camenduru commited on
Commit
0f5dbfe
1 Parent(s): cbbd2e9

thanks to furusu ❤ and alfredplpl ❤

Browse files
diff_control_wd15beta2_canny.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e76d62c67b4a2b193bb46ee2fafcffedf40e06717851e7b9a00c7d555107d440
3
+ size 728499375
diff_control_wd15beta2_canny.yaml ADDED
@@ -0,0 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model:
2
+ target: cldm.cldm.ControlLDM
3
+ params:
4
+ parameterization: "v"
5
+ linear_start: 0.00085
6
+ linear_end: 0.0120
7
+ num_timesteps_cond: 1
8
+ log_every_t: 200
9
+ timesteps: 1000
10
+ first_stage_key: "jpg"
11
+ cond_stage_key: "txt"
12
+ control_key: "hint"
13
+ image_size: 64
14
+ channels: 4
15
+ cond_stage_trainable: false
16
+ conditioning_key: crossattn
17
+ monitor: val/loss_simple_ema
18
+ scale_factor: 0.18215
19
+ use_ema: False
20
+ only_mid_control: False
21
+
22
+ control_stage_config:
23
+ target: cldm.cldm.ControlNet
24
+ params:
25
+ use_checkpoint: True
26
+ image_size: 32 # unused
27
+ in_channels: 4
28
+ hint_channels: 3
29
+ model_channels: 320
30
+ attention_resolutions: [ 4, 2, 1 ]
31
+ num_res_blocks: 2
32
+ channel_mult: [ 1, 2, 4, 4 ]
33
+ num_head_channels: 64 # need to fix for flash-attn
34
+ use_spatial_transformer: True
35
+ use_linear_in_transformer: True
36
+ transformer_depth: 1
37
+ context_dim: 1024
38
+ legacy: False
39
+
40
+ unet_config:
41
+ target: cldm.cldm.ControlledUnetModel
42
+ params:
43
+ use_checkpoint: True
44
+ image_size: 32 # unused
45
+ in_channels: 4
46
+ out_channels: 4
47
+ model_channels: 320
48
+ attention_resolutions: [ 4, 2, 1 ]
49
+ num_res_blocks: 2
50
+ channel_mult: [ 1, 2, 4, 4 ]
51
+ num_head_channels: 64 # need to fix for flash-attn
52
+ use_spatial_transformer: True
53
+ use_linear_in_transformer: True
54
+ transformer_depth: 1
55
+ context_dim: 1024
56
+ legacy: False
57
+
58
+ first_stage_config:
59
+ target: ldm.models.autoencoder.AutoencoderKL
60
+ params:
61
+ embed_dim: 4
62
+ monitor: val/rec_loss
63
+ ddconfig:
64
+ #attn_type: "vanilla-xformers"
65
+ double_z: true
66
+ z_channels: 4
67
+ resolution: 256
68
+ in_channels: 3
69
+ out_ch: 3
70
+ ch: 128
71
+ ch_mult:
72
+ - 1
73
+ - 2
74
+ - 4
75
+ - 4
76
+ num_res_blocks: 2
77
+ attn_resolutions: []
78
+ dropout: 0.0
79
+ lossconfig:
80
+ target: torch.nn.Identity
81
+
82
+ cond_stage_config:
83
+ target: ldm.modules.encoders.modules.FrozenOpenCLIPEmbedder
84
+ params:
85
+ freeze: True
86
+ layer: "penultimate"
diff_control_wd15beta2_depth.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b19cd8dcd2d2f429273747bac8915967df0a355704010cce7fc8f757da532fe2
3
+ size 728499375
diff_control_wd15beta2_depth.yaml ADDED
@@ -0,0 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model:
2
+ target: cldm.cldm.ControlLDM
3
+ params:
4
+ parameterization: "v"
5
+ linear_start: 0.00085
6
+ linear_end: 0.0120
7
+ num_timesteps_cond: 1
8
+ log_every_t: 200
9
+ timesteps: 1000
10
+ first_stage_key: "jpg"
11
+ cond_stage_key: "txt"
12
+ control_key: "hint"
13
+ image_size: 64
14
+ channels: 4
15
+ cond_stage_trainable: false
16
+ conditioning_key: crossattn
17
+ monitor: val/loss_simple_ema
18
+ scale_factor: 0.18215
19
+ use_ema: False
20
+ only_mid_control: False
21
+
22
+ control_stage_config:
23
+ target: cldm.cldm.ControlNet
24
+ params:
25
+ use_checkpoint: True
26
+ image_size: 32 # unused
27
+ in_channels: 4
28
+ hint_channels: 3
29
+ model_channels: 320
30
+ attention_resolutions: [ 4, 2, 1 ]
31
+ num_res_blocks: 2
32
+ channel_mult: [ 1, 2, 4, 4 ]
33
+ num_head_channels: 64 # need to fix for flash-attn
34
+ use_spatial_transformer: True
35
+ use_linear_in_transformer: True
36
+ transformer_depth: 1
37
+ context_dim: 1024
38
+ legacy: False
39
+
40
+ unet_config:
41
+ target: cldm.cldm.ControlledUnetModel
42
+ params:
43
+ use_checkpoint: True
44
+ image_size: 32 # unused
45
+ in_channels: 4
46
+ out_channels: 4
47
+ model_channels: 320
48
+ attention_resolutions: [ 4, 2, 1 ]
49
+ num_res_blocks: 2
50
+ channel_mult: [ 1, 2, 4, 4 ]
51
+ num_head_channels: 64 # need to fix for flash-attn
52
+ use_spatial_transformer: True
53
+ use_linear_in_transformer: True
54
+ transformer_depth: 1
55
+ context_dim: 1024
56
+ legacy: False
57
+
58
+ first_stage_config:
59
+ target: ldm.models.autoencoder.AutoencoderKL
60
+ params:
61
+ embed_dim: 4
62
+ monitor: val/rec_loss
63
+ ddconfig:
64
+ #attn_type: "vanilla-xformers"
65
+ double_z: true
66
+ z_channels: 4
67
+ resolution: 256
68
+ in_channels: 3
69
+ out_ch: 3
70
+ ch: 128
71
+ ch_mult:
72
+ - 1
73
+ - 2
74
+ - 4
75
+ - 4
76
+ num_res_blocks: 2
77
+ attn_resolutions: []
78
+ dropout: 0.0
79
+ lossconfig:
80
+ target: torch.nn.Identity
81
+
82
+ cond_stage_config:
83
+ target: ldm.modules.encoders.modules.FrozenOpenCLIPEmbedder
84
+ params:
85
+ freeze: True
86
+ layer: "penultimate"
diff_control_wd15beta2_pose.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db7281a2026f3eb9ca8df7b6c25ac683ab6581e8968a2100532923d82fd6b6b3
3
+ size 728499375
diff_control_wd15beta2_pose.yaml ADDED
@@ -0,0 +1,86 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ model:
2
+ target: cldm.cldm.ControlLDM
3
+ params:
4
+ parameterization: "v"
5
+ linear_start: 0.00085
6
+ linear_end: 0.0120
7
+ num_timesteps_cond: 1
8
+ log_every_t: 200
9
+ timesteps: 1000
10
+ first_stage_key: "jpg"
11
+ cond_stage_key: "txt"
12
+ control_key: "hint"
13
+ image_size: 64
14
+ channels: 4
15
+ cond_stage_trainable: false
16
+ conditioning_key: crossattn
17
+ monitor: val/loss_simple_ema
18
+ scale_factor: 0.18215
19
+ use_ema: False
20
+ only_mid_control: False
21
+
22
+ control_stage_config:
23
+ target: cldm.cldm.ControlNet
24
+ params:
25
+ use_checkpoint: True
26
+ image_size: 32 # unused
27
+ in_channels: 4
28
+ hint_channels: 3
29
+ model_channels: 320
30
+ attention_resolutions: [ 4, 2, 1 ]
31
+ num_res_blocks: 2
32
+ channel_mult: [ 1, 2, 4, 4 ]
33
+ num_head_channels: 64 # need to fix for flash-attn
34
+ use_spatial_transformer: True
35
+ use_linear_in_transformer: True
36
+ transformer_depth: 1
37
+ context_dim: 1024
38
+ legacy: False
39
+
40
+ unet_config:
41
+ target: cldm.cldm.ControlledUnetModel
42
+ params:
43
+ use_checkpoint: True
44
+ image_size: 32 # unused
45
+ in_channels: 4
46
+ out_channels: 4
47
+ model_channels: 320
48
+ attention_resolutions: [ 4, 2, 1 ]
49
+ num_res_blocks: 2
50
+ channel_mult: [ 1, 2, 4, 4 ]
51
+ num_head_channels: 64 # need to fix for flash-attn
52
+ use_spatial_transformer: True
53
+ use_linear_in_transformer: True
54
+ transformer_depth: 1
55
+ context_dim: 1024
56
+ legacy: False
57
+
58
+ first_stage_config:
59
+ target: ldm.models.autoencoder.AutoencoderKL
60
+ params:
61
+ embed_dim: 4
62
+ monitor: val/rec_loss
63
+ ddconfig:
64
+ #attn_type: "vanilla-xformers"
65
+ double_z: true
66
+ z_channels: 4
67
+ resolution: 256
68
+ in_channels: 3
69
+ out_ch: 3
70
+ ch: 128
71
+ ch_mult:
72
+ - 1
73
+ - 2
74
+ - 4
75
+ - 4
76
+ num_res_blocks: 2
77
+ attn_resolutions: []
78
+ dropout: 0.0
79
+ lossconfig:
80
+ target: torch.nn.Identity
81
+
82
+ cond_stage_config:
83
+ target: ldm.modules.encoders.modules.FrozenOpenCLIPEmbedder
84
+ params:
85
+ freeze: True
86
+ layer: "penultimate"