The system mainly contains three aspects: (1) training of the model under the training set (2) testing of the model under the test set (3) display of the front-end page
First use the existing model on github
(1) The first step is to train the model under the training set and save the model with the best result. You can refer to this blog post https://blog.csdn.net/liuxiaodong400/article/details/83421164
saver = tf.train.Saver()
save_path = saver.save(sess, self.config.modelpath, global_step = epoch)
(2) The second step is to write the test interface, each time a sentence is passed in and the prediction result of the sentence is output
(3) The third step is the front end. The front end uses the Flask framework, which is the connection between the front end and the test interface