Use real-world Electrocardiogram (ECG) data to detect anomalies in a patient heartbeat. We'll build an LSTM Autoencoder, train it on a set of normal heartbeats and classify unseen examples as normal or anomalies.
Subscribe: [ Ссылка ]
Complete tutorial + source code: [ Ссылка ]
GitHub: [ Ссылка ]
📖 Read Hacker's Guide to Machine Learning with Python: [ Ссылка ]
⭐️ Tutorial Contents ⭐️
(04:35) Load the ECG data
(14:09) Exploratory Data Analysis
(23:29) Data preprocessing
(33:30) Build an LSTM Autoencoder with PyTorch
(43:07) Training
(50:58) Loading pre-trained model
(51:53) Choosing a threshold for anomaly detection
(55:36) Finding abnormal heartbeats
#TimeSeries #AnomalyDetection #LSTMAutoencoder #PyTorch #Python
