Premonition Net, a multi-timeline transformer network architecture towards strawberry tabletop yield forecasting

George Onoufriou; Marc Hanheide; Georgios Leontidis

doi:10.1016/j.compag.2023.107784

Premonition Net, a multi-timeline transformer network architecture towards strawberry tabletop yield forecasting

George Onoufriou^*, Marc Hanheide, Georgios Leontidis

^*Corresponding author for this work

University of Lincoln

Research output: Contribution to journal › Article › peer-review

13 Citations (Scopus)

7 Downloads (Pure)

Abstract

Yield forecasting is a critical first step necessary for yield optimisation, with important consequences for the broader food supply chain, procurement, price-negotiation, logistics, and supply. However yield forecasting is notoriously difficult, and oft-inaccurate. Premonition Net is a multi-timeline, time sequence ingesting approach towards processing the past, the present, and premonitions of the future. We show how this structure combined with transformers attains critical yield forecasting proficiency towards improving food security, lowering prices, and reducing waste. We find data availability to be a continued difficulty however using our premonition network and our own collected data we attain yield forecasts 3 weeks ahead with a testing set RMSE loss of 0.08 across our latest season.

Original language	English
Article number	107784
Number of pages	8
Journal	Computers and Electronics in Agriculture
Volume	208
Early online date	23 Mar 2023
DOIs	https://doi.org/10.1016/j.compag.2023.107784
Publication status	Published - 1 May 2023

Bibliographical note

This research was supported by the Biotechnology and Biological Sciences Research Council (BBSRC) studentship, United Kingdom [grant numbers 2155898, BB/S507453/1]

Data Availability Statement

The authors do not have permission to share data.

Keywords

deep learning
Strawberry Yield Forecasting
Time series
transformers

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1016/j.compag.2023.107784Licence: CC BY

Leontidis_compag_2023_finalFinal published version, 983 KBLicence: CC BY

Cite this

@article{e3b68f2af45b4421bafbe6253e179937,

title = "Premonition Net, a multi-timeline transformer network architecture towards strawberry tabletop yield forecasting",

abstract = "Yield forecasting is a critical first step necessary for yield optimisation, with important consequences for the broader food supply chain, procurement, price-negotiation, logistics, and supply. However yield forecasting is notoriously difficult, and oft-inaccurate. Premonition Net is a multi-timeline, time sequence ingesting approach towards processing the past, the present, and premonitions of the future. We show how this structure combined with transformers attains critical yield forecasting proficiency towards improving food security, lowering prices, and reducing waste. We find data availability to be a continued difficulty however using our premonition network and our own collected data we attain yield forecasts 3 weeks ahead with a testing set RMSE loss of 0.08 across our latest season.",

keywords = "deep learning, Strawberry Yield Forecasting, Time series, transformers",

author = "George Onoufriou and Marc Hanheide and Georgios Leontidis",

note = "This research was supported by the Biotechnology and Biological Sciences Research Council (BBSRC) studentship, United Kingdom [grant numbers 2155898, BB/S507453/1]",

year = "2023",

month = may,

day = "1",

doi = "10.1016/j.compag.2023.107784",

language = "English",

volume = "208",

journal = "Computers and Electronics in Agriculture",

issn = "0168-1699",

publisher = "Elsevier",

}

TY - JOUR

T1 - Premonition Net, a multi-timeline transformer network architecture towards strawberry tabletop yield forecasting

AU - Onoufriou, George

AU - Hanheide, Marc

AU - Leontidis, Georgios

N1 - This research was supported by the Biotechnology and Biological Sciences Research Council (BBSRC) studentship, United Kingdom [grant numbers 2155898, BB/S507453/1]

PY - 2023/5/1

Y1 - 2023/5/1

N2 - Yield forecasting is a critical first step necessary for yield optimisation, with important consequences for the broader food supply chain, procurement, price-negotiation, logistics, and supply. However yield forecasting is notoriously difficult, and oft-inaccurate. Premonition Net is a multi-timeline, time sequence ingesting approach towards processing the past, the present, and premonitions of the future. We show how this structure combined with transformers attains critical yield forecasting proficiency towards improving food security, lowering prices, and reducing waste. We find data availability to be a continued difficulty however using our premonition network and our own collected data we attain yield forecasts 3 weeks ahead with a testing set RMSE loss of 0.08 across our latest season.

AB - Yield forecasting is a critical first step necessary for yield optimisation, with important consequences for the broader food supply chain, procurement, price-negotiation, logistics, and supply. However yield forecasting is notoriously difficult, and oft-inaccurate. Premonition Net is a multi-timeline, time sequence ingesting approach towards processing the past, the present, and premonitions of the future. We show how this structure combined with transformers attains critical yield forecasting proficiency towards improving food security, lowering prices, and reducing waste. We find data availability to be a continued difficulty however using our premonition network and our own collected data we attain yield forecasts 3 weeks ahead with a testing set RMSE loss of 0.08 across our latest season.

KW - deep learning

KW - Strawberry Yield Forecasting

KW - Time series

KW - transformers

U2 - 10.1016/j.compag.2023.107784

DO - 10.1016/j.compag.2023.107784

M3 - Article

SN - 0168-1699

VL - 208

JO - Computers and Electronics in Agriculture

JF - Computers and Electronics in Agriculture

M1 - 107784

ER -

Premonition Net, a multi-timeline transformer network architecture towards strawberry tabletop yield forecasting

Abstract

Bibliographical note

Data Availability Statement

Keywords

UN SDGs

Access to Document

Fingerprint

Cite this