adding model files
Browse files- global_sampler_clip/checkpoints/last.ckpt +3 -0
- global_sampler_clip/configs/2022-04-23T05-49-02-lightning.yaml +4 -0
- global_sampler_clip/configs/2022-04-23T05-49-02-project.yaml +62 -0
- local_sampler/checkpoints/last.ckpt +3 -0
- local_sampler_indoor/checkpoints/last.ckpt +3 -0
- local_sampler_indoor/configs/2022-04-26T13-11-31-lightning.yaml +4 -0
- local_sampler_indoor/configs/2022-04-26T13-11-31-project.yaml +89 -0
- local_sampler_outdoor/checkpoints/last.ckpt +3 -0
- local_sampler_outdoor/configs/2022-04-29T20-25-45-lightning.yaml +4 -0
- local_sampler_outdoor/configs/2022-04-29T20-25-45-project.yaml +89 -0
- sritmo.pth +3 -0
global_sampler_clip/checkpoints/last.ckpt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a645439032025805fe831f28c92e45381a40ad9c2760a5a9bd518f05c1f559cb
|
3 |
+
size 1508432949
|
global_sampler_clip/configs/2022-04-23T05-49-02-lightning.yaml
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
lightning:
|
2 |
+
trainer:
|
3 |
+
distributed_backend: ddp
|
4 |
+
gpus: 0,1,2,3,4,5,6,7
|
global_sampler_clip/configs/2022-04-23T05-49-02-project.yaml
ADDED
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
model:
|
2 |
+
base_learning_rate: 4.5e-06
|
3 |
+
target: taming.models.global_sampler.GlobalSamplerWithCLIP
|
4 |
+
params:
|
5 |
+
cond_stage_key: knn
|
6 |
+
transformer_config:
|
7 |
+
target: taming.modules.transformer.mingpt.GPT
|
8 |
+
params:
|
9 |
+
vocab_size: 1024
|
10 |
+
block_size: 133
|
11 |
+
n_layer: 12
|
12 |
+
n_head: 8
|
13 |
+
n_embd: 1024
|
14 |
+
first_stage_config:
|
15 |
+
target: taming.models.vqgan.VQCirModel
|
16 |
+
params:
|
17 |
+
ckpt_path:
|
18 |
+
embed_dim: 256
|
19 |
+
n_embed: 1024
|
20 |
+
ddconfig:
|
21 |
+
double_z: false
|
22 |
+
z_channels: 256
|
23 |
+
resolution: 256
|
24 |
+
in_channels: 3
|
25 |
+
out_ch: 3
|
26 |
+
ch: 128
|
27 |
+
ch_mult:
|
28 |
+
- 1
|
29 |
+
- 1
|
30 |
+
- 2
|
31 |
+
- 2
|
32 |
+
- 4
|
33 |
+
num_res_blocks: 2
|
34 |
+
attn_resolutions:
|
35 |
+
- 16
|
36 |
+
dropout: 0.0
|
37 |
+
lossconfig:
|
38 |
+
target: taming.modules.losses.DummyLoss
|
39 |
+
cond_stage_config:
|
40 |
+
target: taming.modules.misc.coord.CoordStage
|
41 |
+
params:
|
42 |
+
n_embed: 1024
|
43 |
+
down_factor: 16
|
44 |
+
data:
|
45 |
+
target: main.DataModuleFromConfig
|
46 |
+
params:
|
47 |
+
batch_size: 16
|
48 |
+
num_workers: 8
|
49 |
+
train:
|
50 |
+
target: taming.data.custom.CustomTrainHolistic
|
51 |
+
params:
|
52 |
+
training_images_list_file: /home/chenzhaoxi/imgs/train_norot.txt
|
53 |
+
size: 256
|
54 |
+
holistic: 128
|
55 |
+
clip_emb: clip_emb.npy
|
56 |
+
validation:
|
57 |
+
target: taming.data.custom.CustomTestHolistic
|
58 |
+
params:
|
59 |
+
test_images_list_file: /home/chenzhaoxi/imgs/test.txt
|
60 |
+
size: 256
|
61 |
+
holistic: 128
|
62 |
+
clip_emb: clip_emb.npy
|
local_sampler/checkpoints/last.ckpt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c5c1e693f346959b491d84f632c594c15ca88b9f4ec1210ca08f7a9f22b8ade8
|
3 |
+
size 1812376146
|
local_sampler_indoor/checkpoints/last.ckpt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4440bb6c35d1d7f8b76cf615e334c7494e9eb3c9e06a9b049ff707c312f50e1c
|
3 |
+
size 1812376146
|
local_sampler_indoor/configs/2022-04-26T13-11-31-lightning.yaml
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
lightning:
|
2 |
+
trainer:
|
3 |
+
distributed_backend: ddp
|
4 |
+
gpus: 0,1,2,3,4,5,6,7
|
local_sampler_indoor/configs/2022-04-26T13-11-31-project.yaml
ADDED
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
model:
|
2 |
+
base_learning_rate: 4.5e-06
|
3 |
+
target: taming.models.local_sampler.LocalSamplerDualCodebookPE
|
4 |
+
params:
|
5 |
+
cond_stage_key: coord
|
6 |
+
holistic_config:
|
7 |
+
target: taming.models.vqgan.VQCirModel
|
8 |
+
params:
|
9 |
+
ckpt_path:
|
10 |
+
embed_dim: 256
|
11 |
+
n_embed: 1024
|
12 |
+
ddconfig:
|
13 |
+
double_z: false
|
14 |
+
z_channels: 256
|
15 |
+
resolution: 256
|
16 |
+
in_channels: 3
|
17 |
+
out_ch: 3
|
18 |
+
ch: 128
|
19 |
+
ch_mult:
|
20 |
+
- 1
|
21 |
+
- 1
|
22 |
+
- 2
|
23 |
+
- 2
|
24 |
+
- 4
|
25 |
+
num_res_blocks: 2
|
26 |
+
attn_resolutions:
|
27 |
+
- 16
|
28 |
+
dropout: 0.0
|
29 |
+
lossconfig:
|
30 |
+
target: taming.modules.losses.DummyLoss
|
31 |
+
transformer_config:
|
32 |
+
target: taming.modules.transformer.mingpt.GPT
|
33 |
+
params:
|
34 |
+
vocab_size: 1024
|
35 |
+
block_size: 402
|
36 |
+
n_layer: 24
|
37 |
+
n_head: 16
|
38 |
+
n_embd: 1024
|
39 |
+
first_stage_config:
|
40 |
+
target: taming.models.vqgan.VQModel
|
41 |
+
params:
|
42 |
+
ckpt_path:
|
43 |
+
embed_dim: 256
|
44 |
+
n_embed: 1024
|
45 |
+
ddconfig:
|
46 |
+
double_z: false
|
47 |
+
z_channels: 256
|
48 |
+
resolution: 256
|
49 |
+
in_channels: 3
|
50 |
+
out_ch: 3
|
51 |
+
ch: 128
|
52 |
+
ch_mult:
|
53 |
+
- 1
|
54 |
+
- 1
|
55 |
+
- 2
|
56 |
+
- 2
|
57 |
+
- 4
|
58 |
+
num_res_blocks: 2
|
59 |
+
attn_resolutions:
|
60 |
+
- 16
|
61 |
+
dropout: 0.0
|
62 |
+
lossconfig:
|
63 |
+
target: taming.modules.losses.DummyLoss
|
64 |
+
cond_stage_config:
|
65 |
+
target: taming.modules.misc.coord.CoordStage
|
66 |
+
params:
|
67 |
+
n_embed: 1024
|
68 |
+
down_factor: 16
|
69 |
+
data:
|
70 |
+
target: main.DataModuleFromConfig
|
71 |
+
params:
|
72 |
+
batch_size: 24
|
73 |
+
num_workers: 16
|
74 |
+
train:
|
75 |
+
target: taming.data.custom.CustomTrain
|
76 |
+
params:
|
77 |
+
training_images_list_file: /home/chenzhaoxi/imgs/train_indoor.txt
|
78 |
+
size: 256
|
79 |
+
coord: true
|
80 |
+
random_crop: true
|
81 |
+
holistic: 128
|
82 |
+
validation:
|
83 |
+
target: taming.data.custom.CustomTest
|
84 |
+
params:
|
85 |
+
test_images_list_file: /home/chenzhaoxi/imgs/test.txt
|
86 |
+
size: 256
|
87 |
+
coord: true
|
88 |
+
random_crop: true
|
89 |
+
holistic: 128
|
local_sampler_outdoor/checkpoints/last.ckpt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3172f82fe835531bdecf643bc0b538aae85165ff974b0342a972f3c3a81e920e
|
3 |
+
size 1812376146
|
local_sampler_outdoor/configs/2022-04-29T20-25-45-lightning.yaml
ADDED
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
1 |
+
lightning:
|
2 |
+
trainer:
|
3 |
+
distributed_backend: ddp
|
4 |
+
gpus: 0,1,2,3,4,5,6,7
|
local_sampler_outdoor/configs/2022-04-29T20-25-45-project.yaml
ADDED
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
model:
|
2 |
+
base_learning_rate: 4.5e-06
|
3 |
+
target: taming.models.local_sampler.LocalSamplerDualCodebookPE
|
4 |
+
params:
|
5 |
+
cond_stage_key: coord
|
6 |
+
holistic_config:
|
7 |
+
target: taming.models.vqgan.VQCirModel
|
8 |
+
params:
|
9 |
+
ckpt_path:
|
10 |
+
embed_dim: 256
|
11 |
+
n_embed: 1024
|
12 |
+
ddconfig:
|
13 |
+
double_z: false
|
14 |
+
z_channels: 256
|
15 |
+
resolution: 256
|
16 |
+
in_channels: 3
|
17 |
+
out_ch: 3
|
18 |
+
ch: 128
|
19 |
+
ch_mult:
|
20 |
+
- 1
|
21 |
+
- 1
|
22 |
+
- 2
|
23 |
+
- 2
|
24 |
+
- 4
|
25 |
+
num_res_blocks: 2
|
26 |
+
attn_resolutions:
|
27 |
+
- 16
|
28 |
+
dropout: 0.0
|
29 |
+
lossconfig:
|
30 |
+
target: taming.modules.losses.DummyLoss
|
31 |
+
transformer_config:
|
32 |
+
target: taming.modules.transformer.mingpt.GPT
|
33 |
+
params:
|
34 |
+
vocab_size: 1024
|
35 |
+
block_size: 402
|
36 |
+
n_layer: 24
|
37 |
+
n_head: 16
|
38 |
+
n_embd: 1024
|
39 |
+
first_stage_config:
|
40 |
+
target: taming.models.vqgan.VQModel
|
41 |
+
params:
|
42 |
+
ckpt_path:
|
43 |
+
embed_dim: 256
|
44 |
+
n_embed: 1024
|
45 |
+
ddconfig:
|
46 |
+
double_z: false
|
47 |
+
z_channels: 256
|
48 |
+
resolution: 256
|
49 |
+
in_channels: 3
|
50 |
+
out_ch: 3
|
51 |
+
ch: 128
|
52 |
+
ch_mult:
|
53 |
+
- 1
|
54 |
+
- 1
|
55 |
+
- 2
|
56 |
+
- 2
|
57 |
+
- 4
|
58 |
+
num_res_blocks: 2
|
59 |
+
attn_resolutions:
|
60 |
+
- 16
|
61 |
+
dropout: 0.0
|
62 |
+
lossconfig:
|
63 |
+
target: taming.modules.losses.DummyLoss
|
64 |
+
cond_stage_config:
|
65 |
+
target: taming.modules.misc.coord.CoordStage
|
66 |
+
params:
|
67 |
+
n_embed: 1024
|
68 |
+
down_factor: 16
|
69 |
+
data:
|
70 |
+
target: main.DataModuleFromConfig
|
71 |
+
params:
|
72 |
+
batch_size: 24
|
73 |
+
num_workers: 16
|
74 |
+
train:
|
75 |
+
target: taming.data.custom.CustomTrain
|
76 |
+
params:
|
77 |
+
training_images_list_file: /home/chenzhaoxi/imgs/train_outdoor.txt
|
78 |
+
size: 256
|
79 |
+
coord: true
|
80 |
+
random_crop: true
|
81 |
+
holistic: 128
|
82 |
+
validation:
|
83 |
+
target: taming.data.custom.CustomTest
|
84 |
+
params:
|
85 |
+
test_images_list_file: /home/chenzhaoxi/imgs/test.txt
|
86 |
+
size: 256
|
87 |
+
coord: true
|
88 |
+
random_crop: true
|
89 |
+
holistic: 128
|
sritmo.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ab14c0e0a3f810eddcc04f51457fa2e9ca5443670b1620f7b743234c70b2858c
|
3 |
+
size 20491375
|