Nowadays, crop yield prediction is one of the most recent, interesting and challenging tasks due to its dependence on various variable parameters like environmental, weather, soil and climate factors. Machine learning has become one of the important tools for predicting crop yield. This paper presents a machine learning framework for crop yield prediction using crop and weather data. It also compares the performance of potential machine learning methods like regression, decision trees, random forest, support vector machine and gradient boosting to forecast the yield of 80 crops in India for the year 2001 to 2016 using historical data. Furthermore, it has been observed from the results that the root mean square (RMSE) of the random forest method is 9433.7 for the dataset.