Icon

01_​TSA_​with_​LSTM_​Network_​Training

Energy Demand Prediction with LSTM - Training

This workflow trains and applies an LSTM network to predict energy demand using lagged values of a time series as input.

In the Evaluation and Predictions part the trained model is then used for in-sample and out-of-sample forecasting. The forecasted values are compared to the actual values, and the performance of the forecast is reported via scoring metrics and a line plot.




Data Loading Data Cleaning Network Architecture Create Input Vector Evaluating the Predictions Time Series Analysis Demand Prediction LSTM Network - TrainingIn this workflow we build. train, and evaluate an LSTM Network Architecture for demand Prediction in Time SeriesAnalysis. Energyusagedataconvertdate/timeinto Date&Time objectstraining withMSE loss function[1] tensor foroutputLSTM with 100units and Tanhcombine intolistlag 200 valuesout-sample predictionsLSTM.h5in-sample predictions80/20 splittake from toplag 200 valuescombine intolistNode 458[200] tensor for200 lagged inputsoriginal vs. predictedoriginal vs. predictedMissing Value File Reader String to Date&Time Column Filter Keras NetworkLearner Keras Dense Layer Numeric Scorer Keras LSTM Layer Column Aggregator Lag Column out-sample testing Keras NetworkWriter in-sample testing Numeric Scorer Partitioning Lag Column Column Aggregator Missing Value(Apply) Model Writer Keras Input Layer Line Plot (Plotly) Line Plot (Plotly) Timestamp Alignment Data Loading Data Cleaning Network Architecture Create Input Vector Evaluating the Predictions Time Series Analysis Demand Prediction LSTM Network - TrainingIn this workflow we build. train, and evaluate an LSTM Network Architecture for demand Prediction in Time SeriesAnalysis. Energyusagedataconvertdate/timeinto Date&Time objectstraining withMSE loss function[1] tensor foroutputLSTM with 100units and Tanhcombine intolistlag 200 valuesout-sample predictionsLSTM.h5in-sample predictions80/20 splittake from toplag 200 valuescombine intolistNode 458[200] tensor for200 lagged inputsoriginal vs. predictedoriginal vs. predictedMissing Value File Reader String to Date&Time Column Filter Keras NetworkLearner Keras Dense Layer Numeric Scorer Keras LSTM Layer Column Aggregator Lag Column out-sample testing Keras NetworkWriter in-sample testing Numeric Scorer Partitioning Lag Column Column Aggregator Missing Value(Apply) Model Writer Keras Input Layer Line Plot (Plotly) Line Plot (Plotly) Timestamp Alignment

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
On NodePit since: 2020-12-17
Last update: 2024-11-17
Created with KNIME version: v4.2.3
Tags: Time SeriesPredictionEnergy UsageLSTMDeep LearningNeural NetworkKeras

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!