recipes.cad2.task1.ConvTasNet package

Subpackages

• recipes.cad2.task1.ConvTasNet.local package
  • Submodules
  • recipes.cad2.task1.ConvTasNet.local.musdb18_dataset module
    • Compose
    • MUSDB18Dataset
      • MUSDB18Dataset.root
      • MUSDB18Dataset.sources
      • MUSDB18Dataset.suffix
      • MUSDB18Dataset.split
      • MUSDB18Dataset.subset
      • MUSDB18Dataset.segment
      • MUSDB18Dataset.samples_per_track
      • MUSDB18Dataset.random_segments
      • MUSDB18Dataset.source_augmentations
      • MUSDB18Dataset.sample_rate
      • MUSDB18Dataset.tracks
      • MUSDB18Dataset.dataset_name
      • MUSDB18Dataset.get_infos()
      • MUSDB18Dataset.get_tracks()
      • MUSDB18Dataset.root
    • augment_channelswap()
    • augment_gain()
  • recipes.cad2.task1.ConvTasNet.local.system module
    • System
      • System.allow_zero_length_dataloader_with_multiple_devices
      • System.common_step()
      • System.config_to_hparams()
      • System.configure_optimizers()
      • System.default_monitor
      • System.forward()
      • System.lr_scheduler_step()
      • System.on_save_checkpoint()
      • System.on_validation_epoch_end()
      • System.prepare_data_per_node
      • System.train_dataloader()
      • System.training
      • System.training_step()
      • System.val_dataloader()
      • System.validation_step()
    • flatten_dict()
  • recipes.cad2.task1.ConvTasNet.local.tasnet module
    • ChannelwiseLayerNorm
      • ChannelwiseLayerNorm.forward()
      • ChannelwiseLayerNorm.reset_parameters()
    • Chomp1d
      • Chomp1d.forward()
    • ConvTasNetStereo
      • ConvTasNetStereo.forward()
      • ConvTasNetStereo.get_model_args()
      • ConvTasNetStereo.get_state_dict()
      • ConvTasNetStereo.serialize()
      • ConvTasNetStereo.valid_length()
    • Decoder
      • Decoder.forward()
    • DepthwiseSeparableConv
      • DepthwiseSeparableConv.forward()
    • Encoder
      • Encoder.forward()
    • GlobalLayerNorm
      • GlobalLayerNorm.forward()
      • GlobalLayerNorm.reset_parameters()
    • TemporalBlock
      • TemporalBlock.forward()
    • TemporalConvNet
      • TemporalConvNet.forward()
    • chose_norm()
    • overlap_and_add()
  • Module contents
    • Compose
    • ConvTasNetStereo
      • ConvTasNetStereo.forward()
      • ConvTasNetStereo.get_model_args()
      • ConvTasNetStereo.get_state_dict()
      • ConvTasNetStereo.serialize()
      • ConvTasNetStereo.valid_length()
    • MUSDB18Dataset
      • MUSDB18Dataset.root
      • MUSDB18Dataset.sources
      • MUSDB18Dataset.suffix
      • MUSDB18Dataset.split
      • MUSDB18Dataset.subset
      • MUSDB18Dataset.segment
      • MUSDB18Dataset.samples_per_track
      • MUSDB18Dataset.random_segments
      • MUSDB18Dataset.source_augmentations
      • MUSDB18Dataset.sample_rate
      • MUSDB18Dataset.tracks
      • MUSDB18Dataset.dataset_name
      • MUSDB18Dataset.get_infos()
      • MUSDB18Dataset.get_tracks()
      • MUSDB18Dataset.root
    • augment_channelswap()
    • augment_gain()
    • overlap_and_add()

Submodules

recipes.cad2.task1.ConvTasNet.eval module

recipes.cad2.task1.ConvTasNet.eval.main(conf)

recipes.cad2.task1.ConvTasNet.train module

Script to train causal and noncausal ConvTasNet on MUSDB18. Model is trained to separate the vocals from the background music.

recipes.cad2.task1.ConvTasNet.train.create_callbacks(conf: dict, exp_dir: str) → tuple

Create callbacks for the training.

recipes.cad2.task1.ConvTasNet.train.create_datasets_and_loaders(conf: dict) → tuple

Create train and val datasets and loaders.

recipes.cad2.task1.ConvTasNet.train.create_model_and_optimizer(conf: dict) → tuple

Create model, optimizer and lr_scheduler.

recipes.cad2.task1.ConvTasNet.train.create_trainer(conf: dict, callbacks: list) → Trainer

Create PyTorch Lightning trainer.

recipes.cad2.task1.ConvTasNet.train.get_loss_func() → L1Loss

Return the loss function.

recipes.cad2.task1.ConvTasNet.train.load_config(config_path)

recipes.cad2.task1.ConvTasNet.train.main(conf)

recipes.cad2.task1.ConvTasNet.train.save_best_model(system, checkpoint, exp_dir, train_set)

Module contents