The primary factor in a country’s development is agriculture. To effectively manage plant diseases and increase crop production, it is necessary to accurately identify a disease when it first manifests. Machine learning has been used in a wide range of sectors, and it is now being used in agriculture. The diagnosis of illnesses from images of a plant is one use of machine learning in agricultural research. A variety of machine learning algorithms, such as random forest, clustering, Gaussian models, linear / logistic regression, decision trees, Naive Bayes, K-nearest neighbors, support vector machine and deep learning techniques can be used for disease diagnosis in the agricultural sector. The review paper's objective is to examine the various machine and Deep learning models techniques that are effectively used in the agriculture industry