Hugging Face's logo

michaelpiro1
/
new_unet_fromLDM

Diffusers
Safetensors
DDPMPipeline
Model card Files
xet
Community
michaelpiro1 commited on
Commit
97c7b2f
·
verified ·
1 Parent(s): 725726c

Epoch 99

Browse files
Files changed (4) hide show
  1. model_index.json +12 -0
  2. scheduler/scheduler_config.json +19 -0
  3. unet/config.json +75 -0
  4. unet/diffusion_pytorch_model.safetensors +3 -0
model_index.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "DDPMPipeline",
3
+ "_diffusers_version": "0.29.1",
4
+ "scheduler": [
5
+ "diffusers",
6
+ "DDIMScheduler"
7
+ ],
8
+ "unet": [
9
+ "audioldm2",
10
+ "AudioLDM2UNet2DConditionModel"
11
+ ]
12
+ }
scheduler/scheduler_config.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "DDIMScheduler",
3
+ "_diffusers_version": "0.29.1",
4
+ "beta_end": 0.0195,
5
+ "beta_schedule": "scaled_linear",
6
+ "beta_start": 0.0015,
7
+ "clip_sample": false,
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": false,
15
+ "steps_offset": 1,
16
+ "thresholding": false,
17
+ "timestep_spacing": "leading",
18
+ "trained_betas": null
19
+ }
unet/config.json ADDED
@@ -0,0 +1,75 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_class_name": "AudioLDM2UNet2DConditionModel",
3
+ "_diffusers_version": "0.29.1",
4
+ "_name_or_path": "/root/.cache/huggingface/hub/models--cvssp--audioldm2-music/snapshots/0019cb0bd4d314accff72aee1c3abd6f22929ba5/unet",
5
+ "act_fn": "silu",
6
+ "attention_head_dim": 8,
7
+ "block_out_channels": [
8
+ 128,
9
+ 256,
10
+ 384,
11
+ 640
12
+ ],
13
+ "class_embed_type": null,
14
+ "class_embeddings_concat": false,
15
+ "conv_in_kernel": 3,
16
+ "conv_out_kernel": 3,
17
+ "cross_attention_dim": [
18
+ [
19
+ null,
20
+ 768,
21
+ 1024
22
+ ],
23
+ [
24
+ null,
25
+ 768,
26
+ 1024
27
+ ],
28
+ [
29
+ null,
30
+ 768,
31
+ 1024
32
+ ],
33
+ [
34
+ null,
35
+ 768,
36
+ 1024
37
+ ]
38
+ ],
39
+ "down_block_types": [
40
+ "DownBlock2D",
41
+ "CrossAttnDownBlock2D",
42
+ "CrossAttnDownBlock2D",
43
+ "CrossAttnDownBlock2D"
44
+ ],
45
+ "downsample_padding": 1,
46
+ "flip_sin_to_cos": true,
47
+ "freq_shift": 0,
48
+ "in_channels": 16,
49
+ "layers_per_block": 2,
50
+ "mid_block_scale_factor": 1,
51
+ "mid_block_type": "UNetMidBlock2DCrossAttn",
52
+ "norm_eps": 1e-05,
53
+ "norm_num_groups": 32,
54
+ "num_attention_heads": null,
55
+ "num_class_embeds": null,
56
+ "only_cross_attention": false,
57
+ "out_channels": 8,
58
+ "projection_class_embeddings_input_dim": null,
59
+ "resnet_time_scale_shift": "default",
60
+ "sample_size": 256,
61
+ "time_cond_proj_dim": null,
62
+ "time_embedding_act_fn": null,
63
+ "time_embedding_dim": null,
64
+ "time_embedding_type": "positional",
65
+ "timestep_post_act": null,
66
+ "transformer_layers_per_block": 1,
67
+ "up_block_types": [
68
+ "CrossAttnUpBlock2D",
69
+ "CrossAttnUpBlock2D",
70
+ "CrossAttnUpBlock2D",
71
+ "UpBlock2D"
72
+ ],
73
+ "upcast_attention": false,
74
+ "use_linear_projection": false
75
+ }
unet/diffusion_pytorch_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba4123a3311806e9eba8dc2db5d14fe38abda42c59c8fc520b439cc31ae753a8
3
+ size 1388001648