🚀 机器学习7.9

机器学习/结课项目/疫情期间网民情绪识别/covid19/main.py

@@ -1,50 +1,4 @@
-# coding=utf-8
-import pandas as pd
-import jieba
-import numpy as np
-from snownlp import SnowNLP
-from snownlp import sentiment
-import pandas as pd
-import numpy as np
-import jieba
-
-def GetScore(line, score, key):
-    segs = jieba.lcut(line)
-    score_list = [score[key.index(x)] for x in segs if (x in key)]
-    # print(segs)
-    return sum(score_list)
-
-
-BosonNlpScore = pd.read_csv("bosonnlp//BosonNLP_sentiment_score.txt", sep=" ", names=['key', 'score'])
-key = BosonNlpScore['key'].values.tolist()
-score = BosonNlpScore['score'].values.tolist()
-
-sentiment.train('train&test/neg.txt','train&test/pos.txt')
-sentiment.save('/Users/alexfan/anaconda3/lib/python3.7/site-packages/snownlp/sentiment/sentiment.marshal')
-
-'''
-获取test.csv中所有的舆论
-'''
-test = pd.read_csv("train&test/test.csv")
-# print(test)
-test_list = test.values.tolist()
-test_str = []
-for i in test_list:
-    test_str.append(i[0])
-
-'''
-获取test_label.csv的最终结果标识
-'''
-test_label = pd.read_csv("train&test/test_label.csv")
-test_label = test_label["情感倾向"].values.tolist()
-
-result = []
-for comment in test_str[0:100]:
-    #comment_predict = int(GetScore(comment, score, key))
-    # if(comment_predict>0):comment_predict=1
-    # if(comment_predict<0):comment_predict=-1
-    comment = SnowNLP(comment)
-    result.append(comment.sentiments)
-
-print(test_label)
-print(result)
+from splitNegPos import split
+if __name__ == '__main__':
+    splitToTxt = split()
+    splitToTxt.startSplit(1000,1000,1000) #设置读取的训练集中pos、neg、mid样本的大小

机器学习/结课项目/疫情期间网民情绪识别/covid19/splitNegPos.py

@@ -1,34 +1,29 @@
 # coding=utf-8
 import pandas as pd
-import jieba
-import numpy as np
 
-all = pd.read_csv("train&test/train.csv")
-
-pos = all.loc[all['情感倾向']==1]
-pos = pos['微博中文内容'].values.tolist()
-
-neg = all.loc[all['情感倾向']==-1]
-neg = neg['微博中文内容'].values.tolist()
-
-mid = all.loc[all['情感倾向']==0]
-mid = mid['微博中文内容'].values.tolist()
-
-file = open("train&test/pos.txt",'w');
-for comment in pos[0:1000]:
-    file.write(comment)
-    file.write('\n')
-file.close()
-
-file = open("train&test/neg.txt",'w');
-for comment in neg[0:1000]:
-    file.write(comment)
-    file.write('\n')
-file.close()
-
-file = open("train&test/mid.txt",'w');
-for comment in mid:
-    file.write(comment)
-    file.write('\n')
-file.close()
+class split:
+    def __init__(self):
+        self.all = pd.read_csv("train&test/train.csv")
+        self.pos = (self.all.loc[self.all['情感倾向']==1])['微博中文内容'].values.tolist()
+        self.neg = (self.all.loc[self.all['情感倾向']==-1])['微博中文内容'].values.tolist()
+        self.mid = (self.all.loc[self.all['情感倾向']==0])['微博中文内容'].values.tolist()
+
+    def startSplit(self,posnum,negnum,midnum):
+        file = open("train&test/pos.txt", 'w');
+        for comment in self.pos[0:posnum]:
+            file.write(comment)
+            file.write('\n')
+        file.close()
+
+        file = open("train&test/neg.txt",'w');
+        for comment in self.neg[0:negnum]:
+            file.write(comment)
+            file.write('\n')
+        file.close()
+
+        file = open("train&test/mid.txt",'w');
+        for comment in self.mid[0:midnum]:
+            file.write(comment)
+            file.write('\n')
+        file.close()
 

机器学习/结课项目/疫情期间网民情绪识别/covid19/test.py

@@ -1,5 +1,4 @@
 # coding=utf-8
-import 
 def main():
     print("Welcome Back,Alex Fan")
 

机器学习/结课项目/疫情期间网民情绪识别/covid19/train.py

@@ -0,0 +1,39 @@
+# coding=utf-8
+import pandas as pd
+import jieba
+from snownlp import SnowNLP
+from snownlp import sentiment
+
+def GetScore(line, score, key):
+    segs = jieba.lcut(line)
+    score_list = [score[key.index(x)] for x in segs if (x in key)]
+    return sum(score_list)
+
+def bosonNlp():
+    BosonNlpScore = pd.read_csv("bosonnlp//BosonNLP_sentiment_score.txt", sep=" ", names=['key', 'score'])
+    key = BosonNlpScore['key'].values.tolist()
+    score = BosonNlpScore['score'].values.tolist()
+    sentiment.train('train&test/neg.txt','train&test/pos.txt')
+    sentiment.save('/Users/alexfan/anaconda3/lib/python3.7/site-packages/snownlp/sentiment/sentiment.marshal')
+
+def test():
+    '''
+    获取test.csv中所有的舆论
+    '''
+    test = pd.read_csv("train&test/test.csv")
+    # print(test)
+    test_list = test.values.tolist()
+    test_str = []
+    for i in test_list:
+        test_str.append(i[0])
+    '''
+    获取test_label.csv的最终结果标识
+    '''
+    test_label = (pd.read_csv("train&test/test_label.csv"))["情感倾向"].values.tolist()
+    result = []
+    for comment in test_str[0:100]:
+        #comment_predict = int(GetScore(comment, score, key))
+        # if(comment_predict>0):comment_predict=1
+        # if(comment_predict<0):comment_predict=-1
+        comment = SnowNLP(comment)
+        result.append(comment.sentiments)