Contents Menu Expand Light mode Dark mode Auto light/dark, in light mode Auto light/dark, in dark mode Skip to content
musicality documentation
musicality documentation
  • Training & inference workflows
    • inference
  • Configuration
  • Augmentations
    • augmentations
  • Losses
    • tempo_regression
    • tempo_classification
    • beat_only
    • beat_phase
    • beat_position
    • phase_conditioning
    • pos_weight
    • shift_tolerance
  • Metrics
    • f_measure
    • continuity
    • confusion
    • position_accuracy
    • frame_accuracy
    • tempo_acc1
  • Postprocessing
    • postprocess
  • Loaders
    • tempo_dataset
    • beat_dataset
  • Models
    • tcn
    • tempo_net
    • huggingface
    • torch_audio
  • Trainers
    • tempo_module
    • beat_phase_module
    • train
    • train_beat_phase
  • Callbacks
    • error_plot
    • event_metrics
    • metrics_logger
  • Data formats & splits
    • dataformats
      • track_io
    • splitter
Back to top
View this page

train¶

Core training routine for tempo estimation.

Functions

build_callbacks(cfg)

build_dataloaders(cfg)

build_module(cfg)

train(cfg)

build_callbacks(cfg)[source]¶
Parameters:

cfg (DictConfig)

Return type:

list

build_dataloaders(cfg)[source]¶
Parameters:

cfg (DictConfig)

Return type:

tuple[DataLoader, DataLoader, int, int]

build_module(cfg)[source]¶
Parameters:

cfg (DictConfig)

Return type:

TempoModule

train(cfg)[source]¶
Parameters:

cfg (DictConfig)

Return type:

None

Next
train_beat_phase
Previous
beat_phase_module
Copyright © 2026, luczeng
Made with Sphinx and @pradyunsg's Furo
On this page
  • train
    • build_callbacks()
    • build_dataloaders()
    • build_module()
    • train()