_key: "txt"
11
+ control_key: "hint"
12
+ image_size: 64
13
+ channels: 4
14
+ cond_stage_trainable: false
15
+ conditioning_key: crossattn
16
+ monitor: val/loss_simple_ema
17
+ scale_factor: 0.18215
18
+ use_ema: False
19
+ only_mid_control: False
20
+
21
+ control_stage_config:
22
+ target: cldm.cldm.ControlNet
23
+ params:
24
+ image_size: 32 # unused
25
+ in_channels: 4
26
+ hint_channels: 3
27
+ model_channels: 320
28
+ attention_resolutions: [ 4, 2, 1 ]
29
+ num_res_blocks: 2
30
+ channel_mult: [ 1, 2, 4, 4 ]
31
+ num_heads: 8
32
+ use_spatial_transformer: True
33
+ transformer_depth: 1
34
+ context_dim: 768
35
+ use_checkpoint: True
36
+ legacy: False
37
+
38
+ unet_config:
39
+ target: cldm.cldm.ControlledUnetModel
40
+ params:
41
+ image_size: 32 # unused
42
+ in_channels: 4
43
+ out_channels: 4
44
+ model_channels: 320
45
+ attention_resolutions: [ 4, 2, 1 ]
46
+ num_res_blocks: 2
47
+ channel_mult: [ 1, 2, 4, 4 ]
48
+ num_heads: 8
49
+ use_spatial_transformer: True
50
+ transformer_depth: 1
51
+ context_dim: 768
52
+ use_checkpoint: True
53
+ legacy: False
54
+
55
+ first_stage_config:
56
+ target: ldm.models.autoencoder.AutoencoderKL
57
+ params:
58
+ embed_dim: 4
59
+ monitor: val/rec_loss
60
+ ddconfig:
61
+ double_z: true
62
+ z_channels: 4
63
+ resolution: 256
64
+ in_channels: 3
65
+ out_ch: 3
66
+ ch: 128
67
+ ch_mult:
68
+ - 1
69
+ - 2
70
+ - 4
71
+ - 4
72
+ num_res_blocks: 2
73
+ attn_resolutions: []
74
+ dropout: 0.0
75
+ lossconfig:
76
+ target: torch.nn.Identity
77
+
78
+ cond_stage_config:
79
+ target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
control_v11e_sd15_shuffle.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:657f632c35176c467e71c471b55b1a0e94ae1a49d739e1b705624f454e638e1d
3
+ size 1445235365
control_v11e_sd15_shuffle.yaml ADDED
@@ -0,0 +1,80 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model:
2
+ target: cldm.cldm.ControlLDM
3
+ params:
4
+ linear_start: 0.00085
5
+ linear_end: 0.0120
6
+ num_timesteps_cond: 1
7
+ log_every_t: 200
8
+ timesteps: 1000
9
+ first_stage_key: "jpg"
10
+ cond_stage_key: "txt"
11
+ control_key: "hint"
12
+ image_size: 64
13
+ channels: 4
14
+ cond_stage_trainable: false
15
+ conditioning_key: crossattn
16
+ monitor: val/loss_simple_ema
17
+ scale_factor: 0.18215
18
+ use_ema: False
19
+ only_mid_control: False
20
+ global_average_pooling: True
21
+
22
+ control_stage_config:
23
+ target: cldm.cldm.ControlNet
24
+ params:
25
+ image_size: 32 # unused
26
+ in_channels: 4
27
+ hint_channels: 3
28
+ model_channels: 320
29
+ attention_resolutions: [ 4, 2, 1 ]
30
+ num_res_blocks: 2
31
+ channel_mult: [ 1, 2, 4, 4 ]
32
+ num_heads: 8
33
+ use_spatial_transformer: True
34
+ transformer_depth: 1
35
+ context_dim: 768
36
+ use_checkpoint: True
37
+ legacy: False
38
+
39
+ unet_config:
40
+ target: cldm.cldm.ControlledUnetModel
41
+ params:
42
+ image_size: 32 # unused
43
+ in_channels: 4
44
+ out_channels: 4
45
+ model_channels: 320
46
+ attention_resolutions: [ 4, 2, 1 ]
47
+ num_res_blocks: 2
48
+ channel_mult: [ 1, 2, 4, 4 ]
49
+ num_heads: 8
50
+ use_spatial_transformer: True
51
+ transformer_depth: 1
52
+ context_dim: 768
53
+ use_checkpoint: True
54
+ legacy: False
55
+
56
+ first_stage_config:
57
+ target: ldm.models.autoencoder.AutoencoderKL
58
+ params:
59
+ embed_dim: 4
60
+ monitor: val/rec_loss
61
+ ddconfig:
62
+ double_z: true
63
+ z_channels: 4
64
+ resolution: 256
65
+ in_channels: 3
66
+ out_ch: 3
67
+ ch: 128
68
+ ch_mult:
69
+ - 1
70
+ - 2
71
+ - 4
72
+ - 4
73
+ num_res_blocks: 2
74
+ attn_resolutions: []
75
+ dropout: 0.0
76
+ lossconfig:
77
+ target: torch.nn.Identity
78
+
79
+ cond_stage_config:
80
+ target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
control_v11p_sd15_canny.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f99cfe4c70910e38e3fece9918a4979ed7d3dcf9b81cee293e1755363af5406a
3
+ size 1445234681
control_v11p_sd15_canny.yaml ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model:
2
+ target: cldm.cldm.ControlLDM
3
+ params:
4
+ linear_start: 0.00085
5
+ linear_end: 0.0120
6
+ num_timesteps_cond: 1
7
+ log_every_t: 200
8
+ timesteps: 1000
9
+ first_stage_key: "jpg"
10
+ cond_stage_key: "txt"
11
+ control_key: "hint"
12
+ image_size: 64
13
+ channels: 4
14
+ cond_stage_trainable: false
15
+ conditioning_key: crossattn
16
+ monitor: val/loss_simple_ema
17
+ scale_factor: 0.18215
18
+ use_ema: False
19
+ only_mid_control: False
20
+
21
+ control_stage_config:
22
+ target: cldm.cldm.ControlNet
23
+ params:
24
+ image_size: 32 # unused
25
+ in_channels: 4
26
+ hint_channels: 3
27
+ model_channels: 320
28
+ attention_resolutions: [ 4, 2, 1 ]
29
+ num_res_blocks: 2
30
+ channel_mult: [ 1, 2, 4, 4 ]
31
+ num_heads: 8
32
+ use_spatial_transformer: True
33
+ transformer_depth: 1
34
+ context_dim: 768
35
+ use_checkpoint: True
36
+ legacy: False
37
+
38
+ unet_config:
39
+ target: cldm.cldm.ControlledUnetModel
40
+ params:
41
+ image_size: 32 # unused
42
+ in_channels: 4
43
+ out_channels: 4
44
+ model_channels: 320
45
+ attention_resolutions: [ 4, 2, 1 ]
46
+ num_res_blocks: 2
47
+ channel_mult: [ 1, 2, 4, 4 ]
48
+ num_heads: 8
49
+ use_spatial_transformer: True
50
+ transformer_depth: 1
51
+ context_dim: 768
52
+ use_checkpoint: True
53
+ legacy: False
54
+
55
+ first_stage_config:
56
+ target: ldm.models.autoencoder.AutoencoderKL
57
+ params:
58
+ embed_dim: 4
59
+ monitor: val/rec_loss
60
+ ddconfig:
61
+ double_z: true
62
+ z_channels: 4
63
+ resolution: 256
64
+ in_channels: 3
65
+ out_ch: 3
66
+ ch: 128
67
+ ch_mult:
68
+ - 1
69
+ - 2
70
+ - 4
71
+ - 4
72
+ num_res_blocks: 2
73
+ attn_resolutions: []
74
+ dropout: 0.0
75
+ lossconfig:
76
+ target: torch.nn.Identity
77
+
78
+ cond_stage_config:
79
+ target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
control_v11p_sd15_depth.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ed111a879e8236ed3940068251dfe54e8e4543cb1b280f6ff6cb4023adc9d3a
3
+ size 1445234681
control_v11p_sd15_depth.yaml ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model:
2
+ target: cldm.cldm.ControlLDM
3
+ params:
4
+ linear_start: 0.00085
5
+ linear_end: 0.0120
6
+ num_timesteps_cond: 1
7
+ log_every_t: 200
8
+ timesteps: 1000
9
+ first_stage_key: "jpg"
10
+ cond_stage_key: "txt"
11
+ control_key: "hint"
12
+ image_size: 64
13
+ channels: 4
14
+ cond_stage_trainable: false
15
+ conditioning_key: crossattn
16
+ monitor: val/loss_simple_ema
17
+ scale_factor: 0.18215
18
+ use_ema: False
19
+ only_mid_control: False
20
+
21
+ control_stage_config:
22
+ target: cldm.cldm.ControlNet
23
+ params:
24
+ image_size: 32 # unused
25
+ in_channels: 4
26
+ hint_channels: 3
27
+ model_channels: 320
28
+ attention_resolutions: [ 4, 2, 1 ]
29
+ num_res_blocks: 2
30
+ channel_mult: [ 1, 2, 4, 4 ]
31
+ num_heads: 8
32
+ use_spatial_transformer: True
33
+ transformer_depth: 1
34
+ context_dim: 768
35
+ use_checkpoint: True
36
+ legacy: False
37
+
38
+ unet_config:
39
+ target: cldm.cldm.ControlledUnetModel
40
+ params:
41
+ image_size: 32 # unused
42
+ in_channels: 4
43
+ out_channels: 4
44
+ model_channels: 320
45
+ attention_resolutions: [ 4, 2, 1 ]
46
+ num_res_blocks: 2
47
+ channel_mult: [ 1, 2, 4, 4 ]
48
+ num_heads: 8
49
+ use_spatial_transformer: True
50
+ transformer_depth: 1
51
+ context_dim: 768
52
+ use_checkpoint: True
53
+ legacy: False
54
+
55
+ first_stage_config:
56
+ target: ldm.models.autoencoder.AutoencoderKL
57
+ params:
58
+ embed_dim: 4
59
+ monitor: val/rec_loss
60
+ ddconfig:
61
+ double_z: true
62
+ z_channels: 4
63
+ resolution: 256
64
+ in_channels: 3
65
+ out_ch: 3
66
+ ch: 128
67
+ ch_mult:
68
+ - 1
69
+ - 2
70
+ - 4
71
+ - 4
72
+ num_res_blocks: 2
73
+ attn_resolutions: []
74
+ dropout: 0.0
75
+ lossconfig:
76
+ target: torch.nn.Identity
77
+
78
+ cond_stage_config:
79
+ target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
control_v11p_sd15_inpaint.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2a0ff91882a0fbacbded5edeca7babdb6d1bd8f770880a91c177bd44cd58633
3
+ size 1445235365
control_v11p_sd15_inpaint.yaml ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model:
2
+ target: cldm.cldm.ControlLDM
3
+ params:
4
+ linear_start: 0.00085
5
+ linear_end: 0.0120
6
+ num_timesteps_cond: 1
7
+ log_every_t: 200
8
+ timesteps: 1000
9
+ first_stage_key: "jpg"
10
+ cond_stage_key: "txt"
11
+ control_key: "hint"
12
+ image_size: 64
13
+ channels: 4
14
+ cond_stage_trainable: false
15
+ conditioning_key: crossattn
16
+ monitor: val/loss_simple_ema
17
+ scale_factor: 0.18215
18
+ use_ema: False
19
+ only_mid_control: False
20
+
21
+ control_stage_config:
22
+ target: cldm.cldm.ControlNet
23
+ params:
24
+ image_size: 32 # unused
25
+ in_channels: 4
26
+ hint_channels: 3
27
+ model_channels: 320
28
+ attention_resolutions: [ 4, 2, 1 ]
29
+ num_res_blocks: 2
30
+ channel_mult: [ 1, 2, 4, 4 ]
31
+ num_heads: 8
32
+ use_spatial_transformer: True
33
+ transformer_depth: 1
34
+ context_dim: 768
35
+ use_checkpoint: True
36
+ legacy: False
37
+
38
+ unet_config:
39
+ target: cldm.cldm.ControlledUnetModel
40
+ params:
41
+ image_size: 32 # unused
42
+ in_channels: 4
43
+ out_channels: 4
44
+ model_channels: 320
45
+ attention_resolutions: [ 4, 2, 1 ]
46
+ num_res_blocks: 2
47
+ channel_mult: [ 1, 2, 4, 4 ]
48
+ num_heads: 8
49
+ use_spatial_transformer: True
50
+ transformer_depth: 1
51
+ context_dim: 768
52
+ use_checkpoint: True
53
+ legacy: False
54
+
55
+ first_stage_config:
56
+ target: ldm.models.autoencoder.AutoencoderKL
57
+ params:
58
+ embed_dim: 4
59
+ monitor: val/rec_loss
60
+ ddconfig:
61
+ double_z: true
62
+ z_channels: 4
63
+ resolution: 256
64
+ in_channels: 3
65
+ out_ch: 3
66
+ ch: 128
67
+ ch_mult:
68
+ - 1
69
+ - 2
70
+ - 4
71
+ - 4
72
+ num_res_blocks: 2
73
+ attn_resolutions: []
74
+ dropout: 0.0
75
+ lossconfig:
76
+ target: torch.nn.Identity
77
+
78
+ cond_stage_config:
79
+ target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
control_v11p_sd15_lineart.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb5fb912696830da948e5056b50adfa3fcc68ff760c43e4e7a8499853af43c61
3
+ size 1445235365
control_v11p_sd15_lineart.yaml ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model:
2
+ target: cldm.cldm.ControlLDM
3
+ params:
4
+ linear_start: 0.00085
5
+ linear_end: 0.0120
6
+ num_timesteps_cond: 1
7
+ log_every_t: 200
8
+ timesteps: 1000
9
+ first_stage_key: "jpg"
10
+ cond_stage_key: "txt"
11
+ control_key: "hint"
12
+ image_size: 64
13
+ channels: 4
14
+ cond_stage_trainable: false
15
+ conditioning_key: crossattn
16
+ monitor: val/loss_simple_ema
17
+ scale_factor: 0.18215
18
+ use_ema: False
19
+ only_mid_control: False
20
+
21
+ control_stage_config:
22
+ target: cldm.cldm.ControlNet
23
+ params:
24
+ image_size: 32 # unused
25
+ in_channels: 4
26
+ hint_channels: 3
27
+ model_channels: 320
28
+ attention_resolutions: [ 4, 2, 1 ]
29
+ num_res_blocks: 2
30
+ channel_mult: [ 1, 2, 4, 4 ]
31
+ num_heads: 8
32
+ use_spatial_transformer: True
33
+ transformer_depth: 1
34
+ context_dim: 768
35
+ use_checkpoint: True
36
+ legacy: False
37
+
38
+ unet_config:
39
+ target: cldm.cldm.ControlledUnetModel
40
+ params:
41
+ image_size: 32 # unused
42
+ in_channels: 4
43
+ out_channels: 4
44
+ model_channels: 320
45
+ attention_resolutions: [ 4, 2, 1 ]
46
+ num_res_blocks: 2
47
+ channel_mult: [ 1, 2, 4, 4 ]
48
+ num_heads: 8
49
+ use_spatial_transformer: True
50
+ transformer_depth: 1
51
+ context_dim: 768
52
+ use_checkpoint: True
53
+ legacy: False
54
+
55
+ first_stage_config:
56
+ target: ldm.models.autoencoder.AutoencoderKL
57
+ params:
58
+ embed_dim: 4
59
+ monitor: val/rec_loss
60
+ ddconfig:
61
+ double_z: true
62
+ z_channels: 4
63
+ resolution: 256
64
+ in_channels: 3
65
+ out_ch: 3
66
+ ch: 128
67
+ ch_mult:
68
+ - 1
69
+ - 2
70
+ - 4
71
+ - 4
72
+ num_res_blocks: 2
73
+ attn_resolutions: []
74
+ dropout: 0.0
75
+ lossconfig:
76
+ target: torch.nn.Identity
77
+
78
+ cond_stage_config:
79
+ target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
control_v11p_sd15_mlsd.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:22dab177f4c4566ac1c8f68dc2c429b023114e15a644e8c7c0e8f956afe92acf
3
+ size 1445234339
control_v11p_sd15_mlsd.yaml ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model:
2
+ target: cldm.cldm.ControlLDM
3
+ params:
4
+ linear_start: 0.00085
5
+ linear_end: 0.0120
6
+ num_timesteps_cond: 1
7
+ log_every_t: 200
8
+ timesteps: 1000
9
+ first_stage_key: "jpg"
10
+ cond_stage_key: "txt"
11
+ control_key: "hint"
12
+ image_size: 64
13
+ channels: 4
14
+ cond_stage_trainable: false
15
+ conditioning_key: crossattn
16
+ monitor: val/loss_simple_ema
17
+ scale_factor: 0.18215
18
+ use_ema: False
19
+ only_mid_control: False
20
+
21
+ control_stage_config:
22
+ target: cldm.cldm.ControlNet
23
+ params:
24
+ image_size: 32 # unused
25
+ in_channels: 4
26
+ hint_channels: 3
27
+ model_channels: 320
28
+ attention_resolutions: [ 4, 2, 1 ]
29
+ num_res_blocks: 2
30
+ channel_mult: [ 1, 2, 4, 4 ]
31
+ num_heads: 8
32
+ use_spatial_transformer: True
33
+ transformer_depth: 1
34
+ context_dim: 768
35
+ use_checkpoint: True
36
+ legacy: False
37
+
38
+ unet_config:
39
+ target: cldm.cldm.ControlledUnetModel
40
+ params:
41
+ image_size: 32 # unused
42
+ in_channels: 4
43
+ out_channels: 4
44
+ model_channels: 320
45
+ attention_resolutions: [ 4, 2, 1 ]
46
+ num_res_blocks: 2
47
+ channel_mult: [ 1, 2, 4, 4 ]
48
+ num_heads: 8
49
+ use_spatial_transformer: True
50
+ transformer_depth: 1
51
+ context_dim: 768
52
+ use_checkpoint: True
53
+ legacy: False
54
+
55
+ first_stage_config:
56
+ target: ldm.models.autoencoder.AutoencoderKL
57
+ params:
58
+ embed_dim: 4
59
+ monitor: val/rec_loss
60
+ ddconfig:
61
+ double_z: true
62
+ z_channels: 4
63
+ resolution: 256
64
+ in_channels: 3
65
+ out_ch: 3
66
+ ch: 128
67
+ ch_mult:
68
+ - 1
69
+ - 2
70
+ - 4
71
+ - 4
72
+ num_res_blocks: 2
73
+ attn_resolutions: []
74
+ dropout: 0.0
75
+ lossconfig:
76
+ target: torch.nn.Identity
77
+
78
+ cond_stage_config:
79
+ target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
control_v11p_sd15_normalbae.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9608158c204261259f4b5a7815ced8f5c15e5de4e30a9403d07f68f29f81e941
3
+ size 1445236049
control_v11p_sd15_normalbae.yaml ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model:
2
+ target: cldm.cldm.ControlLDM
3
+ params:
4
+ linear_start: 0.00085
5
+ linear_end: 0.0120
6
+ num_timesteps_cond: 1
7
+ log_every_t: 200
8
+ timesteps: 1000
9
+ first_stage_key: "jpg"
10
+ cond_stage_key: "txt"
11
+ control_key: "hint"
12
+ image_size: 64
13
+ channels: 4
14
+ cond_stage_trainable: false
15
+ conditioning_key: crossattn
16
+ monitor: val/loss_simple_ema
17
+ scale_factor: 0.18215
18
+ use_ema: False
19
+ only_mid_control: False
20
+
21
+ control_stage_config:
22
+ target: cldm.cldm.ControlNet
23
+ params:
24
+ image_size: 32 # unused
25
+ in_channels: 4
26
+ hint_channels: 3
27
+ model_channels: 320
28
+ attention_resolutions: [ 4, 2, 1 ]
29
+ num_res_blocks: 2
30
+ channel_mult: [ 1, 2, 4, 4 ]
31
+ num_heads: 8
32
+ use_spatial_transformer: True
33
+ transformer_depth: 1
34
+ context_dim: 768
35
+ use_checkpoint: True
36
+ legacy: False
37
+
38
+ unet_config:
39
+ target: cldm.cldm.ControlledUnetModel
40
+ params:
41
+ image_size: 32 # unused
42
+ in_channels: 4
43
+ out_channels: 4
44
+ model_channels: 320
45
+ attention_resolutions: [ 4, 2, 1 ]
46
+ num_res_blocks: 2
47
+ channel_mult: [ 1, 2, 4, 4 ]
48
+ num_heads: 8
49
+ use_spatial_transformer: True
50
+ transformer_depth: 1
51
+ context_dim: 768
52
+ use_checkpoint: True
53
+ legacy: False
54
+
55
+ first_stage_config:
56
+ target: ldm.models.autoencoder.AutoencoderKL
57
+ params:
58
+ embed_dim: 4
59
+ monitor: val/rec_loss
60
+ ddconfig:
61
+ double_z: true
62
+ z_channels: 4
63
+ resolution: 256
64
+ in_channels: 3
65
+ out_ch: 3
66
+ ch: 128
67
+ ch_mult:
68
+ - 1
69
+ - 2
70
+ - 4
71
+ - 4
72
+ num_res_blocks: 2
73
+ attn_resolutions: []
74
+ dropout: 0.0
75
+ lossconfig:
76
+ target: torch.nn.Identity
77
+
78
+ cond_stage_config:
79
+ target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
control_v11p_sd15_openpose.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db97becd92cd19aff71352a60e93c2508decba3dee64f01f686727b9b406a9dd
3
+ size 1445235707
control_v11p_sd15_openpose.yaml ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model:
2
+ target: cldm.cldm.ControlLDM
3
+ params:
4
+ linear_start: 0.00085
5
+ linear_end: 0.0120
6
+ num_timesteps_cond: 1
7
+ log_every_t: 200
8
+ timesteps: 1000
9
+ first_stage_key: "jpg"
10
+ cond_stage_key: "txt"
11
+ control_key: "hint"
12
+ image_size: 64
13
+ channels: 4
14
+ cond_stage_trainable: false
15
+ conditioning_key: crossattn
16
+ monitor: val/loss_simple_ema
17
+ scale_factor: 0.18215
18
+ use_ema: False
19
+ only_mid_control: False
20
+
21
+ control_stage_config:
22
+ target: cldm.cldm.ControlNet
23
+ params:
24
+ image_size: 32 # unused
25
+ in_channels: 4
26
+ hint_channels: 3
27
+ model_channels: 320
28
+ attention_resolutions: [ 4, 2, 1 ]
29
+ num_res_blocks: 2
30
+ channel_mult: [ 1, 2, 4, 4 ]
31
+ num_heads: 8
32
+ use_spatial_transformer: True
33
+ transformer_depth: 1
34
+ context_dim: 768
35
+ use_checkpoint: True
36
+ legacy: False
37
+
38
+ unet_config:
39
+ target: cldm.cldm.ControlledUnetModel
40
+ params:
41
+ image_size: 32 # unused
42
+ in_channels: 4
43
+ out_channels: 4
44
+ model_channels: 320
45
+ attention_resolutions: [ 4, 2, 1 ]
46
+ num_res_blocks: 2
47
+ channel_mult: [ 1, 2, 4, 4 ]
48
+ num_heads: 8
49
+ use_spatial_transformer: True
50
+ transformer_depth: 1
51
+ context_dim: 768
52
+ use_checkpoint: True
53
+ legacy: False
54
+
55
+ first_stage_config:
56
+ target: ldm.models.autoencoder.AutoencoderKL
57
+ params:
58
+ embed_dim: 4
59
+ monitor: val/rec_loss
60
+ ddconfig:
61
+ double_z: true
62
+ z_channels: 4
63
+ resolution: 256
64
+ in_channels: 3
65
+ out_ch: 3
66
+ ch: 128
67
+ ch_mult:
68
+ - 1
69
+ - 2
70
+ - 4
71
+ - 4
72
+ num_res_blocks: 2
73
+ attn_resolutions: []
74
+ dropout: 0.0
75
+ lossconfig:
76
+ target: torch.nn.Identity
77
+
78
+ cond_stage_config:
79
+ target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
control_v11p_sd15_scribble.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:14d33e8bc14ba679595d976789e4eedc36a3d2d21866ab5896355a7e6677749f
3
+ size 1445235707
control_v11p_sd15_scribble.yaml ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model:
2
+ target: cldm.cldm.ControlLDM
3
+ params:
4
+ linear_start: 0.00085
5
+ linear_end: 0.0120
6
+ num_timesteps_cond: 1
7
+ log_every_t: 200
8
+ timesteps: 1000
9
+ first_stage_key: "jpg"
10
+ cond_stage_key: "txt"
11
+ control_key: "hint"
12
+ image_size: 64
13
+ channels: 4
14
+ cond_stage_trainable: false
15
+ conditioning_key: crossattn
16
+ monitor: val/loss_simple_ema
17
+ scale_factor: 0.18215
18
+ use_ema: False
19
+ only_mid_control: False
20
+
21
+ control_stage_config:
22
+ target: cldm.cldm.ControlNet
23
+ params:
24
+ image_size: 32 # unused
25
+ in_channels: 4
26
+ hint_channels: 3
27
+ model_channels: 320
28
+ attention_resolutions: [ 4, 2, 1 ]
29
+ num_res_blocks: 2
30
+ channel_mult: [ 1, 2, 4, 4 ]
31
+ num_heads: 8
32
+ use_spatial_transformer: True
33
+ transformer_depth: 1
34
+ context_dim: 768
35
+ use_checkpoint: True
36
+ legacy: False
37
+
38
+ unet_config:
39
+ target: cldm.cldm.ControlledUnetModel
40
+ params:
41
+ image_size: 32 # unused
42
+ in_channels: 4
43
+ out_channels: 4
44
+ model_channels: 320
45
+ attention_resolutions: [ 4, 2, 1 ]
46
+ num_res_blocks: 2
47
+ channel_mult: [ 1, 2, 4, 4 ]
48
+ num_heads: 8
49
+ use_spatial_transformer: True
50
+ transformer_depth: 1
51
+ context_dim: 768
52
+ use_checkpoint: True
53
+ legacy: False
54
+
55
+ first_stage_config:
56
+ target: ldm.models.autoencoder.AutoencoderKL
57
+ params:
58
+ embed_dim: 4
59
+ monitor: val/rec_loss
60
+ ddconfig:
61
+ double_z: true
62
+ z_channels: 4
63
+ resolution: 256
64
+ in_channels: 3
65
+ out_ch: 3
66
+ ch: 128
67
+ ch_mult:
68
+ - 1
69
+ - 2
70
+ - 4
71
+ - 4
72
+ num_res_blocks: 2
73
+ attn_resolutions: []
74
+ dropout: 0.0
75
+ lossconfig:
76
+ target: torch.nn.Identity
77
+
78
+ cond_stage_config:
79
+ target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
control_v11p_sd15_seg.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bad6d753886c0edc978e79dfb06083737ab1bd9702e5b7c0f81b1e16a5ed12cc
3
+ size 1445233933
control_v11p_sd15_seg.yaml ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model:
2
+ target: cldm.cldm.ControlLDM
3
+ params:
4
+ linear_start: 0.00085
5
+ linear_end: 0.0120
6
+ num_timesteps_cond: 1
7
+ log_every_t: 200
8
+ timesteps: 1000
9
+ first_stage_key: "jpg"
10
+ cond_stage_key: "txt"
11
+ control_key: "hint"
12
+ image_size: 64
13
+ channels: 4
14
+ cond_stage_trainable: false
15
+ conditioning_key: crossattn
16
+ monitor: val/loss_simple_ema
17
+ scale_factor: 0.18215
18
+ use_ema: False
19
+ only_mid_control: False
20
+
21
+ control_stage_config:
22
+ target: cldm.cldm.ControlNet
23
+ params:
24
+ image_size: 32 # unused
25
+ in_channels: 4
26
+ hint_channels: 3
27
+ model_channels: 320
28
+ attention_resolutions: [ 4, 2, 1 ]
29
+ num_res_blocks: 2
30
+ channel_mult: [ 1, 2, 4, 4 ]
31
+ num_heads: 8
32
+ use_spatial_transformer: True
33
+ transformer_depth: 1
34
+ context_dim: 768
35
+ use_checkpoint: True
36
+ legacy: False
37
+
38
+ unet_config:
39
+ target: cldm.cldm.ControlledUnetModel
40
+ params:
41
+ image_size: 32 # unused
42
+ in_channels: 4
43
+ out_channels: 4
44
+ model_channels: 320
45
+ attention_resolutions: [ 4, 2, 1 ]
46
+ num_res_blocks: 2
47
+ channel_mult: [ 1, 2, 4, 4 ]
48
+ num_heads: 8
49
+ use_spatial_transformer: True
50
+ transformer_depth: 1
51
+ context_dim: 768
52
+ use_checkpoint: True
53
+ legacy: False
54
+
55
+ first_stage_config:
56
+ target: ldm.models.autoencoder.AutoencoderKL
57
+ params:
58
+ embed_dim: 4
59
+ monitor: val/rec_loss
60
+ ddconfig:
61
+ double_z: true
62
+ z_channels: 4
63
+ resolution: 256
64
+ in_channels: 3
65
+ out_ch: 3
66
+ ch: 128
67
+ ch_mult:
68
+ - 1
69
+ - 2
70
+ - 4
71
+ - 4
72
+ num_res_blocks: 2
73
+ attn_resolutions: []
74
+ dropout: 0.0
75
+ lossconfig:
76
+ target: torch.nn.Identity
77
+
78
+ cond_stage_config:
79
+ target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
control_v11p_sd15_softedge.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20759ef71a109c4796e637d86eea8db492ec9772b6d5cb7a9afd077847b403f3
3
+ size 1445235707
control_v11p_sd15_softedge.yaml ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model:
2
+ target: cldm.cldm.ControlLDM
3
+ params:
4
+ linear_start: 0.00085
5
+ linear_end: 0.0120
6
+ num_timesteps_cond: 1
7
+ log_every_t: 200
8
+ timesteps: 1000
9
+ first_stage_key: "jpg"
10
+ cond_stage_key: "txt"
11
+ control_key: "hint"
12
+ image_size: 64
13
+ channels: 4
14
+ cond_stage_trainable: false
15
+ conditioning_key: crossattn
16
+ monitor: val/loss_simple_ema
17
+ scale_factor: 0.18215
18
+ use_ema: False
19
+ only_mid_control: False
20
+
21
+ control_stage_config:
22
+ target: cldm.cldm.ControlNet
23
+ params:
24
+ image_size: 32 # unused
25
+ in_channels: 4
26
+ hint_channels: 3
27
+ model_channels: 320
28
+ attention_resolutions: [ 4, 2, 1 ]
29
+ num_res_blocks: 2
30
+ channel_mult: [ 1, 2, 4, 4 ]
31
+ num_heads: 8
32
+ use_spatial_transformer: True
33
+ transformer_depth: 1
34
+ context_dim: 768
35
+ use_checkpoint: True
36
+ legacy: False
37
+
38
+ unet_config:
39
+ target: cldm.cldm.ControlledUnetModel
40
+ params:
41
+ image_size: 32 # unused
42
+ in_channels: 4
43
+ out_channels: 4
44
+ model_channels: 320
45
+ attention_resolutions: [ 4, 2, 1 ]
46
+ num_res_blocks: 2
47
+ channel_mult: [ 1, 2, 4, 4 ]
48
+ num_heads: 8
49
+ use_spatial_transformer: True
50
+ transformer_depth: 1
51
+ context_dim: 768
52
+ use_checkpoint: True
53
+ legacy: False
54
+
55
+ first_stage_config:
56
+ target: ldm.models.autoencoder.AutoencoderKL
57
+ params:
58
+ embed_dim: 4
59
+ monitor: val/rec_loss
60
+ ddconfig:
61
+ double_z: true
62
+ z_channels: 4
63
+ resolution: 256
64
+ in_channels: 3
65
+ out_ch: 3
66
+ ch: 128
67
+ ch_mult:
68
+ - 1
69
+ - 2
70
+ - 4
71
+ - 4
72
+ num_res_blocks: 2
73
+ attn_resolutions: []
74
+ dropout: 0.0
75
+ lossconfig:
76
+ target: torch.nn.Identity
77
+
78
+ cond_stage_config:
79
+ target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
control_v11p_sd15s2_lineart_anime.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:59169ab86e8ae15ad50208cc5832f1cc598b95391d4482ba742b344c8f7f6904
3
+ size 1445238101
control_v11p_sd15s2_lineart_anime.yaml ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model:
2
+ target: cldm.cldm.ControlLDM
3
+ params:
4
+ linear_start: 0.00085
5
+ linear_end: 0.0120
6
+ num_timesteps_cond: 1
7
+ log_every_t: 200
8
+ timesteps: 1000
9
+ first_stage_key: "jpg"
10
+ cond_stage_key: "txt"
11
+ control_key: "hint"
12
+ image_size: 64
13
+ channels: 4
14
+ cond_stage_trainable: false
15
+ conditioning_key: crossattn
16
+ monitor: val/loss_simple_ema
17
+ scale_factor: 0.18215
18
+ use_ema: False
19
+ only_mid_control: False
20
+
21
+ control_stage_config:
22
+ target: cldm.cldm.ControlNet
23
+ params:
24
+ image_size: 32 # unused
25
+ in_channels: 4
26
+ hint_channels: 3
27
+ model_channels: 320
28
+ attention_resolutions: [ 4, 2, 1 ]
29
+ num_res_blocks: 2
30
+ channel_mult: [ 1, 2, 4, 4 ]
31
+ num_heads: 8
32
+ use_spatial_transformer: True
33
+ transformer_depth: 1
34
+ context_dim: 768
35
+ use_checkpoint: True
36
+ legacy: False
37
+
38
+ unet_config:
39
+ target: cldm.cldm.ControlledUnetModel
40
+ params:
41
+ image_size: 32 # unused
42
+ in_channels: 4
43
+ out_channels: 4
44
+ model_channels: 320
45
+ attention_resolutions: [ 4, 2, 1 ]
46
+ num_res_blocks: 2
47
+ channel_mult: [ 1, 2, 4, 4 ]
48
+ num_heads: 8
49
+ use_spatial_transformer: True
50
+ transformer_depth: 1
51
+ context_dim: 768
52
+ use_checkpoint: True
53
+ legacy: False
54
+
55
+ first_stage_config:
56
+ target: ldm.models.autoencoder.AutoencoderKL
57
+ params:
58
+ embed_dim: 4
59
+ monitor: val/rec_loss
60
+ ddconfig:
61
+ double_z: true
62
+ z_channels: 4
63
+ resolution: 256
64
+ in_channels: 3
65
+ out_ch: 3
66
+ ch: 128
67
+ ch_mult:
68
+ - 1
69
+ - 2
70
+ - 4
71
+ - 4
72
+ num_res_blocks: 2
73
+ attn_resolutions: []
74
+ dropout: 0.0
75
+ lossconfig:
76
+ target: torch.nn.Identity
77
+
78
+ cond_stage_config:
79
+ target: ldm.modules.encoders.modules.FrozenCLIPEmbedder
control_v11u_sd15_tile.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2724efde11112944d9fe65db9cf4513fc11f6c76b3db7d7bb67ac589e03be8e
3
+ size 1445234339
control_v11u_sd15_tile.yaml ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model:
2
+ target: cldm.cldm.ControlLDM
3
+ params:
4
+ linear_start: 0.00085
5
+ linear_end: 0.0120
6
+ num_timesteps_cond: 1
7
+ log_every_t: 200
8
+ timesteps: 1000
9
+ first_stage_key: "jpg"
10
+ cond_stage_key: "txt"
11
+ control_key: "hint"
12
+ image_size: 64
13
+ channels: 4
14
+ cond_stage_trainable: false
15
+ conditioning_key: crossattn
16
+ monitor: val/loss_simple_ema
17
+ scale_factor: 0.18215
18
+ use_ema: False
19
+ only_mid_control: False
20
+
21
+ control_stage_config:
22
+ target: cldm.cldm.ControlNet
23
+ params:
24
+ image_size: 32 # unused
25
+ in_channels: 4
26
+ hint_channels: 3
27
+ model_channels: 320
28
+ attention_resolutions: [ 4, 2, 1 ]
29
+ num_res_blocks: 2
30
+ channel_mult: [ 1, 2, 4, 4 ]
31
+ num_heads: 8
32
+ use_spatial_transformer: True
33
+ transformer_depth: 1
34
+ context_dim: 768
35
+ use_checkpoint: True
36
+ legacy: False
37
+
38
+ unet_config:
39
+ target: cldm.cldm.ControlledUnetModel
40
+ params:
41
+ image_size: 32 # unused
42
+ in_channels: 4
43
+ out_channels: 4
44
+ model_channels: 320
45
+ attention_resolutions: [ 4, 2, 1 ]
46
+ num_res_blocks: 2
47
+ channel_mult: [ 1, 2, 4, 4 ]
48
+ num_heads: 8
49
+ use_spatial_transformer: True
50
+ transformer_depth: 1
51
+ context_dim: 768
52
+ use_checkpoint: True
53
+ legacy: False
54
+
55
+ first_stage_config:
56
+ target: ldm.models.autoencoder.AutoencoderKL
57
+ params:
58
+ embed_dim: 4
59
+ monitor: val/rec_loss
60
+ ddconfig:
61
+ double_z: true
62
+ z_channels: 4
63
+ resolution: 256
64
+ in_channels: 3
65
+ out_ch: 3
66
+ ch: 128
67
+ ch_mult:
68
+ - 1
69
+ - 2
70
+ - 4
71
+ - 4
72
+ num_res_blocks: 2
73
+ attn_resolutions: []
74
+ dropout: 0.0
75
+ lossconfig:
76
+ target: torch.nn.Identity
77
+
78
+ cond_stage_config:
79
+ target: ldm.modules.encoders.modules.FrozenCLIPEmbedder