third_codes/cppjieba
1
0
Fork 0
mirror of https://github.com/yanyiwu/cppjieba.git synced 2025-07-18 00:00:12 +08:00
Code Issues Packages Projects Releases Wiki Activity

use map/set instead of unordered_map/unordered_set to make result stable

Browse Source
This commit is contained in:
yanyiwu 2016-05-03 21:24:40 +08:00
parent 39316114c5
commit f253db0133
2 changed files with 16 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
include/cppjieba/TextRankExtractor.hpp
View File

@ -12,15 +12,18 @@ namespace cppjieba {
public: public:
typedef struct _Word {string word;vector<size_t> offsets;double weight;} Word; // struct Word typedef struct _Word {string word;vector<size_t> offsets;double weight;} Word; // struct Word
private: private:
typedef std::unordered_map<string,Word> WordMap; typedef std::map<string,Word> WordMap;
class WordGraph{ class WordGraph{
private: private:
typedef double Score; typedef double Score;
typedef string Node; typedef string Node;
typedef std::unordered_set<Node> NodeSet; typedef std::set<Node> NodeSet;
typedef std::unordered_map<Node,double> Edges;
typedef std::unordered_map<Node,Edges> Graph; typedef std::map<Node,double> Edges;
typedef std::map<Node,Edges> Graph;
//typedef std::unordered_map<Node,double> Edges;
//typedef std::unordered_map<Node,Edges> Graph;
double d; double d;
Graph graph; Graph graph;

14
test/unittest/textrank_test.cpp
View File

@ -24,14 +24,16 @@ TEST(TextRankExtractorTest, Test1) {
vector<pair<string, double> > words; vector<pair<string, double> > words;
Extractor.Extract(s, words, topN); Extractor.Extract(s, words, topN);
res << words; res << words;
ASSERT_EQ(res, "[\"世界:1\", \"你好:0.514286\"]"); //ASSERT_EQ(res, "[\"世界:1\", \"你好:0.514286\"]");
ASSERT_EQ(res, "[\"\xE4\xB8\x96\xE7\x95\x8C:1\", \"\xE4\xBD\xA0\xE5\xA5\xBD:0.519787\"]");
} }
{ {
vector<TextRankExtractor::Word> words; vector<TextRankExtractor::Word> words;
Extractor.Extract(s, words, topN); Extractor.Extract(s, words, topN);
res << words; res << words;
ASSERT_EQ(res, "[\"世界|[\"6\", \"12\"]|1\", \"你好|[\"0\"]|0.514286\"]"); //ASSERT_EQ(res, "[\"世界|[\"6\", \"12\"]|1\", \"你好|[\"0\"]|0.514286\"]");
ASSERT_EQ(res, "[\"\xE4\xB8\x96\xE7\x95\x8C|[\"6\", \"12\"]|1\", \"\xE4\xBD\xA0\xE5\xA5\xBD|[\"0\"]|0.519787\"]");
} }
} }
@ -42,7 +44,7 @@ TEST(TextRankExtractorTest, Test1) {
size_t topN = 5; size_t topN = 5;
Extractor.Extract(s, wordweights, topN); Extractor.Extract(s, wordweights, topN);
res << wordweights; res << wordweights;
ASSERT_EQ(res, "[\"专业|[\"36\"]|1\", \"CEO|[\"94\"]|0.94764\", \"当上|[\"87\"]|0.79271\", \"手扶拖拉机|[\"21\"]|0.789347\", \"走上|[\"100\"]|0.768261\"]"); ASSERT_EQ(res, "[\"\xE4\xB8\x93\xE4\xB8\x9A|[\"36\"]|1\", \"CEO|[\"94\"]|0.95375\", \"\xE6\x89\x8B\xE6\x89\xB6\xE6\x8B\x96\xE6\x8B\x89\xE6\x9C\xBA|[\"21\"]|0.801701\", \"\xE5\xBD\x93\xE4\xB8\x8A|[\"87\"]|0.798968\", \"\xE8\xB5\xB0\xE4\xB8\x8A|[\"100\"]|0.775505\"]");
// ASSERT_EQ(res, "[\"\xE4\xB8\x93\xE4\xB8\x9A|[\"36\"]|1\", \"CEO|[\"94\"]|0.953149\", \"\xE6\x89\x8B\xE6\x89\xB6\xE6\x8B\x96\xE6\x8B\x89\xE6\x9C\xBA|[\"21\"]|0.794203\", \"\xE5\xBD\x93\xE4\xB8\x8A|[\"87\"]|0.78716\", \"\xE8\xB5\xB0\xE4\xB8\x8A|[\"100\"]|0.767636\"]"); // ASSERT_EQ(res, "[\"\xE4\xB8\x93\xE4\xB8\x9A|[\"36\"]|1\", \"CEO|[\"94\"]|0.953149\", \"\xE6\x89\x8B\xE6\x89\xB6\xE6\x8B\x96\xE6\x8B\x89\xE6\x9C\xBA|[\"21\"]|0.794203\", \"\xE5\xBD\x93\xE4\xB8\x8A|[\"87\"]|0.78716\", \"\xE8\xB5\xB0\xE4\xB8\x8A|[\"100\"]|0.767636\"]");
} }
@ -53,7 +55,8 @@ TEST(TextRankExtractorTest, Test1) {
size_t topN = 5; size_t topN = 5;
Extractor.Extract(s, wordweights, topN); Extractor.Extract(s, wordweights, topN);
res << wordweights; res << wordweights;
ASSERT_EQ(res, "[\"iPhone6|[\"6\"]|1\", \"\xE4\xB8\x80\xE9\x83\xA8|[\"0\"]|0.996126\"]"); ASSERT_EQ(res, "[\"\xE4\xB8\x80\xE9\x83\xA8|[\"0\"]|1\", \"iPhone6|[\"6\"]|0.996126\"]");
//ASSERT_EQ(res, "[\"iPhone6|[\"6\"]|1\", \"\xE4\xB8\x80\xE9\x83\xA8|[\"0\"]|0.996126\"]");
} }
} }
@ -82,6 +85,7 @@ TEST(TextRankExtractorTest, Test2) {
size_t topN = 5; size_t topN = 5;
Extractor.Extract(s, wordweights, topN); Extractor.Extract(s, wordweights, topN);
res << wordweights; res << wordweights;
ASSERT_EQ(res, "[\"iPhone6|[\"6\"]|1\", \"\xE4\xB8\x80\xE9\x83\xA8|[\"0\"]|0.996126\"]"); //ASSERT_EQ(res, "[\"iPhone6|[\"6\"]|1\", \"\xE4\xB8\x80\xE9\x83\xA8|[\"0\"]|0.996126\"]");
ASSERT_EQ(res, "[\"\xE4\xB8\x80\xE9\x83\xA8|[\"0\"]|1\", \"iPhone6|[\"6\"]|0.996126\"]");
} }
} }