YanZongzheng / Crop-Yield-Prediction-Comparison-using-ML-DL-Techniques

In this project, we compare and predict the yield of five crops (wheat, barley, jowar, rapeseed & mustard, and bajra) in Rajasthan (district-wise) using three machine learning techniques: random forest, lasso regression and SVM, and two deep learning techniques: gradient descent and RNN LSTM. To apply the models to our data, we divided it into training and testing datasets. Each model is tested twice: once with only "area" and "production" in mind, and then again with additional factors (rainfall and soil type) in mind to predict crop yield. To find the model that most accurately predicts the yield, R2 score, Root Mean Squared Error (RMSE) and Mean Average Error (MAE) are calculated for each model.

Crop-Yield-Prediction-Comparison-using-ML-DL-Techniques

In this project, we compare and predict the yield of five crops (wheat, barley, jowar, rapeseed & mustard, and bajra) in Rajasthan (district-wise) using three machine learning techniques: random forest, lasso regression and SVM, and two deep learning techniques: gradient descent and RNN LSTM. To apply the models to our data, we divided it into training and testing datasets. Each model is tested twice: once with only "area" and "production" in mind, and then again with additional factors (rainfall and soil type) in mind to predict crop yield. To find the model that most accurately predicts the yield, R2 score, Root Mean Squared Error (RMSE) and Mean Average Error (MAE) are calculated for each model.