Upload 4 files
Browse files
extracted-depth-sd21-diff-half.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ed33ec158b6ea3bf61494ee71adc16d45886d5f6611d108345a184ebe7d08d81
|
3 |
+
size 728494536
|
extracted-depth-sd21-diff-half.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"
|
extracted-depth-sd21-diff.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:74fd83d9b3ffde17f3eb4de31934bac32c96ff5e97765930f3e2dae6258a61b0
|
3 |
+
size 1456951265
|
extracted-depth-sd21-diff.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"
|