Big data has already occupied a lot in the information society. The application of big data to intelligent agriculture is the core development direction for maximizing the utilization of agricultural data information, and the deep learning method can more effectively extract abstract information from big data and convert it into useful knowledge, thus supporting the development of intelligent agriculture from different dimensions. In this paper, a CNN-RNN model is constructed based on cloud computing technology, and the parallel neural network model divided by training set is adopted to design the batch gradient descent algorithm based on deep unsupervised learning and BP algorithm based on Map-Reduce. An experiment verifies the feasibility of deep unsupervised learning neural network based on cloud computing and verifies that the optimize algorithm proposed in this paper can better increase the training efficiency of neural network.
Neural networks , Unsupervised learning
Software And Hardware
• Hardware: Processor: i3 ,i5 RAM: 4GB Hard disk: 16 GB • Software: operating System : Windws2000/XP/7/8/10 Anaconda,jupyter,spyder,flask Frontend :-python Backend:- MYSQL