Implementation:Sktime Pytorch forecasting sLSTMNetwork

Knowledge Sources	Sktime_Pytorch_forecasting
Domains	Time_Series, Forecasting, Deep_Learning
Last Updated	2026-02-08 08:00 GMT

Overview

sLSTMNetwork is a complete stabilized LSTM network combining stacked sLSTM layers with a fully connected output layer for prediction.

Description

sLSTMNetwork extends nn.Module and composes an sLSTMLayer (stacked stabilized LSTM layers) with a final nn.Linear fully connected layer. The sLSTM layers process the input sequence recurrently with exponential gating, and the FC layer maps the last time step's hidden output to the desired output dimension. This provides an end-to-end trainable network architecture based on the sLSTM cell with stabilized gating.

Usage

Use sLSTMNetwork as a standalone recurrent forecasting model when you want a complete stabilized LSTM-based architecture. It is suitable for time series prediction tasks where training stability from exponential gating is beneficial.

Code Reference

Source Location

Repository: Sktime_Pytorch_forecasting
File: pytorch_forecasting/layers/_recurrent/_slstm/network.py
Lines: 1-91

Signature

class sLSTMNetwork(nn.Module):
    def __init__(
        self,
        input_size,
        hidden_size,
        num_layers,
        output_size,
        dropout=0.0,
        use_layer_norm=True,
    ):
    def forward(self, x, h=None, c=None):
    def init_hidden(self, batch_size, device=None):

Import

from pytorch_forecasting.layers._recurrent._slstm.network import sLSTMNetwork

I/O Contract

Inputs

init

Name	Type	Required	Description
input_size	int	Yes	Number of features in the input.
hidden_size	int	Yes	Number of features in the hidden state of each sLSTM layer.
num_layers	int	Yes	Number of stacked sLSTM layers in the network.
output_size	int	Yes	Number of features in the output prediction.
dropout	float	No	Dropout probability for the input of each sLSTM layer. Defaults to 0.0.
use_layer_norm	bool	No	Whether to use layer normalization in each sLSTM layer. Defaults to True.

forward

Name	Type	Required	Description
x	torch.Tensor	Yes	Input sequence tensor of shape (seq_len, batch_size, input_size).
h	list of torch.Tensor or None	No	Initial hidden states for each layer. If None, initialized to zeros.
c	list of torch.Tensor or None	No	Initial cell states for each layer. If None, initialized to zeros.

Outputs

forward

Name	Type	Description
output	torch.Tensor	Final output tensor from the fully connected layer, produced from the last time step. Shape is (batch_size, output_size).
(h, c)	tuple of lists	Final hidden and cell states for each layer.

Usage Examples

import torch
from pytorch_forecasting.layers._recurrent._slstm.network import sLSTMNetwork

model = sLSTMNetwork(
    input_size=32,
    hidden_size=64,
    num_layers=2,
    output_size=10,
    dropout=0.1,
    use_layer_norm=True,
)

seq_len, batch_size = 20, 16
x = torch.randn(seq_len, batch_size, 32)
output, (h, c) = model(x)
# output shape: (16, 10)  -- last time step projected to output_size

Related Pages

Page Connections

Double-click a node to navigate. Hold to expand connections.

Principle

Implementation

Heuristic

Environment