File size: 1,918 Bytes
1ca8b82 |
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 |
datasets:
train:
cls: DivideAndRemasterRandomChunkDataset
kwargs:
split: train
subset: ${....commons.datasets.subset}
stems: ${....commons.datasets.stems}
npy_memmap: ${....commons.datasets.npy_memmap}
full_audio_length_seconds: ${....commons.datasets.full_audio_length_seconds}
data_root: ${....commons.datasets.data_root}
chunk_size_seconds: ${....commons.datasets.chunk_size_seconds}
target_dataset_length: 16384 #10240
aligned_chunking: false
auto_include_mixture: true
recompute_mixture: true
val:
cls: DivideAndRemasterDeterministicChunkDataset
kwargs:
split: val
subset: ${....commons.datasets.subset}
stems: ${....commons.datasets.stems}
npy_memmap: ${....commons.datasets.npy_memmap}
full_audio_length_seconds: ${....commons.datasets.full_audio_length_seconds}
data_root: ${....commons.datasets.data_root}
chunk_size_seconds: ${....commons.datasets.chunk_size_seconds}
hop_size_seconds: ${....commons.datasets.hop_size_seconds}
aligned_chunking: true
auto_include_mixture: true
recompute_mixture: true
test:
cls: DivideAndRemasterFullTrackDataset
kwargs:
split: test
subset: ${....commons.datasets.subset}
stems: ${....commons.datasets.stems}
npy_memmap: ${....commons.datasets.npy_memmap}
full_audio_length_seconds: ${....commons.datasets.full_audio_length_seconds}
data_root: ${....commons.datasets.data_root}
commons:
datasets:
subset: multi-v2b
stems:
- speech
- music
- sfx
npy_memmap: true
full_audio_length_seconds: 60.0
chunk_size_seconds: 8.0
hop_size_seconds: 8.0
data_root: /fsx_vfx/projects/aa-cass/data/dnr-variants/dnr-v3-com-48k-smad-multi-v2
datamodule:
cls: DivideAndRemasterDataModule
kwargs:
batch_size_per_gpu: 8
num_workers: 8 |