Namerlight commited on
Commit
9474c35
1 Parent(s): 4c43fd3

Upload folder using huggingface_hub

Browse files
logs/train_unet/events.out.tfevents.1701857104.438b51897877.2574.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a85c54f2f98a8a90f40768730f4cfde5c2791e9730502b6dd521fcb86312778f
3
+ size 513730
logs/train_unet/events.out.tfevents.1701863381.438b51897877.2719.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ddfb2c343b3e8c8b7818292c0665331fa1092c7b2e4fc9f9c6c77bd8df551be3
3
+ size 88
logs/train_unet/events.out.tfevents.1701863414.438b51897877.2763.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df2ea0a74cfdbd10635880845b29b55f4a9ced90a1041bbb80eed45b3d4be90a
3
+ size 88
logs/train_unet/events.out.tfevents.1701863443.438b51897877.2807.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:879b0956b0a7c1072a09e2acbc4ea453fff4db9a7a9e98de62daf54990eb622c
3
+ size 88
logs/train_unet/events.out.tfevents.1701863542.438b51897877.2876.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:03d9d71d77dc7dda17b63c2eec0c57a0ee238021fcefce062b9efae0f199275f
3
+ size 88
logs/train_unet/events.out.tfevents.1701863916.438b51897877.165.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:18cd4c0465438021b094f0542e68a18d757f399bbe0cdefc1bfc68450c2ec9f2
3
+ size 88
logs/train_unet/events.out.tfevents.1701863951.438b51897877.215.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:90e0bfbfcc1103500c7bc6c28cbf83a2090581d253e194cf3f531c2d627bf8ed
3
+ size 1784580
mel/mel_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "Mel",
3
+ "_diffusers_version": "0.24.0",
4
+ "hop_length": 512,
5
+ "n_fft": 2048,
6
+ "n_iter": 32,
7
+ "sample_rate": 22050,
8
+ "top_db": 80,
9
+ "x_res": 128,
10
+ "y_res": 128
11
+ }
model_index.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "AudioDiffusionPipeline",
3
+ "_diffusers_version": "0.24.0",
4
+ "mel": [
5
+ "audio_diffusion",
6
+ "Mel"
7
+ ],
8
+ "scheduler": [
9
+ "diffusers",
10
+ "DDIMScheduler"
11
+ ],
12
+ "unet": [
13
+ "diffusers",
14
+ "UNet2DConditionModel"
15
+ ],
16
+ "vqvae": [
17
+ null,
18
+ null
19
+ ]
20
+ }
scheduler/scheduler_config.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "DDIMScheduler",
3
+ "_diffusers_version": "0.24.0",
4
+ "beta_end": 0.02,
5
+ "beta_schedule": "linear",
6
+ "beta_start": 0.0001,
7
+ "clip_sample": true,
8
+ "clip_sample_range": 1.0,
9
+ "dynamic_thresholding_ratio": 0.995,
10
+ "num_train_timesteps": 1000,
11
+ "prediction_type": "epsilon",
12
+ "rescale_betas_zero_snr": false,
13
+ "sample_max_value": 1.0,
14
+ "set_alpha_to_one": true,
15
+ "steps_offset": 0,
16
+ "thresholding": false,
17
+ "timestep_spacing": "leading",
18
+ "trained_betas": null
19
+ }
unet/config.json ADDED
@@ -0,0 +1,70 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "UNet2DConditionModel",
3
+ "_diffusers_version": "0.24.0",
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
+ "attention_type": "default",
10
+ "block_out_channels": [
11
+ 128,
12
+ 256,
13
+ 512,
14
+ 512
15
+ ],
16
+ "center_input_sample": false,
17
+ "class_embed_type": null,
18
+ "class_embeddings_concat": false,
19
+ "conv_in_kernel": 3,
20
+ "conv_out_kernel": 3,
21
+ "cross_attention_dim": 100,
22
+ "cross_attention_norm": null,
23
+ "down_block_types": [
24
+ "CrossAttnDownBlock2D",
25
+ "CrossAttnDownBlock2D",
26
+ "CrossAttnDownBlock2D",
27
+ "DownBlock2D"
28
+ ],
29
+ "downsample_padding": 1,
30
+ "dropout": 0.0,
31
+ "dual_cross_attention": false,
32
+ "encoder_hid_dim": null,
33
+ "encoder_hid_dim_type": null,
34
+ "flip_sin_to_cos": true,
35
+ "freq_shift": 0,
36
+ "in_channels": 1,
37
+ "layers_per_block": 2,
38
+ "mid_block_only_cross_attention": null,
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
+ "out_channels": 1,
47
+ "projection_class_embeddings_input_dim": null,
48
+ "resnet_out_scale_factor": 1.0,
49
+ "resnet_skip_time_act": false,
50
+ "resnet_time_scale_shift": "default",
51
+ "reverse_transformer_layers_per_block": null,
52
+ "sample_size": [
53
+ 128,
54
+ 128
55
+ ],
56
+ "time_cond_proj_dim": null,
57
+ "time_embedding_act_fn": null,
58
+ "time_embedding_dim": null,
59
+ "time_embedding_type": "positional",
60
+ "timestep_post_act": null,
61
+ "transformer_layers_per_block": 1,
62
+ "up_block_types": [
63
+ "UpBlock2D",
64
+ "CrossAttnUpBlock2D",
65
+ "CrossAttnUpBlock2D",
66
+ "CrossAttnUpBlock2D"
67
+ ],
68
+ "upcast_attention": false,
69
+ "use_linear_projection": false
70
+ }
unet/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af217b3d7e965626c3332e515428e16fcfa7ea4f62524a1b1cb1a4b731d8343c
3
+ size 542321276