LSTM

Public technical reference documentation from Graiphic. For current SOTA installation and toolkit guides, visit the documentation center.

Description

Adds the weights of the LSTM layer to the weights table. Type : polymorphic.

Documentation illustration (illustration unavailable in the archive).

Input parameters

Weights in : array

name : string, name of layer.
Documentation illustration (illustration unavailable in the archive). weights : variant, weights values.

Documentation illustration (illustration unavailable in the archive).

name : string, name of layer.
input_weights : array, 2D values. input_weights = [features, 4*units].
hidden_weights : array, 2D values. hidden_weights = [units, 4*units].
biases : array, 1D values. biases = [4*units].

Output parameters

Weights out : array

name : string, name of layer.
Documentation illustration (illustration unavailable in the archive). weights : variant, weights values.

Documentation illustration (illustration unavailable in the archive).

Dimension

input_weights = [features, 4*units]

The size depends on the LSTM layer input and the units parameter.
For example, if the input has a size of [batch = 10, timesteps = 8, features = 5] and units a value of 3 then input_weights will have a size of [features = 5, 4*units = 3].
Another example, if the input has a size of [batch = 15, timesteps = 8, features = 6] and units a value of 2 then input_weights will have a size of [features = 6, 4*units = 2].

hidden_weights = [units, 4*units].

The size depends on the units parameter of the LSTM layer.
For example, if units has a value of 6 then hidden_weights will have a size of [units = 6, 4*units = 6].
Another example, if units has a value of 3 then hidden_weights will have a size of [units = 3, 4*units = 3].

biases = [4*units]

The size depends on the units parameter of the LSTM layer.
For example, if units has a value of 6, then biases will have a size of [4*units = 6].
Another example, if units has a value of 3, then biases will have a size of [4*units = 3].

Example

All these exemples are snippets PNG, you can drop these Snippet onto the block diagram and get the depicted code added to your VI (Do not forget to install Deep Learning library to run it).