From a05e753d11d9c36983ec4e58c421dbcf86d1dcd4 Mon Sep 17 00:00:00 2001
From: Xian Shi <40013335+R1ckShi@users.noreply.github.com>
Date: 星期二, 17 十月 2023 16:47:27 +0800
Subject: [PATCH] Merge branch 'main' into dev_onnx
---
funasr/runtime/onnxruntime/src/tokenizer.h | 16 +++++++++++++++-
1 files changed, 15 insertions(+), 1 deletions(-)
diff --git a/funasr/runtime/onnxruntime/src/tokenizer.h b/funasr/runtime/onnxruntime/src/tokenizer.h
index 4ddd359..149161b 100644
--- a/funasr/runtime/onnxruntime/src/tokenizer.h
+++ b/funasr/runtime/onnxruntime/src/tokenizer.h
@@ -5,13 +5,21 @@
#pragma once
#include <yaml-cpp/yaml.h>
+#include "cppjieba/DictTrie.hpp"
+#include "cppjieba/HMMModel.hpp"
+#include "cppjieba/Jieba.hpp"
+namespace funasr {
class CTokenizer {
private:
bool m_ready = false;
vector<string> m_id2token,m_id2punc;
map<string, int> m_token2id,m_punc2id;
+
+ cppjieba::DictTrie *jieba_dict_trie_;
+ cppjieba::HMMModel *jieba_model_;
+ cppjieba::Jieba jieba_processor_;
public:
@@ -27,7 +35,13 @@
string Id2Punc(int n_punc_id);
vector<int> Punc2Ids(vector<string> input);
vector<string> SplitChineseString(const string& str_info);
+ vector<string> SplitChineseJieba(const string& str_info);
void StrSplit(const string& str, const char split, vector<string>& res);
void Tokenize(const char* str_info, vector<string>& str_out, vector<int>& id_out);
-
+ bool IsPunc(string& Punc);
+ bool seg_jieba = false;
+ void SetJiebaRes(cppjieba::DictTrie *dict, cppjieba::HMMModel *hmm);
+ void JiebaInit(std::string punc_config);
};
+
+} // namespace funasr
--
Gitblit v1.9.1