Pytorch LSTM — shapes of input, hidden state, cell state and output

PyTorch LSTM — Shapes of Input, Hidden State, Cell State And Output

In Pytorch, to use an LSTM (with nn.LSTM()), we need to understand how the tensors representing the input time series, hidden state vector and cell state vector should be shaped. In this article, let us assume you are working with multivariate time series. Each multivariate time series in the dataset contains multiple univariate time series.

The following are the differences from pytorch’s LSTMCell discussed in the …

article deep learning hidden lstm machine learning neural networks pytorch recurrent neural network series state time series vector