/signalML

@CoditEU Project to train a real time anomaly detection model. MelSpectrogram extraction from .wav -> Data augmentation (rolling, stretching, etc...) -> Generating multi class label (specific failure) with PCA+KMeans -> CNN training on multiclass labeled MelSpectrograms -> Configuring HTTP endpoint with Flask & deploying with Docker -> Uploading to ACR (Azure Container Registry) & testing with ACI (Azure Container Instance)

Primary LanguagePythonApache License 2.0Apache-2.0

Watchers