DPMHM (prononced "Deep HM") stands for Diagnosis and Prognosis for Machine Health Monitoring. It is a library written in Python and Tensorflow for benchmark test of deep-learning based algorithms in the context of machine health monitoring (MHM), featuring
- A collection of open-access database of vibration signals, standardized and formatted using
tensorflow-datasets - Pipelines for automatic preprocessing & transformation of data
- Baseline algorithms: Supervised/Unsupervised/Self-Supervised representation learning, Transfer learning/Few-shot learning for fault diagnosis, Remaining Useful Life (RUL) prediction etc.
Documentation: https://yanncalec.github.io/dpmhm/