Contents
What is LSTM HT and output?
LSTM cell has two kind of output: cell state and prediction, cell state is key factor for prediction. ht is consist of Ct,ht-1, xi. The first step in our LSTM is to decide what information we’re going to throw away from the cell state. sigmoid function squash value to 0~1 ，means let how much information through.
The output of an LSTM cell or layer of cells is called the hidden state. This is confusing, because each LSTM cell retains an internal state that is not output, called the cell state, or c. The LSTM hidden state output for the last time step. The LSTM hidden state output for the last time step (again).
What does h and C mean in a LSTM cell?
The LSTM cell also uses h (t-1) and c (t-1) as inputs. Here, the t-1 means from the previous time step. The c represents the cell state, so h (t-1) and c (t-1) are the previous output values and the previous state values. The interior of the LSTM cell architecture looks complicated. And it is, but not as complicated as you might guess.
How does LSTM cell do element wise multiplication?
so LSTM cell takes the previous memory state Ct-1 and does element wise multiplication with forget gate (f) if forget gate value is 1 then previous memory state is completely passed to the cell ( Remember f gate gives values between 0 and 1 ) Now with current memory state Ct we calculate new memory state from input state and C layer.
How is the previous memory state passed to LSTM cell?
I hope you get the idea. so LSTM cell takes the previous memory state Ct-1 and does element wise multiplication with forget gate (f) if forget gate value is 1 then previous memory state is completely passed to the cell ( Remember f gate gives values between 0 and 1 )
What are the gates of the LSTM cell?
Equations (1), (2) and (3) define three gates: a forget gate, an input gate and an output gate. Each gate is a vector of values between 0.0 and 1.0, which are used to determine how much information to forget (or, equivalently, remember) in each input-output cycle.