Crop Yield Prediction Using Deep Neural Networks

Khaki, Saeed; Wang, Lizhi

doi:10.3389/fpls.2019.00621

Cited by 471 publications

(293 citation statements)

References 36 publications

Supporting

Mentioning

244

Contrasting

Order By: Relevance

“…Farmers can utilize the yield prediction to make knowledgeable management and financial decisions. Performances of new hybrids can be predicted in new and untested locations (Khaki and Wang, 2019). However, successful crop yield prediction is very difficult due to many complex factors.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A CNN-RNN Framework for Crop Yield Prediction

2020

Self Cite

View full text Add to dashboard Cite

Crop yield prediction is extremely challenging due to its dependence on multiple factors such as crop genotype, environmental factors, management practices, and their interactions. This paper presents a deep learning framework using convolutional neural networks (CNNs) and recurrent neural networks (RNNs) for crop yield prediction based on environmental data and management practices. The proposed CNN-RNN model, along with other popular methods such as random forest (RF), deep fully connected neural networks (DFNN), and LASSO, was used to forecast corn and soybean yield across the entire Corn Belt (including 13 states) in the United States for years 2016, 2017, and 2018 using historical data. The new model achieved a root-mean-square-error (RMSE) 9% and 8% of their respective average yields, substantially outperforming all other methods that were tested. The CNN-RNN has three salient features that make it a potentially useful method for other crop yield prediction studies. (1) The CNN-RNN model was designed to capture the time dependencies of environmental factors and the genetic improvement of seeds over time without having their genotype information.(2) The model demonstrated the capability to generalize the yield prediction to untested environments without significant drop in the prediction accuracy. (3) Coupled with the backpropagation method, the model could reveal the extent to which weather conditions, accuracy of weather predictions, soil conditions, and management practices were able to explain the variation in the crop yields.

show abstract

Section: Introductionmentioning

confidence: 99%

“…More recently, deep learning methods have been applied for the crop yield prediction. Khaki and Wang (2019) designed a deep neural network model to predict corn yield across 2,247 locations between 2008 and 2016. Their model was found to outperform other methods such as Lasso, shallow neural networks, and regression tree.…”

Section: Introductionmentioning

confidence: 99%

A CNN-RNN Framework for Crop Yield Prediction

2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…The latter have enjoyed wide applications in various ecological classification problems and predictive modeling (Rumpf et al 2010, Shekoofa et al 2014, Crane-Droesch 2018, Karimzadeh and Olafsson 2019, Pham and Olafsson 2019a, 2019b because of their adeptness to deal with nonlinear relationships, high-order interactions and non-normal data (De'ath and Fabricius 2000). Such methods include regularized regressions (Hoerl and Kennard 1970, Tibshirani 1996, Zou and Hastie 2005, tree-based models (Shekoofa et al 2014), Support Vector Machines (Basak et al 2007, Karimi et al 2008, Neural Networks (Liu et al 2001, Crane-Droesch 2018, Khaki and Khalilzadeh 2019, Khaki and Wang 2019 and others.…”

Section: Introductionmentioning

confidence: 99%

Maize yield and nitrate loss prediction with machine learning algorithms

Shahhosseini

Martinez-Feria

et al. 2019

Environ. Res. Lett.

142

View full text Add to dashboard Cite

Pre-growing season prediction of crop production outcomes such as grain yields and nitrogen (N) losses can provide insights to farmers and agronomists to make decisions. Simulation crop models can assist in scenario planning, but their use is limited because of data requirements and long runtimes. Thus, there is a need for more computationally expedient approaches to scale up predictions. We evaluated the potential of four machine learning (ML) algorithms (LASSO Regression, Ridge Regression, random forests, Extreme Gradient Boosting, and their ensembles) as meta-models for a cropping systems simulator (APSIM) to inform future decision support tool development. We asked:(1) How well do ML meta-models predict maize yield and N losses using pre-season information? (2) How many data are needed to train ML algorithms to achieve acceptable predictions? (3) Which input data variables are most important for accurate prediction? And (4) do ensembles of ML meta-models improve prediction? The simulated dataset included more than three million data including genotype, environment and management scenarios. XGBoost was the most accurate ML model in predicting yields with a relative mean square error (RRMSE) of 13.5%, and Random forests most accurately predicted N loss at planting time, with a RRMSE of 54%. ML meta-models reasonably reproduced simulated maize yields using the information available at planting, but not N loss. They also differed in their sensitivities to the size of the training dataset. Across all ML models, yield prediction error decreased by 10%-40% as the training dataset increased from 0.5 to 1.8 million data points, whereas N loss prediction error showed no consistent pattern. ML models also differed in their sensitivities to input variables (weather, soil properties, management, initial conditions), thus depending on the data availability researchers may use a different ML model. Modest prediction improvements resulted from ML ensembles. These results can help accelerate progress in coupling simulation models and ML toward developing dynamic decision support tools for pre-season management.

show abstract

“…In addition to epistasis, another example of interactions relating to breeding is the interaction between genes and environments, which is prominent particularly in plants. Whereas mixed effect models [41,42] and physiological model-based approaches [43,44] are often used to predict gene-by-environment interactions, several authors proposed neural networks that used both environmental covariates (i.e., environment identifiers or climate covariates) and genotype information (i.e.., genotype identifiers, marker genotypes, or decomposed genetic relationship matrices) as inputs [45][46][47]. In ref.…”

Section: Beyond Genomic Predictionmentioning

confidence: 99%

“…In ref. [47], a narrow and deep network (21 hidden layers and 50 units per layer) was used to predict the yield of maize. Although such a model structure is not supported from the results of the present study, the authors adopted the residual learning framework using shortcut connections, which realizes extremely deep architectures [48].…”

Section: Beyond Genomic Predictionmentioning

confidence: 99%

Predicting genotypic values associated with gene interactions using neural networks: A simulation study for investigating factors affecting prediction accuracy

Onogi

2019

Preprint

View full text Add to dashboard Cite

Genomic prediction has been applied to various species of plants and livestock to enhance breeding efficacy. Neural networks including deep neural networks are attractive candidates to predict phenotypic values. However, the properties of neural networks in predicting non-additive effects have not been clarified. In this simulation study, factors affecting the prediction of genetic values associated with gene interactions (i.e., epistasis) were investigated using multilayer perceptron. The results suggested that (1) redundant markers should be pruned, although markers in LD with QTLs are less harmful, (2) predicting epistatic genetic values with neural networks in real populations would be infeasible using training populations of 1000 samples, (3) neural networks with two or fewer hidden layers and a sufficient number of units per hidden layer would be useful, particularly when a certain number of interactions is involved, and (4) neural networks have greater capability to predict epistatic genetic values than random forests, although neural networks are more sensitive to training population size and the level of epistatic genetic variance. These lessons also would be applicable to other regression problems in which interactions between explanatory variables are expected, e.g., gene-by-environment interactions.

show abstract

Crop Yield Prediction Using Deep Neural Networks

Cited by 471 publications

References 36 publications

A CNN-RNN Framework for Crop Yield Prediction

A CNN-RNN Framework for Crop Yield Prediction

Maize yield and nitrate loss prediction with machine learning algorithms

Predicting genotypic values associated with gene interactions using neural networks: A simulation study for investigating factors affecting prediction accuracy

Contact Info

Product

Resources

About