sadaco.dataman.icbhi package

Submodules

sadaco.dataman.icbhi.demo module

sadaco.dataman.icbhi.demo.min_max_scale(samples, min, max)[source]

sadaco.dataman.icbhi.demo.load_input(input_path)[source]

class sadaco.dataman.icbhi.demo.demo_helper(master_cfg, model_cfg)[source]

Bases: object

do_inference(input_path, return_raw=False)[source]

do_explanation(input_path, method, cls)[source]

sadaco.dataman.icbhi.dummy module

class sadaco.dataman.icbhi.dummy.RespiDatasetSTFT(*args, **kwds)[source]

Bases: torch.utils.data.dataset.Dataset

__getitem__(index)[source]: returns: image, audio, nframes where image is a FloatTensor of size (3, H, W) audio is a FloatTensor of size (N_freq, N_frames) for spectrogram, or (N_frames) for waveform nframes is an integer

initialize(paths, multi_label)[source]

to_multi_hot(ann)[source]

to_one_hot(ann)[source]

to_int(ann)[source]

recover(mag, phase)[source]

sadaco.dataman.icbhi.icbhi module

class sadaco.dataman.icbhi.icbhi.ICBHI_Dataset(*args, **kwds)[source]

Bases: sadaco.dataman.base.BaseDataset

parse_label(sample, num_label=None, multi_label=False)[source]

Module contents