How to improve LSTM classification of time series

Question

I am training a Bidirectional LSTM for classification of time series for the purposes of anomaly detection. I create my sequences of input and labels (1 or 0) as such:

<