File size: 2,849 Bytes
3978e51 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 |
audio:
chunk_size: 352800
dim_f: 1024
dim_t: 256
hop_length: 441
n_fft: 2048
num_channels: 2
sample_rate: 44100
min_mean_abs: 0.000
model:
dim: 384
depth: 6
stereo: true
num_stems: 4
time_transformer_depth: 1
freq_transformer_depth: 1
linear_transformer_depth: 0
num_bands: 60
dim_head: 64
heads: 8
attn_dropout: 0
ff_dropout: 0
flash_attn: True
dim_freqs_in: 1025
sample_rate: 44100 # needed for mel filter bank from librosa
stft_n_fft: 2048
stft_hop_length: 441
stft_win_length: 2048
stft_normalized: False
mask_estimator_depth: 2
multi_stft_resolution_loss_weight: 1.0
multi_stft_resolutions_window_sizes: !!python/tuple
- 4096
- 2048
- 1024
- 512
- 256
multi_stft_hop_size: 147
multi_stft_normalized: False
mlp_expansion_factor: 4 # Probably too big (requires a lot of memory for weights)
use_torch_checkpoint: False # it allows to greatly reduce GPU memory consumption during training (not fully tested)
skip_connection: False # Enable skip connection between transformer blocks - can solve problem with gradients and probably faster training
training:
batch_size: 1
gradient_accumulation_steps: 1
grad_clip: 0
instruments:
- drums
- bass
- other
- vocals
lr: 1.0e-05
patience: 2
reduce_factor: 0.95
target_instrument: null
num_epochs: 1000
num_steps: 1000
augmentation: false # enable augmentations by audiomentations and pedalboard
augmentation_type: null
use_mp3_compress: false # Deprecated
augmentation_mix: false # Mix several stems of the same type with some probability
augmentation_loudness: false # randomly change loudness of each stem
augmentation_loudness_type: 1 # Type 1 or 2
augmentation_loudness_min: 0
augmentation_loudness_max: 0
q: 0.95
coarse_loss_clip: false
ema_momentum: 0.999
optimizer: adam
other_fix: false # it's needed for checking on multisong dataset if other is actually instrumental
use_amp: true # enable or disable usage of mixed precision (float16) - usually it must be true
augmentations:
enable: true # enable or disable all augmentations (to fast disable if needed)
loudness: true # randomly change loudness of each stem on the range (loudness_min; loudness_max)
loudness_min: 0.5
loudness_max: 1.5
mixup: true # mix several stems of same type with some probability (only works for dataset types: 1, 2, 3)
mixup_probs:
!!python/tuple # 2 additional stems of the same type (1st with prob 0.2, 2nd with prob 0.02)
- 0.2
- 0.02
mixup_loudness_min: 0.5
mixup_loudness_max: 1.5
all:
channel_shuffle: 0.5 # Set 0 or lower to disable
random_inverse: 0.1 # inverse track (better lower probability)
random_polarity: 0.5 # polarity change (multiply waveform to -1)
inference:
batch_size: 4
dim_t: 256
num_overlap: 2 |