/Multivariate-Classification

Primary LanguagePythonMIT LicenseMIT

Multivariate-Classification

Multivariate Time Series Classification Using LSTM

Multivariate Time Series Classification using LSTM - Keras

How to Run: main.py and Multi-variate-Time-series-Data.xlsx need to be on the same folder. The script will create four images

dataset Stats: Multi-variate-Time-series-Data.xlsx

  • Total Number of Time Series : 205
  • Data For Training : 72%
  • Data For Validation : 8%
  • Data For Testing : 20%

Method Overview : I have used Keras framework and an LSTM Network to design the model Train-Test Data Generation win_size = 89 represents time series size. num_var = 14 represents the feature vector size. Based no this I extract segmentsof 89*14 array. FUrtherr, we feed this time series to LSTM sequntially.

Model Sumary

Parameter setting split_ratio = 0.8 learning rate=0.001 nb_epoch=50 batch_size=64

Output:

  1. Performance of model on test set,
Test ROC Test Confusion matrix
  1. Performance of model on train set,
Train ROC Train Confusion matrix