About this document
Coding LSTM by marcusjoshua.tornea is a document available to read on EtoBox.
This document discusses various LSTM variants trained on a multivariate dataset of Beijing air pollution, focusing on pm2.5 levels and several meteorological features. It details the data preparation process, including imputation for missing values and feature selection, as well as the performance metrics of different LSTM models, including vanilla, stacked, bidirectional, CNN-LSTM, ConvLSTM, and others. The results highlight the challenges of multi-step forecasting and the varying performance of models bas
- Author
- marcusjoshua.tornea
- Language
- EN