src.utils

Submodules

src.utils.loaders

src.utils.loaders.load_dataset(dataset_class, dataset_folder, dataset_config)[source]

This is a helper function that looks in the src.dataset module.

Parameters

  • dataset_class (str) – Name of the dataset class you want to

  • instantiate (e.g. Scaper, MixSourceFolder) –

  • dataset_folder (str) – Folder you want to load the data from.

  • dataset_config (dict) – Configuration of the dataset

Returns

Instantiated DatasetClass given the parameters.

Return type

torch.utils.data.Dataset

src.utils.loaders.load_model(model_config)[source]

Loads a deep SeparationModel given a model configuration.

Parameters

  • model_config (dict) – Model configuration with a ‘class’ key. The rest of the keys

  • put into the 'args'. (get) –

Returns

Instantiated deep model given the parameters.

Return type

SeparationModel

src.utils.parallel

src.utils.parallel.parallel_process(array, function, n_jobs=4, use_kwargs=False, front_num=0)[source]

A parallel version of the map function with a progress bar.

Parameters

  • array (array-like) – An array to iterate over.

  • function (function) – A python function to apply to the elements of array

  • n_jobs (int, default=16) – The number of cores to use

  • use_kwargs (boolean, default=False) – Whether to consider the elements of array as dictionaries of keyword arguments to function

  • front_num (int, default=3) – The number of iterations to run serially before kicking off the parallel job. Useful for catching bugs

Returns

A list containing the results of each function on each of the arguments in the array:

[function(array[0]), function(array[1]), ...].

Return type

list

src.utils.seed

src.utils.seed.seed(s=0)[source]

Seeds every possible random number generator with a specific seed to guarantee the same results with the same seed (hopefully). Currently supported RNGs are:

torch.manual_seed
torch.cuda.manual_seed
np.random.seed
random.seed

If you use RNGs outside of this list in your project, then there’s no guarantee of reproducibility.

Parameters

s (int, optional) – Seed to use. Defaults to 0.

Module contents