Instructions to use sebascorreia/audio-diffusion2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use sebascorreia/audio-diffusion2 with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("sebascorreia/audio-diffusion2", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
Commit ·
18ce9bb
1
Parent(s): 24cc43d
Epoch 0
Browse files- logs/train_unet/events.out.tfevents.1690365386.b9f85c4c9c92.1745.0 +3 -0
- logs/train_unet/events.out.tfevents.1690365476.b9f85c4c9c92.2375.0 +3 -0
- logs/train_unet/events.out.tfevents.1690365668.b9f85c4c9c92.3261.0 +3 -0
- mel/mel_config.json +11 -0
- model_index.json +20 -0
- scheduler/scheduler_config.json +19 -0
- unet/config.json +51 -0
- unet/diffusion_pytorch_model.bin +3 -0
logs/train_unet/events.out.tfevents.1690365386.b9f85c4c9c92.1745.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9d9c4f135c8aaab2670660291d816e5132d9aafb0ead639228090d140fd882e6
|
| 3 |
+
size 88
|
logs/train_unet/events.out.tfevents.1690365476.b9f85c4c9c92.2375.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b2344e2ad546e044780eb87144f46e61c3aa281ce81c590cc7bd8e77e6711e1a
|
| 3 |
+
size 73080
|
logs/train_unet/events.out.tfevents.1690365668.b9f85c4c9c92.3261.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3bcaf86a244ee5bbca5a6ff3f58da3f9c1d139686677578511f93efeb46df860
|
| 3 |
+
size 943355
|
mel/mel_config.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_class_name": "Mel",
|
| 3 |
+
"_diffusers_version": "0.18.2",
|
| 4 |
+
"hop_length": 512,
|
| 5 |
+
"n_fft": 2048,
|
| 6 |
+
"n_iter": 32,
|
| 7 |
+
"sample_rate": 22050,
|
| 8 |
+
"top_db": 80,
|
| 9 |
+
"x_res": 256,
|
| 10 |
+
"y_res": 256
|
| 11 |
+
}
|
model_index.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_class_name": "AudioDiffusionPipeline",
|
| 3 |
+
"_diffusers_version": "0.18.2",
|
| 4 |
+
"mel": [
|
| 5 |
+
"audio_diffusion",
|
| 6 |
+
"Mel"
|
| 7 |
+
],
|
| 8 |
+
"scheduler": [
|
| 9 |
+
"diffusers",
|
| 10 |
+
"DDIMScheduler"
|
| 11 |
+
],
|
| 12 |
+
"unet": [
|
| 13 |
+
"diffusers",
|
| 14 |
+
"UNet2DModel"
|
| 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": "0.18.2",
|
| 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 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_class_name": "UNet2DModel",
|
| 3 |
+
"_diffusers_version": "0.18.2",
|
| 4 |
+
"act_fn": "silu",
|
| 5 |
+
"add_attention": true,
|
| 6 |
+
"attention_head_dim": 8,
|
| 7 |
+
"block_out_channels": [
|
| 8 |
+
128,
|
| 9 |
+
128,
|
| 10 |
+
256,
|
| 11 |
+
256,
|
| 12 |
+
512,
|
| 13 |
+
512
|
| 14 |
+
],
|
| 15 |
+
"center_input_sample": false,
|
| 16 |
+
"class_embed_type": null,
|
| 17 |
+
"down_block_types": [
|
| 18 |
+
"DownBlock2D",
|
| 19 |
+
"DownBlock2D",
|
| 20 |
+
"DownBlock2D",
|
| 21 |
+
"DownBlock2D",
|
| 22 |
+
"AttnDownBlock2D",
|
| 23 |
+
"DownBlock2D"
|
| 24 |
+
],
|
| 25 |
+
"downsample_padding": 1,
|
| 26 |
+
"downsample_type": "conv",
|
| 27 |
+
"flip_sin_to_cos": true,
|
| 28 |
+
"freq_shift": 0,
|
| 29 |
+
"in_channels": 1,
|
| 30 |
+
"layers_per_block": 2,
|
| 31 |
+
"mid_block_scale_factor": 1,
|
| 32 |
+
"norm_eps": 1e-05,
|
| 33 |
+
"norm_num_groups": 32,
|
| 34 |
+
"num_class_embeds": null,
|
| 35 |
+
"out_channels": 1,
|
| 36 |
+
"resnet_time_scale_shift": "default",
|
| 37 |
+
"sample_size": [
|
| 38 |
+
256,
|
| 39 |
+
256
|
| 40 |
+
],
|
| 41 |
+
"time_embedding_type": "positional",
|
| 42 |
+
"up_block_types": [
|
| 43 |
+
"UpBlock2D",
|
| 44 |
+
"AttnUpBlock2D",
|
| 45 |
+
"UpBlock2D",
|
| 46 |
+
"UpBlock2D",
|
| 47 |
+
"UpBlock2D",
|
| 48 |
+
"UpBlock2D"
|
| 49 |
+
],
|
| 50 |
+
"upsample_type": "conv"
|
| 51 |
+
}
|
unet/diffusion_pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e7e2878b17225494b9f4ef6aa57c3fb5bf95c80ad8bc5365183e2ab8760af9b8
|
| 3 |
+
size 454849533
|