The objective of this study is thus to find suitable time series machine learning models of forecasting for cereals, vegetables, fruits, and wheat production in Turkey. Turkey, situated at the crossroads of Asia and Europe, is among the larger countries of the region with a population of over 85 million in terms of territory. Agriculture employs about a quarter of the workforce and generates most of the income and employment in rural areas. We examined five machine learning algorithms, including autoregressive integrated moving average (ARIMA), Prophet, elastic-net regularized generalized linear (GLMNet), random forest, and eXtreme Gradient Boost (XGBoost) using R programming. The performance of the algorithms was evaluated using the mean absolute percent error (MAPE). As a result, the algorithms that give the best estimates based on the MAPE error metrics were found as ARIMA and GLMNet.