OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Lstm Fcn

Lstm Fcn is a Python codebase that implements Keras LSTM-FCN models for time-series classification, combining temporal convolutional layers with Long Short-Term Memory recurrent networks. It includes an attention-based ALSTM-FCN variant and documents a separate multivariate MLSTM-FCN model for data with multiple variables.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Lstm Fcn — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use the repository when you need an implementation and training workflow for LSTM-FCN time-series classification, especially when working with the UCR datasets or comparing the documented attention variant.

Who it is for

  • Researchers and developers building time-series classification models with Python and Keras.
  • Practitioners evaluating LSTM-FCN and attention-based variants on UCR benchmark datasets.
  • Users exploring multivariate time-series classification through the separately linked MLSTM-FCN project.

Documented uses

  • Build an LSTM-FCN or ALSTM-FCN Keras model using the repository's model functions.
  • Train and evaluate classification models across the listed UCR datasets.
  • Compare model configurations that vary the number of LSTM or attention cells.
  • Use provided weight files for evaluation and later analysis.
Before you choose

Requirements & limits

  • The documented stack is Keras with the TensorFlow backend; Theano and CNTK are unsupported.
  • The README describes a dedicated multivariate model in a separate MLSTM-FCN repository because standard LSTM-FCN may not perform optimally when applied directly to multivariate datasets.
  • The training script can overwrite existing weight files when training is run.
  • The README describes evaluation using provided code and weights, with visualization parameters tied to dataset ID, model function, and cell count.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up the repository and UCR datasets

Prepare the codebase and dataset files in the locations described by the repository instructions.

Before you start

  • The LSTM-FCN repository
  • Python and pip
  • The UCR time-series dataset archive

Steps

  1. Download the repository and install its required libraries with the command below.

    Documented command
    pip install -r requirements.txt

  2. Extract the UCR dataset archive into a folder.

  3. Copy extract_all_datasets.py into that folder and run it to create the _data folder.

  4. Move the extracted dataset files into the repository's Data directory.

Expected outcome

The requirements are installed and the extracted dataset files are placed in the repository's Data directory.

Guide limitations

  • The documented model development uses Keras with the TensorFlow backend; Theano and CNTK are not supported.
  • This setup guide does not cover running training or evaluation.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
6 November 2026
Original website title
titu1994/LSTM-FCN · Official homepage title · retained in its source language
How we review sourcesSuggest a correction