From f85f9cb48861240687d3133a7b165c90b97c3e04 Mon Sep 17 00:00:00 2001 From: Stevezhangz Date: Mon, 26 Apr 2021 02:16:03 +0800 Subject: [PATCH] Update README.md --- README.md | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index d15212c..82e3f85 100644 --- a/README.md +++ b/README.md @@ -1,8 +1,16 @@ # BERT-pytorch -Repeted by pytorch, without pre-train. -Now, could be trained by mask word2idx. -And, I am foucsing on the conversation training. +Introduction: + +This mechine could be trained by "train_demo.py" + And there are mainly two datasets demo, one is a json file about poem, another is a conversation demo created by myself. + However I don't recommand to use those demo_datas to train, I prefer use formal datasets. + +Funtune method could be found in "Bert_finetune.py", funtune examples mainly include two. + First is the word classify prediction, could be found in ["bert_for_word_classify.py"]{https://codechina.csdn.net/captainAAAjohn/BERT-pytorch/-/blob/main/bert_for_word_classify.py} + Second is the sentences classify prediction, could be found in [" bert_for_sentence_classify.py"]{https://codechina.csdn.net/captainAAAjohn/BERT-pytorch/-/blob/main/bert_for_sentence_classify.py} + +Next, I will enrich the language generation as well as conversation process. # How to use -- GitLab