Hugging Face's logo

DaveLoay
/
Riffusion_FineTuning_Tutorial

Text-to-Image
Diffusers
TensorBoard
English
StableDiffusionPipeline
Model card Files Metrics Community
DaveLoay commited on
Commit
20e3d1c
·
1 Parent(s): 4556d04

End of training

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. checkpoint-1000/optimizer.bin +3 -0
  2. checkpoint-1000/random_states_0.pkl +3 -0
  3. checkpoint-1000/scaler.pt +3 -0
  4. checkpoint-1000/scheduler.bin +3 -0
  5. checkpoint-1000/unet/config.json +62 -0
  6. checkpoint-1000/unet/diffusion_pytorch_model.bin +3 -0
  7. checkpoint-1000/unet_ema/config.json +69 -0
  8. checkpoint-1000/unet_ema/diffusion_pytorch_model.bin +3 -0
  9. checkpoint-1500/optimizer.bin +3 -0
  10. checkpoint-1500/random_states_0.pkl +3 -0
  11. checkpoint-1500/scaler.pt +3 -0
  12. checkpoint-1500/scheduler.bin +3 -0
  13. checkpoint-1500/unet/config.json +62 -0
  14. checkpoint-1500/unet/diffusion_pytorch_model.bin +3 -0
  15. checkpoint-1500/unet_ema/config.json +69 -0
  16. checkpoint-1500/unet_ema/diffusion_pytorch_model.bin +3 -0
  17. checkpoint-2000/optimizer.bin +3 -0
  18. checkpoint-2000/random_states_0.pkl +3 -0
  19. checkpoint-2000/scaler.pt +3 -0
  20. checkpoint-2000/scheduler.bin +3 -0
  21. checkpoint-2000/unet/config.json +62 -0
  22. checkpoint-2000/unet/diffusion_pytorch_model.bin +3 -0
  23. checkpoint-2000/unet_ema/config.json +69 -0
  24. checkpoint-2000/unet_ema/diffusion_pytorch_model.bin +3 -0
  25. checkpoint-2500/optimizer.bin +3 -0
  26. checkpoint-2500/random_states_0.pkl +3 -0
  27. checkpoint-2500/scaler.pt +3 -0
  28. checkpoint-2500/scheduler.bin +3 -0
  29. checkpoint-2500/unet/config.json +62 -0
  30. checkpoint-2500/unet/diffusion_pytorch_model.bin +3 -0
  31. checkpoint-2500/unet_ema/config.json +69 -0
  32. checkpoint-2500/unet_ema/diffusion_pytorch_model.bin +3 -0
  33. checkpoint-3000/optimizer.bin +3 -0
  34. checkpoint-3000/random_states_0.pkl +3 -0
  35. checkpoint-3000/scaler.pt +3 -0
  36. checkpoint-3000/scheduler.bin +3 -0
  37. checkpoint-3000/unet/config.json +62 -0
  38. checkpoint-3000/unet/diffusion_pytorch_model.bin +3 -0
  39. checkpoint-3000/unet_ema/config.json +69 -0
  40. checkpoint-3000/unet_ema/diffusion_pytorch_model.bin +3 -0
  41. checkpoint-3500/optimizer.bin +3 -0
  42. checkpoint-3500/random_states_0.pkl +3 -0
  43. checkpoint-3500/scaler.pt +3 -0
  44. checkpoint-3500/scheduler.bin +3 -0
  45. checkpoint-3500/unet/config.json +62 -0
  46. checkpoint-3500/unet/diffusion_pytorch_model.bin +3 -0
  47. checkpoint-3500/unet_ema/config.json +69 -0
  48. checkpoint-3500/unet_ema/diffusion_pytorch_model.bin +3 -0
  49. checkpoint-4000/optimizer.bin +3 -0
  50. checkpoint-4000/random_states_0.pkl +3 -0
checkpoint-1000/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:222cce1d28224dfaa8b7b4b7274d8266a40129af3ab1b71ac269e369a3ff152a
3
+ size 6876749715
checkpoint-1000/random_states_0.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:468a72147c4149031396a60a73fa0be37c82b3f983760d182010a9dd005a2df0
3
+ size 14663
checkpoint-1000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:68cff80b680ddf6e7abbef98b5f336b97f9b5963e2209307f639383870e8cc71
3
+ size 557
checkpoint-1000/scheduler.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f34912fa092355c5a950e4967ea28cbb3556238a7857baa5999d228cde0ae4c3
3
+ size 563
checkpoint-1000/unet/config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "UNet2DConditionModel",
3
+ "_diffusers_version": "0.17.0.dev0",
4
+ "_name_or_path": "riffusion/riffusion-model-v1",
5
+ "act_fn": "silu",
6
+ "addition_embed_type": null,
7
+ "addition_embed_type_num_heads": 64,
8
+ "attention_head_dim": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "center_input_sample": false,
16
+ "class_embed_type": null,
17
+ "class_embeddings_concat": false,
18
+ "conv_in_kernel": 3,
19
+ "conv_out_kernel": 3,
20
+ "cross_attention_dim": 768,
21
+ "cross_attention_norm": null,
22
+ "down_block_types": [
23
+ "CrossAttnDownBlock2D",
24
+ "CrossAttnDownBlock2D",
25
+ "CrossAttnDownBlock2D",
26
+ "DownBlock2D"
27
+ ],
28
+ "downsample_padding": 1,
29
+ "dual_cross_attention": false,
30
+ "encoder_hid_dim": null,
31
+ "encoder_hid_dim_type": null,
32
+ "flip_sin_to_cos": true,
33
+ "freq_shift": 0,
34
+ "in_channels": 4,
35
+ "layers_per_block": 2,
36
+ "mid_block_only_cross_attention": null,
37
+ "mid_block_scale_factor": 1,
38
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
39
+ "norm_eps": 1e-05,
40
+ "norm_num_groups": 32,
41
+ "num_class_embeds": null,
42
+ "only_cross_attention": false,
43
+ "out_channels": 4,
44
+ "projection_class_embeddings_input_dim": null,
45
+ "resnet_out_scale_factor": 1.0,
46
+ "resnet_skip_time_act": false,
47
+ "resnet_time_scale_shift": "default",
48
+ "sample_size": 32,
49
+ "time_cond_proj_dim": null,
50
+ "time_embedding_act_fn": null,
51
+ "time_embedding_dim": null,
52
+ "time_embedding_type": "positional",
53
+ "timestep_post_act": null,
54
+ "up_block_types": [
55
+ "UpBlock2D",
56
+ "CrossAttnUpBlock2D",
57
+ "CrossAttnUpBlock2D",
58
+ "CrossAttnUpBlock2D"
59
+ ],
60
+ "upcast_attention": false,
61
+ "use_linear_projection": false
62
+ }
checkpoint-1000/unet/diffusion_pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6063e9ebee3d65387c624572ecff2dd926cb9a41e61da321a42a88e5f112ab41
3
+ size 3438375973
checkpoint-1000/unet_ema/config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "UNet2DConditionModel",
3
+ "_diffusers_version": "0.17.0.dev0",
4
+ "_name_or_path": "riffusion/riffusion-model-v1",
5
+ "act_fn": "silu",
6
+ "addition_embed_type": null,
7
+ "addition_embed_type_num_heads": 64,
8
+ "attention_head_dim": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "center_input_sample": false,
16
+ "class_embed_type": null,
17
+ "class_embeddings_concat": false,
18
+ "conv_in_kernel": 3,
19
+ "conv_out_kernel": 3,
20
+ "cross_attention_dim": 768,
21
+ "cross_attention_norm": null,
22
+ "decay": 0.9999,
23
+ "down_block_types": [
24
+ "CrossAttnDownBlock2D",
25
+ "CrossAttnDownBlock2D",
26
+ "CrossAttnDownBlock2D",
27
+ "DownBlock2D"
28
+ ],
29
+ "downsample_padding": 1,
30
+ "dual_cross_attention": false,
31
+ "encoder_hid_dim": null,
32
+ "encoder_hid_dim_type": null,
33
+ "flip_sin_to_cos": true,
34
+ "freq_shift": 0,
35
+ "in_channels": 4,
36
+ "inv_gamma": 1.0,
37
+ "layers_per_block": 2,
38
+ "mid_block_only_cross_attention": null,
39
+ "mid_block_scale_factor": 1,
40
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
41
+ "min_decay": 0.0,
42
+ "norm_eps": 1e-05,
43
+ "norm_num_groups": 32,
44
+ "num_class_embeds": null,
45
+ "only_cross_attention": false,
46
+ "optimization_step": 1000,
47
+ "out_channels": 4,
48
+ "power": 0.6666666666666666,
49
+ "projection_class_embeddings_input_dim": null,
50
+ "resnet_out_scale_factor": 1.0,
51
+ "resnet_skip_time_act": false,
52
+ "resnet_time_scale_shift": "default",
53
+ "sample_size": 32,
54
+ "time_cond_proj_dim": null,
55
+ "time_embedding_act_fn": null,
56
+ "time_embedding_dim": null,
57
+ "time_embedding_type": "positional",
58
+ "timestep_post_act": null,
59
+ "up_block_types": [
60
+ "UpBlock2D",
61
+ "CrossAttnUpBlock2D",
62
+ "CrossAttnUpBlock2D",
63
+ "CrossAttnUpBlock2D"
64
+ ],
65
+ "upcast_attention": false,
66
+ "update_after_step": 0,
67
+ "use_ema_warmup": false,
68
+ "use_linear_projection": false
69
+ }
checkpoint-1000/unet_ema/diffusion_pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a03f06f91c8f85deab5a8605fb55e871e076fdd010f2edb4a88ea80f5af9e876
3
+ size 3438366373
checkpoint-1500/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2d51dc9a7b416892a263afbe8cec646cf6bf8236ae74830f90623b44bf5edfd
3
+ size 6876749715
checkpoint-1500/random_states_0.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d13b784797447a9360a73eaa258e73af30d89ead261776baa9db5a7979d8f08
3
+ size 14663
checkpoint-1500/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:203a72d6c29f42a0e2964fdddc8d7a98df1eccee78fea9de0fa416613390f5c6
3
+ size 557
checkpoint-1500/scheduler.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bab8983b1117c68e4e9a600eacf54bed4c3ac40f58cbf8864e53243107b904fb
3
+ size 563
checkpoint-1500/unet/config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "UNet2DConditionModel",
3
+ "_diffusers_version": "0.17.0.dev0",
4
+ "_name_or_path": "riffusion/riffusion-model-v1",
5
+ "act_fn": "silu",
6
+ "addition_embed_type": null,
7
+ "addition_embed_type_num_heads": 64,
8
+ "attention_head_dim": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "center_input_sample": false,
16
+ "class_embed_type": null,
17
+ "class_embeddings_concat": false,
18
+ "conv_in_kernel": 3,
19
+ "conv_out_kernel": 3,
20
+ "cross_attention_dim": 768,
21
+ "cross_attention_norm": null,
22
+ "down_block_types": [
23
+ "CrossAttnDownBlock2D",
24
+ "CrossAttnDownBlock2D",
25
+ "CrossAttnDownBlock2D",
26
+ "DownBlock2D"
27
+ ],
28
+ "downsample_padding": 1,
29
+ "dual_cross_attention": false,
30
+ "encoder_hid_dim": null,
31
+ "encoder_hid_dim_type": null,
32
+ "flip_sin_to_cos": true,
33
+ "freq_shift": 0,
34
+ "in_channels": 4,
35
+ "layers_per_block": 2,
36
+ "mid_block_only_cross_attention": null,
37
+ "mid_block_scale_factor": 1,
38
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
39
+ "norm_eps": 1e-05,
40
+ "norm_num_groups": 32,
41
+ "num_class_embeds": null,
42
+ "only_cross_attention": false,
43
+ "out_channels": 4,
44
+ "projection_class_embeddings_input_dim": null,
45
+ "resnet_out_scale_factor": 1.0,
46
+ "resnet_skip_time_act": false,
47
+ "resnet_time_scale_shift": "default",
48
+ "sample_size": 32,
49
+ "time_cond_proj_dim": null,
50
+ "time_embedding_act_fn": null,
51
+ "time_embedding_dim": null,
52
+ "time_embedding_type": "positional",
53
+ "timestep_post_act": null,
54
+ "up_block_types": [
55
+ "UpBlock2D",
56
+ "CrossAttnUpBlock2D",
57
+ "CrossAttnUpBlock2D",
58
+ "CrossAttnUpBlock2D"
59
+ ],
60
+ "upcast_attention": false,
61
+ "use_linear_projection": false
62
+ }
checkpoint-1500/unet/diffusion_pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72a96997d8574920ed0d589f6a2a09dfa1576366813ac7c71348d49ba7dc6c54
3
+ size 3438375973
checkpoint-1500/unet_ema/config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "UNet2DConditionModel",
3
+ "_diffusers_version": "0.17.0.dev0",
4
+ "_name_or_path": "riffusion/riffusion-model-v1",
5
+ "act_fn": "silu",
6
+ "addition_embed_type": null,
7
+ "addition_embed_type_num_heads": 64,
8
+ "attention_head_dim": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "center_input_sample": false,
16
+ "class_embed_type": null,
17
+ "class_embeddings_concat": false,
18
+ "conv_in_kernel": 3,
19
+ "conv_out_kernel": 3,
20
+ "cross_attention_dim": 768,
21
+ "cross_attention_norm": null,
22
+ "decay": 0.9999,
23
+ "down_block_types": [
24
+ "CrossAttnDownBlock2D",
25
+ "CrossAttnDownBlock2D",
26
+ "CrossAttnDownBlock2D",
27
+ "DownBlock2D"
28
+ ],
29
+ "downsample_padding": 1,
30
+ "dual_cross_attention": false,
31
+ "encoder_hid_dim": null,
32
+ "encoder_hid_dim_type": null,
33
+ "flip_sin_to_cos": true,
34
+ "freq_shift": 0,
35
+ "in_channels": 4,
36
+ "inv_gamma": 1.0,
37
+ "layers_per_block": 2,
38
+ "mid_block_only_cross_attention": null,
39
+ "mid_block_scale_factor": 1,
40
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
41
+ "min_decay": 0.0,
42
+ "norm_eps": 1e-05,
43
+ "norm_num_groups": 32,
44
+ "num_class_embeds": null,
45
+ "only_cross_attention": false,
46
+ "optimization_step": 1500,
47
+ "out_channels": 4,
48
+ "power": 0.6666666666666666,
49
+ "projection_class_embeddings_input_dim": null,
50
+ "resnet_out_scale_factor": 1.0,
51
+ "resnet_skip_time_act": false,
52
+ "resnet_time_scale_shift": "default",
53
+ "sample_size": 32,
54
+ "time_cond_proj_dim": null,
55
+ "time_embedding_act_fn": null,
56
+ "time_embedding_dim": null,
57
+ "time_embedding_type": "positional",
58
+ "timestep_post_act": null,
59
+ "up_block_types": [
60
+ "UpBlock2D",
61
+ "CrossAttnUpBlock2D",
62
+ "CrossAttnUpBlock2D",
63
+ "CrossAttnUpBlock2D"
64
+ ],
65
+ "upcast_attention": false,
66
+ "update_after_step": 0,
67
+ "use_ema_warmup": false,
68
+ "use_linear_projection": false
69
+ }
checkpoint-1500/unet_ema/diffusion_pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2639f4a28ab9f4400823f54d36cd8c3511c5f399edaeaa1275da9bb5ff99c3f
3
+ size 3438366373
checkpoint-2000/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45a311b2fdee40dd72f69c969bd5beaf32e27639c1136634192fc690ab26ec9d
3
+ size 6876749715
checkpoint-2000/random_states_0.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d2cb8edcc169934cf1fdc4cb8d88fdc39ab68a0019a7c167ed2213e5ef4526b
3
+ size 14663
checkpoint-2000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd2de9749828adacdf103bf6e9592702bb7067a2c1df27dd62ab38c1eb8c070f
3
+ size 557
checkpoint-2000/scheduler.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dfe94afd21b04fe331bf93a36e6de4869425c4da4a32303493f7f8b96dd6ca61
3
+ size 563
checkpoint-2000/unet/config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "UNet2DConditionModel",
3
+ "_diffusers_version": "0.17.0.dev0",
4
+ "_name_or_path": "riffusion/riffusion-model-v1",
5
+ "act_fn": "silu",
6
+ "addition_embed_type": null,
7
+ "addition_embed_type_num_heads": 64,
8
+ "attention_head_dim": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "center_input_sample": false,
16
+ "class_embed_type": null,
17
+ "class_embeddings_concat": false,
18
+ "conv_in_kernel": 3,
19
+ "conv_out_kernel": 3,
20
+ "cross_attention_dim": 768,
21
+ "cross_attention_norm": null,
22
+ "down_block_types": [
23
+ "CrossAttnDownBlock2D",
24
+ "CrossAttnDownBlock2D",
25
+ "CrossAttnDownBlock2D",
26
+ "DownBlock2D"
27
+ ],
28
+ "downsample_padding": 1,
29
+ "dual_cross_attention": false,
30
+ "encoder_hid_dim": null,
31
+ "encoder_hid_dim_type": null,
32
+ "flip_sin_to_cos": true,
33
+ "freq_shift": 0,
34
+ "in_channels": 4,
35
+ "layers_per_block": 2,
36
+ "mid_block_only_cross_attention": null,
37
+ "mid_block_scale_factor": 1,
38
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
39
+ "norm_eps": 1e-05,
40
+ "norm_num_groups": 32,
41
+ "num_class_embeds": null,
42
+ "only_cross_attention": false,
43
+ "out_channels": 4,
44
+ "projection_class_embeddings_input_dim": null,
45
+ "resnet_out_scale_factor": 1.0,
46
+ "resnet_skip_time_act": false,
47
+ "resnet_time_scale_shift": "default",
48
+ "sample_size": 32,
49
+ "time_cond_proj_dim": null,
50
+ "time_embedding_act_fn": null,
51
+ "time_embedding_dim": null,
52
+ "time_embedding_type": "positional",
53
+ "timestep_post_act": null,
54
+ "up_block_types": [
55
+ "UpBlock2D",
56
+ "CrossAttnUpBlock2D",
57
+ "CrossAttnUpBlock2D",
58
+ "CrossAttnUpBlock2D"
59
+ ],
60
+ "upcast_attention": false,
61
+ "use_linear_projection": false
62
+ }
checkpoint-2000/unet/diffusion_pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c92645889a507c3f524cac72ca3dc14756e7b972f0ae7c4732c7e8e91451accd
3
+ size 3438375973
checkpoint-2000/unet_ema/config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "UNet2DConditionModel",
3
+ "_diffusers_version": "0.17.0.dev0",
4
+ "_name_or_path": "riffusion/riffusion-model-v1",
5
+ "act_fn": "silu",
6
+ "addition_embed_type": null,
7
+ "addition_embed_type_num_heads": 64,
8
+ "attention_head_dim": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "center_input_sample": false,
16
+ "class_embed_type": null,
17
+ "class_embeddings_concat": false,
18
+ "conv_in_kernel": 3,
19
+ "conv_out_kernel": 3,
20
+ "cross_attention_dim": 768,
21
+ "cross_attention_norm": null,
22
+ "decay": 0.9999,
23
+ "down_block_types": [
24
+ "CrossAttnDownBlock2D",
25
+ "CrossAttnDownBlock2D",
26
+ "CrossAttnDownBlock2D",
27
+ "DownBlock2D"
28
+ ],
29
+ "downsample_padding": 1,
30
+ "dual_cross_attention": false,
31
+ "encoder_hid_dim": null,
32
+ "encoder_hid_dim_type": null,
33
+ "flip_sin_to_cos": true,
34
+ "freq_shift": 0,
35
+ "in_channels": 4,
36
+ "inv_gamma": 1.0,
37
+ "layers_per_block": 2,
38
+ "mid_block_only_cross_attention": null,
39
+ "mid_block_scale_factor": 1,
40
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
41
+ "min_decay": 0.0,
42
+ "norm_eps": 1e-05,
43
+ "norm_num_groups": 32,
44
+ "num_class_embeds": null,
45
+ "only_cross_attention": false,
46
+ "optimization_step": 2000,
47
+ "out_channels": 4,
48
+ "power": 0.6666666666666666,
49
+ "projection_class_embeddings_input_dim": null,
50
+ "resnet_out_scale_factor": 1.0,
51
+ "resnet_skip_time_act": false,
52
+ "resnet_time_scale_shift": "default",
53
+ "sample_size": 32,
54
+ "time_cond_proj_dim": null,
55
+ "time_embedding_act_fn": null,
56
+ "time_embedding_dim": null,
57
+ "time_embedding_type": "positional",
58
+ "timestep_post_act": null,
59
+ "up_block_types": [
60
+ "UpBlock2D",
61
+ "CrossAttnUpBlock2D",
62
+ "CrossAttnUpBlock2D",
63
+ "CrossAttnUpBlock2D"
64
+ ],
65
+ "upcast_attention": false,
66
+ "update_after_step": 0,
67
+ "use_ema_warmup": false,
68
+ "use_linear_projection": false
69
+ }
checkpoint-2000/unet_ema/diffusion_pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4fe95873a21bb642022b065697c3dff8687b21a7538bd7c13398bc2eb007b2d
3
+ size 3438366373
checkpoint-2500/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2f61539be406a4e4016b8c59f3ae3a5e99afe4e87df97b45b7f7dbf8bea506c
3
+ size 6876749715
checkpoint-2500/random_states_0.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5d4d487c8a51beb0f84a15ce239790b1596b6916e5edb2227377dec6eceaa71
3
+ size 14663
checkpoint-2500/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0fbcebc8f5487b0c117b5dd47f2ea304af3eebf408d297118d9307e1223927e1
3
+ size 557
checkpoint-2500/scheduler.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:22b1db9332b702ad998c57267be2dca991aa8ef4e47b4e515dc19c7f20063a70
3
+ size 563
checkpoint-2500/unet/config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "UNet2DConditionModel",
3
+ "_diffusers_version": "0.17.0.dev0",
4
+ "_name_or_path": "riffusion/riffusion-model-v1",
5
+ "act_fn": "silu",
6
+ "addition_embed_type": null,
7
+ "addition_embed_type_num_heads": 64,
8
+ "attention_head_dim": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "center_input_sample": false,
16
+ "class_embed_type": null,
17
+ "class_embeddings_concat": false,
18
+ "conv_in_kernel": 3,
19
+ "conv_out_kernel": 3,
20
+ "cross_attention_dim": 768,
21
+ "cross_attention_norm": null,
22
+ "down_block_types": [
23
+ "CrossAttnDownBlock2D",
24
+ "CrossAttnDownBlock2D",
25
+ "CrossAttnDownBlock2D",
26
+ "DownBlock2D"
27
+ ],
28
+ "downsample_padding": 1,
29
+ "dual_cross_attention": false,
30
+ "encoder_hid_dim": null,
31
+ "encoder_hid_dim_type": null,
32
+ "flip_sin_to_cos": true,
33
+ "freq_shift": 0,
34
+ "in_channels": 4,
35
+ "layers_per_block": 2,
36
+ "mid_block_only_cross_attention": null,
37
+ "mid_block_scale_factor": 1,
38
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
39
+ "norm_eps": 1e-05,
40
+ "norm_num_groups": 32,
41
+ "num_class_embeds": null,
42
+ "only_cross_attention": false,
43
+ "out_channels": 4,
44
+ "projection_class_embeddings_input_dim": null,
45
+ "resnet_out_scale_factor": 1.0,
46
+ "resnet_skip_time_act": false,
47
+ "resnet_time_scale_shift": "default",
48
+ "sample_size": 32,
49
+ "time_cond_proj_dim": null,
50
+ "time_embedding_act_fn": null,
51
+ "time_embedding_dim": null,
52
+ "time_embedding_type": "positional",
53
+ "timestep_post_act": null,
54
+ "up_block_types": [
55
+ "UpBlock2D",
56
+ "CrossAttnUpBlock2D",
57
+ "CrossAttnUpBlock2D",
58
+ "CrossAttnUpBlock2D"
59
+ ],
60
+ "upcast_attention": false,
61
+ "use_linear_projection": false
62
+ }
checkpoint-2500/unet/diffusion_pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:259ba0857d3a9f596c236b1d13ac247d9b0b7d0daad2c19bca44afbc42254e1c
3
+ size 3438375973
checkpoint-2500/unet_ema/config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "UNet2DConditionModel",
3
+ "_diffusers_version": "0.17.0.dev0",
4
+ "_name_or_path": "riffusion/riffusion-model-v1",
5
+ "act_fn": "silu",
6
+ "addition_embed_type": null,
7
+ "addition_embed_type_num_heads": 64,
8
+ "attention_head_dim": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "center_input_sample": false,
16
+ "class_embed_type": null,
17
+ "class_embeddings_concat": false,
18
+ "conv_in_kernel": 3,
19
+ "conv_out_kernel": 3,
20
+ "cross_attention_dim": 768,
21
+ "cross_attention_norm": null,
22
+ "decay": 0.9999,
23
+ "down_block_types": [
24
+ "CrossAttnDownBlock2D",
25
+ "CrossAttnDownBlock2D",
26
+ "CrossAttnDownBlock2D",
27
+ "DownBlock2D"
28
+ ],
29
+ "downsample_padding": 1,
30
+ "dual_cross_attention": false,
31
+ "encoder_hid_dim": null,
32
+ "encoder_hid_dim_type": null,
33
+ "flip_sin_to_cos": true,
34
+ "freq_shift": 0,
35
+ "in_channels": 4,
36
+ "inv_gamma": 1.0,
37
+ "layers_per_block": 2,
38
+ "mid_block_only_cross_attention": null,
39
+ "mid_block_scale_factor": 1,
40
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
41
+ "min_decay": 0.0,
42
+ "norm_eps": 1e-05,
43
+ "norm_num_groups": 32,
44
+ "num_class_embeds": null,
45
+ "only_cross_attention": false,
46
+ "optimization_step": 2500,
47
+ "out_channels": 4,
48
+ "power": 0.6666666666666666,
49
+ "projection_class_embeddings_input_dim": null,
50
+ "resnet_out_scale_factor": 1.0,
51
+ "resnet_skip_time_act": false,
52
+ "resnet_time_scale_shift": "default",
53
+ "sample_size": 32,
54
+ "time_cond_proj_dim": null,
55
+ "time_embedding_act_fn": null,
56
+ "time_embedding_dim": null,
57
+ "time_embedding_type": "positional",
58
+ "timestep_post_act": null,
59
+ "up_block_types": [
60
+ "UpBlock2D",
61
+ "CrossAttnUpBlock2D",
62
+ "CrossAttnUpBlock2D",
63
+ "CrossAttnUpBlock2D"
64
+ ],
65
+ "upcast_attention": false,
66
+ "update_after_step": 0,
67
+ "use_ema_warmup": false,
68
+ "use_linear_projection": false
69
+ }
checkpoint-2500/unet_ema/diffusion_pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9de771422d95a4a2da87ea9a3d1576a5b6ea25bfa614e731bd668d8e001078eb
3
+ size 3438366373
checkpoint-3000/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:899d79341280cf8ab25b598282ed77eb901d0df222deaae707b7926303db323a
3
+ size 6876749715
checkpoint-3000/random_states_0.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:44e0ce5d222d66132c6a4d5e0d9dff2ebb1d586c7a60cba1e6562443072539ad
3
+ size 14663
checkpoint-3000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb1f9398b77268202e8e1465734a63d123b1ef11c27f20f2473677e9883a6869
3
+ size 557
checkpoint-3000/scheduler.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3d6ec0338061ab706b1805a9e2966ca126b26619cd90bb73133127681b9de370
3
+ size 563
checkpoint-3000/unet/config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "UNet2DConditionModel",
3
+ "_diffusers_version": "0.17.0.dev0",
4
+ "_name_or_path": "riffusion/riffusion-model-v1",
5
+ "act_fn": "silu",
6
+ "addition_embed_type": null,
7
+ "addition_embed_type_num_heads": 64,
8
+ "attention_head_dim": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "center_input_sample": false,
16
+ "class_embed_type": null,
17
+ "class_embeddings_concat": false,
18
+ "conv_in_kernel": 3,
19
+ "conv_out_kernel": 3,
20
+ "cross_attention_dim": 768,
21
+ "cross_attention_norm": null,
22
+ "down_block_types": [
23
+ "CrossAttnDownBlock2D",
24
+ "CrossAttnDownBlock2D",
25
+ "CrossAttnDownBlock2D",
26
+ "DownBlock2D"
27
+ ],
28
+ "downsample_padding": 1,
29
+ "dual_cross_attention": false,
30
+ "encoder_hid_dim": null,
31
+ "encoder_hid_dim_type": null,
32
+ "flip_sin_to_cos": true,
33
+ "freq_shift": 0,
34
+ "in_channels": 4,
35
+ "layers_per_block": 2,
36
+ "mid_block_only_cross_attention": null,
37
+ "mid_block_scale_factor": 1,
38
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
39
+ "norm_eps": 1e-05,
40
+ "norm_num_groups": 32,
41
+ "num_class_embeds": null,
42
+ "only_cross_attention": false,
43
+ "out_channels": 4,
44
+ "projection_class_embeddings_input_dim": null,
45
+ "resnet_out_scale_factor": 1.0,
46
+ "resnet_skip_time_act": false,
47
+ "resnet_time_scale_shift": "default",
48
+ "sample_size": 32,
49
+ "time_cond_proj_dim": null,
50
+ "time_embedding_act_fn": null,
51
+ "time_embedding_dim": null,
52
+ "time_embedding_type": "positional",
53
+ "timestep_post_act": null,
54
+ "up_block_types": [
55
+ "UpBlock2D",
56
+ "CrossAttnUpBlock2D",
57
+ "CrossAttnUpBlock2D",
58
+ "CrossAttnUpBlock2D"
59
+ ],
60
+ "upcast_attention": false,
61
+ "use_linear_projection": false
62
+ }
checkpoint-3000/unet/diffusion_pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:656aade4d9d3593cf7334e92bd8137a1ac4a2ebb2680825bb7c3d09687164195
3
+ size 3438375973
checkpoint-3000/unet_ema/config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "UNet2DConditionModel",
3
+ "_diffusers_version": "0.17.0.dev0",
4
+ "_name_or_path": "riffusion/riffusion-model-v1",
5
+ "act_fn": "silu",
6
+ "addition_embed_type": null,
7
+ "addition_embed_type_num_heads": 64,
8
+ "attention_head_dim": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "center_input_sample": false,
16
+ "class_embed_type": null,
17
+ "class_embeddings_concat": false,
18
+ "conv_in_kernel": 3,
19
+ "conv_out_kernel": 3,
20
+ "cross_attention_dim": 768,
21
+ "cross_attention_norm": null,
22
+ "decay": 0.9999,
23
+ "down_block_types": [
24
+ "CrossAttnDownBlock2D",
25
+ "CrossAttnDownBlock2D",
26
+ "CrossAttnDownBlock2D",
27
+ "DownBlock2D"
28
+ ],
29
+ "downsample_padding": 1,
30
+ "dual_cross_attention": false,
31
+ "encoder_hid_dim": null,
32
+ "encoder_hid_dim_type": null,
33
+ "flip_sin_to_cos": true,
34
+ "freq_shift": 0,
35
+ "in_channels": 4,
36
+ "inv_gamma": 1.0,
37
+ "layers_per_block": 2,
38
+ "mid_block_only_cross_attention": null,
39
+ "mid_block_scale_factor": 1,
40
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
41
+ "min_decay": 0.0,
42
+ "norm_eps": 1e-05,
43
+ "norm_num_groups": 32,
44
+ "num_class_embeds": null,
45
+ "only_cross_attention": false,
46
+ "optimization_step": 3000,
47
+ "out_channels": 4,
48
+ "power": 0.6666666666666666,
49
+ "projection_class_embeddings_input_dim": null,
50
+ "resnet_out_scale_factor": 1.0,
51
+ "resnet_skip_time_act": false,
52
+ "resnet_time_scale_shift": "default",
53
+ "sample_size": 32,
54
+ "time_cond_proj_dim": null,
55
+ "time_embedding_act_fn": null,
56
+ "time_embedding_dim": null,
57
+ "time_embedding_type": "positional",
58
+ "timestep_post_act": null,
59
+ "up_block_types": [
60
+ "UpBlock2D",
61
+ "CrossAttnUpBlock2D",
62
+ "CrossAttnUpBlock2D",
63
+ "CrossAttnUpBlock2D"
64
+ ],
65
+ "upcast_attention": false,
66
+ "update_after_step": 0,
67
+ "use_ema_warmup": false,
68
+ "use_linear_projection": false
69
+ }
checkpoint-3000/unet_ema/diffusion_pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee58da74433f0ca4d60b60894723d66714c91377ae1d6bcf9d6e86cb82924e6f
3
+ size 3438366373
checkpoint-3500/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7054101f9759a82cde6803202862af0f9c8014c2770938dd9f14910c72893719
3
+ size 6876749715
checkpoint-3500/random_states_0.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:92d31895e6fe491199de2ea5c0878f9c10e9698975e15a95cb85852166d13444
3
+ size 14663
checkpoint-3500/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4cfe72973031c5a482f6e9ed0cb664a6aa7f63116e294e2b1f72d360d9b033c
3
+ size 557
checkpoint-3500/scheduler.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b516ff84772ee7d9b4425cc1c7028178b9ec20a018c7a7b587a5e2996b63d080
3
+ size 563
checkpoint-3500/unet/config.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "UNet2DConditionModel",
3
+ "_diffusers_version": "0.17.0.dev0",
4
+ "_name_or_path": "riffusion/riffusion-model-v1",
5
+ "act_fn": "silu",
6
+ "addition_embed_type": null,
7
+ "addition_embed_type_num_heads": 64,
8
+ "attention_head_dim": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "center_input_sample": false,
16
+ "class_embed_type": null,
17
+ "class_embeddings_concat": false,
18
+ "conv_in_kernel": 3,
19
+ "conv_out_kernel": 3,
20
+ "cross_attention_dim": 768,
21
+ "cross_attention_norm": null,
22
+ "down_block_types": [
23
+ "CrossAttnDownBlock2D",
24
+ "CrossAttnDownBlock2D",
25
+ "CrossAttnDownBlock2D",
26
+ "DownBlock2D"
27
+ ],
28
+ "downsample_padding": 1,
29
+ "dual_cross_attention": false,
30
+ "encoder_hid_dim": null,
31
+ "encoder_hid_dim_type": null,
32
+ "flip_sin_to_cos": true,
33
+ "freq_shift": 0,
34
+ "in_channels": 4,
35
+ "layers_per_block": 2,
36
+ "mid_block_only_cross_attention": null,
37
+ "mid_block_scale_factor": 1,
38
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
39
+ "norm_eps": 1e-05,
40
+ "norm_num_groups": 32,
41
+ "num_class_embeds": null,
42
+ "only_cross_attention": false,
43
+ "out_channels": 4,
44
+ "projection_class_embeddings_input_dim": null,
45
+ "resnet_out_scale_factor": 1.0,
46
+ "resnet_skip_time_act": false,
47
+ "resnet_time_scale_shift": "default",
48
+ "sample_size": 32,
49
+ "time_cond_proj_dim": null,
50
+ "time_embedding_act_fn": null,
51
+ "time_embedding_dim": null,
52
+ "time_embedding_type": "positional",
53
+ "timestep_post_act": null,
54
+ "up_block_types": [
55
+ "UpBlock2D",
56
+ "CrossAttnUpBlock2D",
57
+ "CrossAttnUpBlock2D",
58
+ "CrossAttnUpBlock2D"
59
+ ],
60
+ "upcast_attention": false,
61
+ "use_linear_projection": false
62
+ }
checkpoint-3500/unet/diffusion_pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f536ec67de2967198fa8448f24d04c7fd6e7c25ad1f8c9ba32aef2f591680a67
3
+ size 3438375973
checkpoint-3500/unet_ema/config.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "UNet2DConditionModel",
3
+ "_diffusers_version": "0.17.0.dev0",
4
+ "_name_or_path": "riffusion/riffusion-model-v1",
5
+ "act_fn": "silu",
6
+ "addition_embed_type": null,
7
+ "addition_embed_type_num_heads": 64,
8
+ "attention_head_dim": 8,
9
+ "block_out_channels": [
10
+ 320,
11
+ 640,
12
+ 1280,
13
+ 1280
14
+ ],
15
+ "center_input_sample": false,
16
+ "class_embed_type": null,
17
+ "class_embeddings_concat": false,
18
+ "conv_in_kernel": 3,
19
+ "conv_out_kernel": 3,
20
+ "cross_attention_dim": 768,
21
+ "cross_attention_norm": null,
22
+ "decay": 0.9999,
23
+ "down_block_types": [
24
+ "CrossAttnDownBlock2D",
25
+ "CrossAttnDownBlock2D",
26
+ "CrossAttnDownBlock2D",
27
+ "DownBlock2D"
28
+ ],
29
+ "downsample_padding": 1,
30
+ "dual_cross_attention": false,
31
+ "encoder_hid_dim": null,
32
+ "encoder_hid_dim_type": null,
33
+ "flip_sin_to_cos": true,
34
+ "freq_shift": 0,
35
+ "in_channels": 4,
36
+ "inv_gamma": 1.0,
37
+ "layers_per_block": 2,
38
+ "mid_block_only_cross_attention": null,
39
+ "mid_block_scale_factor": 1,
40
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
41
+ "min_decay": 0.0,
42
+ "norm_eps": 1e-05,
43
+ "norm_num_groups": 32,
44
+ "num_class_embeds": null,
45
+ "only_cross_attention": false,
46
+ "optimization_step": 3500,
47
+ "out_channels": 4,
48
+ "power": 0.6666666666666666,
49
+ "projection_class_embeddings_input_dim": null,
50
+ "resnet_out_scale_factor": 1.0,
51
+ "resnet_skip_time_act": false,
52
+ "resnet_time_scale_shift": "default",
53
+ "sample_size": 32,
54
+ "time_cond_proj_dim": null,
55
+ "time_embedding_act_fn": null,
56
+ "time_embedding_dim": null,
57
+ "time_embedding_type": "positional",
58
+ "timestep_post_act": null,
59
+ "up_block_types": [
60
+ "UpBlock2D",
61
+ "CrossAttnUpBlock2D",
62
+ "CrossAttnUpBlock2D",
63
+ "CrossAttnUpBlock2D"
64
+ ],
65
+ "upcast_attention": false,
66
+ "update_after_step": 0,
67
+ "use_ema_warmup": false,
68
+ "use_linear_projection": false
69
+ }
checkpoint-3500/unet_ema/diffusion_pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f8b5743dc141397f08f15f609af6d5fe9536357165930faa8569d0f6ef0c77a
3
+ size 3438366373
checkpoint-4000/optimizer.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e64081e64a0b63f6dc2293b063eb83d1a53cf71d4182ec8c18b35355a38cf2e3
3
+ size 6876749715
checkpoint-4000/random_states_0.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:402fc84a1e8091d8f68cea31aaaf2638d5bad61d5e43c32f033ff4fd1499d081
3
+ size 14663