LSTM for Time-series: Chaos in the AI Industry

Author(s): Michelangiolo Mazzeschi

Originally published on Towards AI.

Machine Learning

LSTM is one of the most common neural networks for forecasting time-series. Unfortunately, this kind of neural network is complex (data preparation in particular), not only because time-series data requires a specific structure to be fit into an AI (such as the sliding windows technique), but because of the key steps into creating an LSTM (data preparation, reshaping…) articulate into a variance of paths that depends entirely on the habits of the engineer.

When I scavenge the Internet in search of code to study, I always find it difficult to incorporate the code of another programmer in my project. Specifically for… Read the full blog for free on Medium.

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI