Spring Boot integrates tensorflow to implement image detection

1. What is TensorFlow?

The origin of the name TensorFlow is the flow of tensors in a computational graph, as shown in the figure. In addition to automatically helping you find gradients, it also provides a variety of common operations (OP, i.e., nodes of the graph), common loss functions, and optimization algorithms.

• TensorFlow is an open-source software library for high-performance numerical computing. With its flexible architecture, users can easily deploy computing jobs to multiple platforms (CPUs, GPUs, TPUs) and devices (desktops, server clusters, mobile devices, edge devices, etc.).https://www.tensorflow.org/tutorials/?hl=zh-cnwww.tensorflow.org/tutorials/?hl=zh-cn(opens new window)
• TensorFlow is an open-source machine learning library for research and production. TensorFlow provides a variety of APIs for beginners and experts to develop in desktop, mobile, web, and cloud environments.
• TensorFlow is calculated using data flow graphs, so first we have to create a data flow graph, and then put our data (data in the form of tensors) in the data flow graph for calculation. Nodes represent mathematical operations in a graph, and edges represent arrays of multidimensional data that are interconnected between nodes, i.e., tensors. When training a model, tensor will continuously flow from one node in the data flow graph to another, which is where the name TensorFlow comes from. Tensor: There are many kinds of tensors. A tensor of the order zero is a pure quantity or scalar, which is a numeric value. For example, [1], the first-order tensor is a vector, such as a one-dimensional [1, 2, 3], a second-order tensor is a matrix, such as a two-dimensional [[1, 2, 3],[4, 5, 6],[7, 8, 9]], and so on, and a third-order three-dimensional … The computational process of a tensor flowing from one end of a flow diagram to the other. It vividly describes the flow, transmission, analysis, and processing patterns of complex data structures in artificial neural networks.

In machine learning, numerical values are usually composed of four types: (1) Scalar: that is, a numerical value, which is the smallest unit of calculation, such as “1” or “3.2”, etc. (2) Vector: A one-dimensional array composed of some scalar quantities…