Hugging Face's logo

sylvia-boxue
/
models_v2

Text-to-Image
Diffusers
TensorBoard
Safetensors
stable-diffusion
stable-diffusion-diffusers
controlnet
diffusers-training
Model card Files Metrics Community
sylvia-boxue commited on
Commit
fd5edf8
•
1 Parent(s): 2f662b0

End of training

Browse files
Files changed (43) hide show
  1. .gitattributes +2 -0
  2. README.md +45 -0
  3. checkpoint-1000/controlnet/config.json +51 -0
  4. checkpoint-1000/controlnet/diffusion_pytorch_model.safetensors +3 -0
  5. checkpoint-1000/optimizer.bin +3 -0
  6. checkpoint-1000/random_states_0.pkl +3 -0
  7. checkpoint-1000/scheduler.bin +3 -0
  8. checkpoint-1500/controlnet/config.json +51 -0
  9. checkpoint-1500/controlnet/diffusion_pytorch_model.safetensors +3 -0
  10. checkpoint-1500/optimizer.bin +3 -0
  11. checkpoint-1500/random_states_0.pkl +3 -0
  12. checkpoint-1500/scheduler.bin +3 -0
  13. checkpoint-2000/controlnet/config.json +51 -0
  14. checkpoint-2000/controlnet/diffusion_pytorch_model.safetensors +3 -0
  15. checkpoint-2000/optimizer.bin +3 -0
  16. checkpoint-2000/random_states_0.pkl +3 -0
  17. checkpoint-2000/scheduler.bin +3 -0
  18. checkpoint-2500/controlnet/config.json +51 -0
  19. checkpoint-2500/controlnet/diffusion_pytorch_model.safetensors +3 -0
  20. checkpoint-2500/optimizer.bin +3 -0
  21. checkpoint-2500/random_states_0.pkl +3 -0
  22. checkpoint-2500/scheduler.bin +3 -0
  23. checkpoint-3000/controlnet/config.json +51 -0
  24. checkpoint-3000/controlnet/diffusion_pytorch_model.safetensors +3 -0
  25. checkpoint-3000/optimizer.bin +3 -0
  26. checkpoint-3000/random_states_0.pkl +3 -0
  27. checkpoint-3000/scheduler.bin +3 -0
  28. checkpoint-500/controlnet/config.json +51 -0
  29. checkpoint-500/controlnet/diffusion_pytorch_model.safetensors +3 -0
  30. checkpoint-500/optimizer.bin +3 -0
  31. checkpoint-500/random_states_0.pkl +3 -0
  32. checkpoint-500/scheduler.bin +3 -0
  33. config.json +51 -0
  34. diffusion_pytorch_model.safetensors +3 -0
  35. image_control.png +0 -0
  36. images_0.png +3 -0
  37. images_1.png +3 -0
  38. logs/train_controlnet/1724147255.4815252/events.out.tfevents.1724147255.aiot-4090-hk01.3621311.1 +3 -0
  39. logs/train_controlnet/1724147255.4824226/hparams.yml +50 -0
  40. logs/train_controlnet/1724147283.4051206/events.out.tfevents.1724147283.aiot-4090-hk01.3622357.1 +3 -0
  41. logs/train_controlnet/1724147283.4062078/hparams.yml +50 -0
  42. logs/train_controlnet/events.out.tfevents.1724147255.aiot-4090-hk01.3621311.0 +3 -0
  43. logs/train_controlnet/events.out.tfevents.1724147283.aiot-4090-hk01.3622357.0 +3 -0
.gitattributes CHANGED
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ images_0.png filter=lfs diff=lfs merge=lfs -text
37
+ images_1.png filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: runwayml/stable-diffusion-v1-5
3
+ library_name: diffusers
4
+ license: creativeml-openrail-m
5
+ tags:
6
+ - stable-diffusion
7
+ - stable-diffusion-diffusers
8
+ - text-to-image
9
+ - diffusers
10
+ - controlnet
11
+ - diffusers-training
12
+ inference: true
13
+ ---
14
+
15
+ <!-- This model card has been generated automatically according to the information the training script had access to. You
16
+ should probably proofread and complete it, then remove this comment. -->
17
+
18
+
19
+ # controlnet-sylvia-boxue/models_v2
20
+
21
+ These are controlnet weights trained on runwayml/stable-diffusion-v1-5 with new type of conditioning.
22
+ You can find some example images below.
23
+
24
+ prompt: A thermal image of an indoor scene with human body as the primary heat source
25
+ ![images_0)](./images_0.png)
26
+ prompt: A thermal image of an indoor scene with human body as the primary heat source
27
+ ![images_1)](./images_1.png)
28
+
29
+
30
+
31
+ ## Intended uses & limitations
32
+
33
+ #### How to use
34
+
35
+ ```python
36
+ # TODO: add an example code snippet for running this diffusion pipeline
37
+ ```
38
+
39
+ #### Limitations and bias
40
+
41
+ [TODO: provide examples of latent issues and potential remediations]
42
+
43
+ ## Training details
44
+
45
+ [TODO: describe the data used to train the model]
checkpoint-1000/controlnet/config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "ControlNetModel",
3
+ "_diffusers_version": "0.31.0.dev0",
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": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "class_embed_type": null,
16
+ "conditioning_channels": 3,
17
+ "conditioning_embedding_out_channels": [
18
+ 16,
19
+ 32,
20
+ 96,
21
+ 256
22
+ ],
23
+ "controlnet_conditioning_channel_order": "rgb",
24
+ "cross_attention_dim": 768,
25
+ "down_block_types": [
26
+ "CrossAttnDownBlock2D",
27
+ "CrossAttnDownBlock2D",
28
+ "CrossAttnDownBlock2D",
29
+ "DownBlock2D"
30
+ ],
31
+ "downsample_padding": 1,
32
+ "encoder_hid_dim": null,
33
+ "encoder_hid_dim_type": null,
34
+ "flip_sin_to_cos": true,
35
+ "freq_shift": 0,
36
+ "global_pool_conditions": false,
37
+ "in_channels": 4,
38
+ "layers_per_block": 2,
39
+ "mid_block_scale_factor": 1,
40
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
41
+ "norm_eps": 1e-05,
42
+ "norm_num_groups": 32,
43
+ "num_attention_heads": null,
44
+ "num_class_embeds": null,
45
+ "only_cross_attention": false,
46
+ "projection_class_embeddings_input_dim": null,
47
+ "resnet_time_scale_shift": "default",
48
+ "transformer_layers_per_block": 1,
49
+ "upcast_attention": false,
50
+ "use_linear_projection": false
51
+ }
checkpoint-1000/controlnet/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4cccedc9de6fc04379ce8f0b74556a9dc89b0ba5fbc05ff9a56d4b8c2acbe6e
3
+ size 1445157120
checkpoint-1000/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b80b37157ddeb855069f0664c8de65a86b2d63717df23268e0bd2a0cc169d38f
3
+ size 2890515549
checkpoint-1000/random_states_0.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe0e8a5e0811c86d398ff7f444746f36f4c6b46833345c4f33daa7d7926b7da1
3
+ size 14567
checkpoint-1000/scheduler.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e8c4d64372d3cf1a4c56e9eb4a7d849f8d2bd7ed35075b211556d09c3222db9
3
+ size 559
checkpoint-1500/controlnet/config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "ControlNetModel",
3
+ "_diffusers_version": "0.31.0.dev0",
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": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "class_embed_type": null,
16
+ "conditioning_channels": 3,
17
+ "conditioning_embedding_out_channels": [
18
+ 16,
19
+ 32,
20
+ 96,
21
+ 256
22
+ ],
23
+ "controlnet_conditioning_channel_order": "rgb",
24
+ "cross_attention_dim": 768,
25
+ "down_block_types": [
26
+ "CrossAttnDownBlock2D",
27
+ "CrossAttnDownBlock2D",
28
+ "CrossAttnDownBlock2D",
29
+ "DownBlock2D"
30
+ ],
31
+ "downsample_padding": 1,
32
+ "encoder_hid_dim": null,
33
+ "encoder_hid_dim_type": null,
34
+ "flip_sin_to_cos": true,
35
+ "freq_shift": 0,
36
+ "global_pool_conditions": false,
37
+ "in_channels": 4,
38
+ "layers_per_block": 2,
39
+ "mid_block_scale_factor": 1,
40
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
41
+ "norm_eps": 1e-05,
42
+ "norm_num_groups": 32,
43
+ "num_attention_heads": null,
44
+ "num_class_embeds": null,
45
+ "only_cross_attention": false,
46
+ "projection_class_embeddings_input_dim": null,
47
+ "resnet_time_scale_shift": "default",
48
+ "transformer_layers_per_block": 1,
49
+ "upcast_attention": false,
50
+ "use_linear_projection": false
51
+ }
checkpoint-1500/controlnet/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7fe619653baab09134d16d0198609e431820d6b22467b5bfc6be78f1c436e357
3
+ size 1445157120
checkpoint-1500/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b855e533578629a72750bb40c2189e4b726b335c9d99487c1ac07ae0b8c0278
3
+ size 2890515549
checkpoint-1500/random_states_0.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b47c6ff9d80d01744ebf27f381235e3e1ddc78354ebc0e1925b358996064d41
3
+ size 14503
checkpoint-1500/scheduler.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af3055db812c559ccd800d4df8e3fd39b6eedb05f1409578279f9a113adb5cde
3
+ size 559
checkpoint-2000/controlnet/config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "ControlNetModel",
3
+ "_diffusers_version": "0.31.0.dev0",
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": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "class_embed_type": null,
16
+ "conditioning_channels": 3,
17
+ "conditioning_embedding_out_channels": [
18
+ 16,
19
+ 32,
20
+ 96,
21
+ 256
22
+ ],
23
+ "controlnet_conditioning_channel_order": "rgb",
24
+ "cross_attention_dim": 768,
25
+ "down_block_types": [
26
+ "CrossAttnDownBlock2D",
27
+ "CrossAttnDownBlock2D",
28
+ "CrossAttnDownBlock2D",
29
+ "DownBlock2D"
30
+ ],
31
+ "downsample_padding": 1,
32
+ "encoder_hid_dim": null,
33
+ "encoder_hid_dim_type": null,
34
+ "flip_sin_to_cos": true,
35
+ "freq_shift": 0,
36
+ "global_pool_conditions": false,
37
+ "in_channels": 4,
38
+ "layers_per_block": 2,
39
+ "mid_block_scale_factor": 1,
40
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
41
+ "norm_eps": 1e-05,
42
+ "norm_num_groups": 32,
43
+ "num_attention_heads": null,
44
+ "num_class_embeds": null,
45
+ "only_cross_attention": false,
46
+ "projection_class_embeddings_input_dim": null,
47
+ "resnet_time_scale_shift": "default",
48
+ "transformer_layers_per_block": 1,
49
+ "upcast_attention": false,
50
+ "use_linear_projection": false
51
+ }
checkpoint-2000/controlnet/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e529631546af9c7e2aee68a9e4c35a1444d9aa4f704bd2bf538c0c56d175ba4
3
+ size 1445157120
checkpoint-2000/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:11b4aef4ec6466b9992ddd39fb9ce96526a26fec14b646e8719bf15cebc5dfaf
3
+ size 2890515549
checkpoint-2000/random_states_0.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e8e2be8a3a16c05ceff5999a1a416bbe352fd84b8d7fd0d5bb0c4f7be474926
3
+ size 14567
checkpoint-2000/scheduler.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6c995ccb04e2615676160d574336f1fedb3fbd4741e7650e6f2b0273cccaab2
3
+ size 559
checkpoint-2500/controlnet/config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "ControlNetModel",
3
+ "_diffusers_version": "0.31.0.dev0",
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": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "class_embed_type": null,
16
+ "conditioning_channels": 3,
17
+ "conditioning_embedding_out_channels": [
18
+ 16,
19
+ 32,
20
+ 96,
21
+ 256
22
+ ],
23
+ "controlnet_conditioning_channel_order": "rgb",
24
+ "cross_attention_dim": 768,
25
+ "down_block_types": [
26
+ "CrossAttnDownBlock2D",
27
+ "CrossAttnDownBlock2D",
28
+ "CrossAttnDownBlock2D",
29
+ "DownBlock2D"
30
+ ],
31
+ "downsample_padding": 1,
32
+ "encoder_hid_dim": null,
33
+ "encoder_hid_dim_type": null,
34
+ "flip_sin_to_cos": true,
35
+ "freq_shift": 0,
36
+ "global_pool_conditions": false,
37
+ "in_channels": 4,
38
+ "layers_per_block": 2,
39
+ "mid_block_scale_factor": 1,
40
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
41
+ "norm_eps": 1e-05,
42
+ "norm_num_groups": 32,
43
+ "num_attention_heads": null,
44
+ "num_class_embeds": null,
45
+ "only_cross_attention": false,
46
+ "projection_class_embeddings_input_dim": null,
47
+ "resnet_time_scale_shift": "default",
48
+ "transformer_layers_per_block": 1,
49
+ "upcast_attention": false,
50
+ "use_linear_projection": false
51
+ }
checkpoint-2500/controlnet/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:96f9fedd3645b890e33e7f21ab682651c8e8a55185f9fbdd014543d5c5863cba
3
+ size 1445157120
checkpoint-2500/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:abb845d406f666c015cf3bab3664dbd14e87c549c7a5d295a1bcb4be3d0a4f14
3
+ size 2890515549
checkpoint-2500/random_states_0.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed2f18e3e05fb0c28447a0ef4523d753a8163cd79287784b428c065115662ce6
3
+ size 14567
checkpoint-2500/scheduler.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4857238eeb379b56d934df534f5ed6ca85661c7c76ec0a363a2b9aab297ee286
3
+ size 559
checkpoint-3000/controlnet/config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "ControlNetModel",
3
+ "_diffusers_version": "0.31.0.dev0",
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": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "class_embed_type": null,
16
+ "conditioning_channels": 3,
17
+ "conditioning_embedding_out_channels": [
18
+ 16,
19
+ 32,
20
+ 96,
21
+ 256
22
+ ],
23
+ "controlnet_conditioning_channel_order": "rgb",
24
+ "cross_attention_dim": 768,
25
+ "down_block_types": [
26
+ "CrossAttnDownBlock2D",
27
+ "CrossAttnDownBlock2D",
28
+ "CrossAttnDownBlock2D",
29
+ "DownBlock2D"
30
+ ],
31
+ "downsample_padding": 1,
32
+ "encoder_hid_dim": null,
33
+ "encoder_hid_dim_type": null,
34
+ "flip_sin_to_cos": true,
35
+ "freq_shift": 0,
36
+ "global_pool_conditions": false,
37
+ "in_channels": 4,
38
+ "layers_per_block": 2,
39
+ "mid_block_scale_factor": 1,
40
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
41
+ "norm_eps": 1e-05,
42
+ "norm_num_groups": 32,
43
+ "num_attention_heads": null,
44
+ "num_class_embeds": null,
45
+ "only_cross_attention": false,
46
+ "projection_class_embeddings_input_dim": null,
47
+ "resnet_time_scale_shift": "default",
48
+ "transformer_layers_per_block": 1,
49
+ "upcast_attention": false,
50
+ "use_linear_projection": false
51
+ }
checkpoint-3000/controlnet/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f1fa69fa7064e88376b146ea3f18698be62214b2728ec105905bc6114646f9c0
3
+ size 1445157120
checkpoint-3000/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83856cc84d20b14160a5b3066feea6eec93e3a5712838cac722a5f6abc154ec8
3
+ size 2890515549
checkpoint-3000/random_states_0.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aea1ac133b01648097ae0c79724c36429d629211c4ef5e5622ee0f9b00e9adba
3
+ size 14567
checkpoint-3000/scheduler.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2a7ce4d519c487f1bbc2c86cc7af032e1439d4975686c1113bebcc2c1bf2c07
3
+ size 559
checkpoint-500/controlnet/config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "ControlNetModel",
3
+ "_diffusers_version": "0.31.0.dev0",
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": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "class_embed_type": null,
16
+ "conditioning_channels": 3,
17
+ "conditioning_embedding_out_channels": [
18
+ 16,
19
+ 32,
20
+ 96,
21
+ 256
22
+ ],
23
+ "controlnet_conditioning_channel_order": "rgb",
24
+ "cross_attention_dim": 768,
25
+ "down_block_types": [
26
+ "CrossAttnDownBlock2D",
27
+ "CrossAttnDownBlock2D",
28
+ "CrossAttnDownBlock2D",
29
+ "DownBlock2D"
30
+ ],
31
+ "downsample_padding": 1,
32
+ "encoder_hid_dim": null,
33
+ "encoder_hid_dim_type": null,
34
+ "flip_sin_to_cos": true,
35
+ "freq_shift": 0,
36
+ "global_pool_conditions": false,
37
+ "in_channels": 4,
38
+ "layers_per_block": 2,
39
+ "mid_block_scale_factor": 1,
40
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
41
+ "norm_eps": 1e-05,
42
+ "norm_num_groups": 32,
43
+ "num_attention_heads": null,
44
+ "num_class_embeds": null,
45
+ "only_cross_attention": false,
46
+ "projection_class_embeddings_input_dim": null,
47
+ "resnet_time_scale_shift": "default",
48
+ "transformer_layers_per_block": 1,
49
+ "upcast_attention": false,
50
+ "use_linear_projection": false
51
+ }
checkpoint-500/controlnet/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4fa44f45fca8c78784172e4817469fb63883a926f1fc175ccd721469665c3687
3
+ size 1445157120
checkpoint-500/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0ff9171e51e59d7003b993636f118cdf2272379eb35b971ebb88878dfbbd1af
3
+ size 2890515549
checkpoint-500/random_states_0.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6214dd30e62668be93d8b7c6722ebcef83ca66f7d69cc39a2ae00b59931283a
3
+ size 14567
checkpoint-500/scheduler.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8b1890ce121d340d95c51c08c1e3cb35cf570495feafe7bebe027e63d66af7ac
3
+ size 559
config.json ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "ControlNetModel",
3
+ "_diffusers_version": "0.31.0.dev0",
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": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "class_embed_type": null,
16
+ "conditioning_channels": 3,
17
+ "conditioning_embedding_out_channels": [
18
+ 16,
19
+ 32,
20
+ 96,
21
+ 256
22
+ ],
23
+ "controlnet_conditioning_channel_order": "rgb",
24
+ "cross_attention_dim": 768,
25
+ "down_block_types": [
26
+ "CrossAttnDownBlock2D",
27
+ "CrossAttnDownBlock2D",
28
+ "CrossAttnDownBlock2D",
29
+ "DownBlock2D"
30
+ ],
31
+ "downsample_padding": 1,
32
+ "encoder_hid_dim": null,
33
+ "encoder_hid_dim_type": null,
34
+ "flip_sin_to_cos": true,
35
+ "freq_shift": 0,
36
+ "global_pool_conditions": false,
37
+ "in_channels": 4,
38
+ "layers_per_block": 2,
39
+ "mid_block_scale_factor": 1,
40
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
41
+ "norm_eps": 1e-05,
42
+ "norm_num_groups": 32,
43
+ "num_attention_heads": null,
44
+ "num_class_embeds": null,
45
+ "only_cross_attention": false,
46
+ "projection_class_embeddings_input_dim": null,
47
+ "resnet_time_scale_shift": "default",
48
+ "transformer_layers_per_block": 1,
49
+ "upcast_attention": false,
50
+ "use_linear_projection": false
51
+ }
diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fa304720a51b9dc764ed08a4a5ca2e9f07b75301677eb80f43a21634371997df
3
+ size 1445157120
image_control.png ADDED
images_0.png ADDED

Git LFS Details

  • SHA256: f990db83d90d362e45ba1fe357a2821c193384482509d187a7dbd82ea745789b
  • Pointer size: 132 Bytes
  • Size of remote file: 1.44 MB
images_1.png ADDED

Git LFS Details

  • SHA256: 2f848d7cd19b5a57a8b48d759a49f50d31aae0c4ded531c72cdc58920d1e1f1b
  • Pointer size: 132 Bytes
  • Size of remote file: 1.54 MB
logs/train_controlnet/1724147255.4815252/events.out.tfevents.1724147255.aiot-4090-hk01.3621311.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2c1200b7f83c40567a67d9ea9723be5f98ee54c8a8ec22d1326da217dd2f5f08
3
+ size 2322
logs/train_controlnet/1724147255.4824226/hparams.yml ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ adam_beta1: 0.9
2
+ adam_beta2: 0.999
3
+ adam_epsilon: 1.0e-08
4
+ adam_weight_decay: 0.01
5
+ allow_tf32: false
6
+ cache_dir: null
7
+ caption_column: prompt
8
+ checkpointing_steps: 500
9
+ checkpoints_total_limit: null
10
+ conditioning_image_column: source1
11
+ controlnet_model_name_or_path: null
12
+ dataloader_num_workers: 0
13
+ dataset_config_name: null
14
+ dataset_name: null
15
+ enable_xformers_memory_efficient_attention: false
16
+ gradient_accumulation_steps: 4
17
+ gradient_checkpointing: false
18
+ hub_model_id: null
19
+ hub_token: null
20
+ image_column: target
21
+ learning_rate: 1.0e-05
22
+ logging_dir: logs
23
+ lr_num_cycles: 1
24
+ lr_power: 1.0
25
+ lr_scheduler: constant
26
+ lr_warmup_steps: 500
27
+ max_grad_norm: 1.0
28
+ max_train_samples: null
29
+ max_train_steps: 3358
30
+ mixed_precision: null
31
+ num_train_epochs: 2
32
+ num_validation_images: 4
33
+ output_dir: models_v2
34
+ pretrained_model_name_or_path: runwayml/stable-diffusion-v1-5
35
+ proportion_empty_prompts: 0
36
+ push_to_hub: true
37
+ report_to: tensorboard
38
+ resolution: 512
39
+ resume_from_checkpoint: null
40
+ revision: null
41
+ scale_lr: false
42
+ seed: 1234
43
+ set_grads_to_none: false
44
+ tokenizer_name: null
45
+ tracker_project_name: train_controlnet
46
+ train_batch_size: 1
47
+ train_data_dir: /home/zhiyuan/boxue/dataset/train/data
48
+ use_8bit_adam: false
49
+ validation_steps: 100
50
+ variant: null
logs/train_controlnet/1724147283.4051206/events.out.tfevents.1724147283.aiot-4090-hk01.3622357.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc4e7aea0a27560d2f7f37724c51dcd7bfd12cc66e015fcf1f0853cce4553241
3
+ size 2322
logs/train_controlnet/1724147283.4062078/hparams.yml ADDED
@@ -0,0 +1,50 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ adam_beta1: 0.9
2
+ adam_beta2: 0.999
3
+ adam_epsilon: 1.0e-08
4
+ adam_weight_decay: 0.01
5
+ allow_tf32: false
6
+ cache_dir: null
7
+ caption_column: prompt
8
+ checkpointing_steps: 500
9
+ checkpoints_total_limit: null
10
+ conditioning_image_column: source1
11
+ controlnet_model_name_or_path: null
12
+ dataloader_num_workers: 0
13
+ dataset_config_name: null
14
+ dataset_name: null
15
+ enable_xformers_memory_efficient_attention: false
16
+ gradient_accumulation_steps: 4
17
+ gradient_checkpointing: false
18
+ hub_model_id: null
19
+ hub_token: null
20
+ image_column: target
21
+ learning_rate: 1.0e-05
22
+ logging_dir: logs
23
+ lr_num_cycles: 1
24
+ lr_power: 1.0
25
+ lr_scheduler: constant
26
+ lr_warmup_steps: 500
27
+ max_grad_norm: 1.0
28
+ max_train_samples: null
29
+ max_train_steps: 3358
30
+ mixed_precision: null
31
+ num_train_epochs: 2
32
+ num_validation_images: 4
33
+ output_dir: models_v2
34
+ pretrained_model_name_or_path: runwayml/stable-diffusion-v1-5
35
+ proportion_empty_prompts: 0
36
+ push_to_hub: true
37
+ report_to: tensorboard
38
+ resolution: 512
39
+ resume_from_checkpoint: null
40
+ revision: null
41
+ scale_lr: false
42
+ seed: 1234
43
+ set_grads_to_none: false
44
+ tokenizer_name: null
45
+ tracker_project_name: train_controlnet
46
+ train_batch_size: 1
47
+ train_data_dir: /home/zhiyuan/boxue/dataset/train/data
48
+ use_8bit_adam: false
49
+ validation_steps: 100
50
+ variant: null
logs/train_controlnet/events.out.tfevents.1724147255.aiot-4090-hk01.3621311.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42a2ab7e6a97c4d97e7eb5117b5e5f6023217aac8b4d222f07e6989ef7e7991f
3
+ size 1962
logs/train_controlnet/events.out.tfevents.1724147283.aiot-4090-hk01.3622357.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:38d5c4e0ba354fadee2f6f57dd5233a5588629937406daf589f0e58a7765f3d9
3
+ size 108858088