audio_test / model_config_float_conditioning_dit_all.json
“fred-dev”
added repo files
5915064
{
"model_type": "diffusion_cond",
"sample_size": 1048576,
"sample_rate": 44100,
"audio_channels": 1,
"model": {
"pretransform": {
"type": "autoencoder",
"iterate_batch": true,
"config": {
"encoder": {
"type": "dac",
"config": {
"in_channels": 1,
"latent_dim": 128,
"d_model": 128,
"strides": [4, 4, 8, 8]
}
},
"decoder": {
"type": "dac",
"config": {
"out_channels": 1,
"latent_dim": 64,
"channels": 1536,
"rates": [8, 8, 4, 4]
}
},
"bottleneck": {
"type": "vae"
},
"latent_dim": 64,
"downsampling_ratio": 1024,
"io_channels": 1
}
},
"conditioning": {
"configs": [
{
"id": "latitude",
"type": "number",
"config": {
"min_val": -54.617412,
"max_val": -10.13994
}
},
{
"id": "longitude",
"type": "number",
"config": {
"min_val": 96.8233,
"max_val": 167.9619
}
},
{
"id": "temperature",
"type": "number",
"config": {
"min_val": -10.0,
"max_val": 55.0
}
},
{
"id": "humidity",
"type": "number",
"config": {
"min_val": 1,
"max_val": 100.0
}
},
{
"id": "wind_speed",
"type": "number",
"config": {
"min_val": 0,
"max_val": 50.0
}
},
{
"id": "pressure",
"type": "number",
"config": {
"min_val": 800.0,
"max_val": 1200.0
}
},
{
"id": "minutes_of_day",
"type": "number",
"config": {
"min_val": 0,
"max_val": 1439
}
},
{
"id": "day_of_year",
"type": "number",
"config": {
"min_val": 1,
"max_val": 366
}
},
{
"id": "seconds_start",
"type": "number",
"config": {
"min_val": 0,
"max_val": 512
}
},
{
"id": "seconds_total",
"type": "number",
"config": {
"min_val": 0,
"max_val": 512
}
}
],
"cond_dim": 768
},
"diffusion": {
"cross_attention_cond_ids": ["latitude", "longitude", "temperature", "humidity", "wind_speed", "pressure", "minutes_of_day", "day_of_year","seconds_start", "seconds_total"],
"global_cond_ids": ["seconds_start", "seconds_total"],
"type": "dit",
"config": {
"io_channels": 64,
"embed_dim": 768,
"depth": 24,
"num_heads": 24,
"cond_token_dim": 768,
"global_cond_dim": 1536,
"project_cond_tokens": false,
"transformer_type": "continuous_transformer"
}
},
"io_channels": 64
},
"training": {
"use_ema": true,
"log_loss_info": false,
"optimizer_configs": {
"diffusion": {
"optimizer": {
"type": "AdamW",
"config": {
"lr": 5e-5,
"betas": [0.9, 0.999],
"weight_decay": 1e-3
}
},
"scheduler": {
"type": "InverseLR",
"config": {
"inv_gamma": 1000000,
"power": 0.5,
"warmup": 0.99
}
}
}
},
"demo": {
"demo_every": 2500,
"demo_steps": 100,
"num_demos": 3,
"demo_cfg_scales": [3, 5, 7],
"demo_cond": [
{
"latitude": -24.005512,
"longitude": 133.368348,
"temperature": 25.5,
"humidity": 60,
"wind_speed": 8,
"pressure": 1000,
"minutes_of_day": 400,
"day_of_year": 110,
"seconds_start": 0,
"seconds_total": 22
},
{
"latitude": -26.987815,
"longitude": 153.129068,
"temperature": 31.5,
"humidity": 70,
"wind_speed": 12,
"pressure": 1010,
"minutes_of_day": 600,
"day_of_year": 57,
"seconds_start": 0,
"seconds_total": 22
},
{
"latitude": -12.546364,
"longitude": 130.919605,
"temperature": 28.5,
"humidity": 60,
"wind_speed": 18,
"pressure": 1015,
"minutes_of_day": 1140,
"day_of_year": 280,
"seconds_start": 0,
"seconds_total": 22
}
]
}
}
}