Result and conclusion

Here we have used three methods:

1. Random forecast
2. Support vector regression
3. Deep neural network

We have used the parameters of mean absolute error (MAE) and mean square error (MSE) and compared different models. We can see from the given bar graph that a good result is shown by deep neural network and also the MAE and MSE have lowest value in the represented in model.

Now we will see the graph of the neural network. In this graph, we can see directly that the MSE and MAE are decreasing and the accuracy is increasing with the increasing number of crops. Now we will see the CNN case which is used for counting the images of crops (Figure 5.15).

In CNN, you can see that with the increasing number of crops MSE is decreasing when the results between test data and train data were compared. According to the performance of model, we can say that the model is good. So, we can conclude that deep neural network process performs best for accuracy of crop yield prediction problem.