Added trained continuer pipeline

logs/train_continuer/events.out.tfevents.1684026981.127x12.395141.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bbabfb31b45cf6c795fbb444d66563b77410ba0d779e047a54c5d2c044d62f71
+size 11673405

mel/mel_config.json

@@ -0,0 +1,11 @@
+{
+  "_class_name": "Mel",
+  "_diffusers_version": "0.16.1",
+  "hop_length": 512,
+  "n_fft": 2048,
+  "n_iter": 32,
+  "sample_rate": 22050,
+  "top_db": 80,
+  "x_res": 256,
+  "y_res": 256
+}

model_index.json

@@ -0,0 +1,20 @@
+{
+  "_class_name": "AudioDiffusionPipeline",
+  "_diffusers_version": "0.16.1",
+  "mel": [
+    "audio_diffusion",
+    "Mel"
+  ],
+  "scheduler": [
+    "diffusers",
+    "DDPMScheduler"
+  ],
+  "unet": [
+    "diffusers",
+    "UNet2DModel"
+  ],
+  "vqvae": [
+    "diffusers",
+    "AutoencoderKL"
+  ]
+}

scheduler/scheduler_config.json

@@ -0,0 +1,16 @@
+{
+  "_class_name": "DDPMScheduler",
+  "_diffusers_version": "0.16.1",
+  "beta_end": 0.02,
+  "beta_schedule": "linear",
+  "beta_start": 0.0001,
+  "clip_sample": true,
+  "clip_sample_range": 1.0,
+  "dynamic_thresholding_ratio": 0.995,
+  "num_train_timesteps": 1000,
+  "prediction_type": "epsilon",
+  "sample_max_value": 1.0,
+  "thresholding": false,
+  "trained_betas": null,
+  "variance_type": "fixed_small"
+}

unet/config.json

@@ -0,0 +1,49 @@
+{
+  "_class_name": "UNet2DModel",
+  "_diffusers_version": "0.16.1",
+  "act_fn": "silu",
+  "add_attention": true,
+  "attention_head_dim": 8,
+  "block_out_channels": [
+    128,
+    128,
+    256,
+    256,
+    512,
+    512
+  ],
+  "center_input_sample": false,
+  "class_embed_type": null,
+  "down_block_types": [
+    "DownBlock2D",
+    "DownBlock2D",
+    "DownBlock2D",
+    "DownBlock2D",
+    "DownBlock2D",
+    "DownBlock2D"
+  ],
+  "downsample_padding": 1,
+  "flip_sin_to_cos": true,
+  "freq_shift": 0,
+  "in_channels": 3,
+  "layers_per_block": 2,
+  "mid_block_scale_factor": 1,
+  "norm_eps": 1e-05,
+  "norm_num_groups": 32,
+  "num_class_embeds": null,
+  "out_channels": 1,
+  "resnet_time_scale_shift": "default",
+  "sample_size": [
+    32,
+    32
+  ],
+  "time_embedding_type": "positional",
+  "up_block_types": [
+    "UpBlock2D",
+    "UpBlock2D",
+    "UpBlock2D",
+    "UpBlock2D",
+    "UpBlock2D",
+    "UpBlock2D"
+  ]
+}

unet/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f46482642ab61f510ea8d1453993a8a71dfee923b660f5c0246716e632be0fb5
+size 433806081

vqvae/config.json

@@ -0,0 +1,31 @@
+{
+  "_class_name": "AutoencoderKL",
+  "_diffusers_version": "0.16.1",
+  "_name_or_path": "//home/kroohpar/.cache/huggingface/hub/models--teticio--latent-audio-diffusion-256/snapshots/e52078e262cf9f5683e953b7e7be2349d90c3a8c/vqvae",
+  "act_fn": "silu",
+  "block_out_channels": [
+    128,
+    256,
+    512,
+    512
+  ],
+  "down_block_types": [
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D",
+    "DownEncoderBlock2D"
+  ],
+  "in_channels": 1,
+  "latent_channels": 1,
+  "layers_per_block": 2,
+  "norm_num_groups": 32,
+  "out_channels": 1,
+  "sample_size": 256,
+  "scaling_factor": 0.18215,
+  "up_block_types": [
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D",
+    "UpDecoderBlock2D"
+  ]
+}

vqvae/diffusion_pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f8029799a5d21c04675913de95b3e74db77b1506696c8d67b3facd59f5eee49
+size 334530673