yongzhuo / Keras-TextClassification

中文长文本分类、短句子分类、多标签分类、两句子相似度(Chinese Text Classification of Keras NLP, multi-label classify, or sentence classify, long or short),字词句向量嵌入层(embeddings)和网络层(graph)构建基类,FastText,TextCNN,CharCNN,TextRNN, RCNN, DCNN, DPCNN, VDCNN, CRNN, Bert, Xlnet, Albert, Attention, DeepMoji, HAN, 胶囊网络-CapsuleNet, Transformer-encode, Seq2seq, SWEM, LEAM, TextGCN
https://blog.csdn.net/rensihui
MIT License
1.77k stars 404 forks source link

训练好的模型一般如何部署呢? #69

Closed luoyudong593 closed 3 years ago

luoyudong593 commented 3 years ago

已经按照教程训练了一个分类模型,想了解一般以什么方式部署到线上呢(接口、脚本等),望大佬指点一下,有相关的文档参考也好,谢谢~

yongzhuo commented 3 years ago

1.简单的可以直接写flask接口调用predict; 2.也可以转为.pb文件等用tf-serving

luoyudong593 commented 3 years ago

好的,感谢~