How sleep-revolution/esadafileconsumer works
Main Components
a. Data Input Layer:
• Entry point through RunOnOneFile.py and Processor.py
• File input handling capabilities
b. Core Processing Pipeline:
Into three main processing blocks:
Preprocessing Module
• Multiple preprocessing filters as subcomponents:
• Bandpass filter (bpfilter.py)
• Chebyshev high-pass filter (chebyhpfilter.py)
• Signal harmonisation (harmonisation.py)
• IQR standardization (iqrstd.py)
• Resampling (resample.py)
• Reshaping (reshape.py)
• Spectrogram generation (spectrograms.py)
Model Module
• Mixture models component (MixtureModels.py)
• Main model implementation (model.py)
• API integration (noxsasapi.py)
Data Processing Module
• Data generation (datagenerator.py)
• Part signal processing (partsignal.py)
• Prediction handling (prediction.py)
• Predictors management (predictors.py)
c. Utility Layer:
• Configuration management (yamlutils.py)
• Experiment saving (save_xp.py)
• System introspection (introspection.py)
[Input Layer]
• ↓
[Preprocessing Layer]
• ↓
[Model Layer]
• ↓
[Data Processing Layer]
• ↓
[Output/Results]
Additional Elements
• Version/dependency information from requirements.txt
• Docker container boundaries
• Entry points and exit points clearly
Implementation Notes
• Maintain consistent spacing and alignment
• Clear labels for all components
• Brief descriptions for complex interactions