update document categorisation

app.py

+from flask import Flask
+from flask import request
+from flask_cors import CORS
+import os
+import question as qu
+import db
+import json
+import ast
+from operator import attrgetter
+import pickle
+from sklearn.feature_extraction.text import CountVectorizer
+from sklearn.feature_extraction.text import TfidfTransformer
+
+app  = Flask("__main__")
+CORS(app)
+
+current_path =  os.path.abspath(os.path.join(os.path.dirname(__file__)))
+model = current_path+'\\finalized_model.pickle'
+vec_count = current_path+'\\vec_count.pkl'
+tfidf = current_path+'\\tfidf.pkl'
+input_path = current_path+'\input.txt'
+
+#function to generate questions
+@app.route('/generateQuestions', methods=['GET']) 
+def generateQuestions():
+    text = request.args.get('text')
+    print(text)
+    qList, aList, shortAList = qu.createQuestions(text)
+
+    values = {
+            "questions": qList,
+            "answers": aList,
+            "short": shortAList
+    }
+    return values
+
+#function to insert questions
+@app.route('/insertQuestions', methods=['GET']) 
+def insertQuestions():
+    questions = json.loads(request.args.get('question'))
+    answers = json.loads(request.args.get('answers'))
+    shortAnswers = json.loads(request.args.get('short'))
+    selectedQuestions = json.loads(request.args.get('selected'))
+    name = request.args.get('name')
+
+    new_qs = []
+    new_as = []
+    new_shrt = []
+
+    for index in selectedQuestions:
+        index = int(index)
+        new_qs.append(questions[index])
+        new_as.append(answers[index])
+        new_shrt.append(shortAnswers[index])
+        
+    con = db.connection()
+    cursor = con.cursor()
+    sql = "INSERT INTO questions (qs_name, qs_list,as_list,as_shrt_list) VALUES (%s, %s, %s, %s)" 
+    val = [name,str(new_qs),str(new_as),str(new_shrt)]
+    cursor.execute(sql,val)
+    val = con.commit()
+
+    return str(1)
+
+#function to log
+@app.route('/upload/login', methods=['GET']) 
+def login():
+    uname = request.args.get('uname')
+    pwd = request.args.get('pwd')
+        
+    con = db.connection()
+    cursor = con.cursor()
+    selectData = "select * from users where uname='%s'"%uname #retrieve data from mysql db
+    cursor.execute(selectData)   
+    result = cursor.fetchall()
+    print(result[0])
+    user = False
+
+    if(result[0][1] == pwd):
+        user = True
+
+    values = {
+        "user" : user,
+        "type" : result[0][2]
+    }
+
+    return values
+
+#function to load questions name list
+@app.route('/loadQuestions', methods=['GET']) 
+def loadQuestions():      
+    con = db.connection()
+    cursor = con.cursor()
+    selectData = "select qs_name from questions" #retrieve data from mysql db
+    cursor.execute(selectData)   
+    result = cursor.fetchall()
+    print(result)
+
+    values = {
+        "name" : result,
+    }
+
+    return values
+
+#function to load questions
+@app.route('/getQuestions', methods=['GET']) 
+def getQuestions():  
+    qname = request.args.get('name')   
+    
+    con = db.connection()
+    cursor = con.cursor()
+    selectData = "select * from questions where qs_name='%s'"%qname #retrieve data from mysql db
+    cursor.execute(selectData)   
+    result = cursor.fetchall()
+
+    values = {
+        "questions" : ast.literal_eval(result[0][1]),
+        "answers" : ast.literal_eval(result[0][2]),
+        "short" : ast.literal_eval(result[0][3]),
+    }
+  
+    return values
+
+#function to load questions
+@app.route('/insertResult', methods=['GET']) 
+def insertResult():  
+    uname = request.args.get('uname')  
+    correct = request.args.get('correct_an')   
+    wrong = request.args.get('wrong_an')    
+    qname = request.args.get('qname') 
+    
+    con = db.connection()
+    cursor = con.cursor()
+    sql = "INSERT INTO result (uname,correct,wrong,qname) VALUES (%s, %s, %s,%s)" 
+    val = [uname,correct,wrong,qname]
+    cursor.execute(sql,val)
+    val = con.commit()
+
+    return "1"
+
+#function to load questions
+@app.route('/loadResult', methods=['GET']) 
+def loadResult():  
+    qname = request.args.get('qname')  
+  
+    con = db.connection()
+    cursor = con.cursor()
+    selectData = "select * from result where qname='%s'"%qname #retrieve data from mysql db
+    cursor.execute(selectData)   
+    result = cursor.fetchall()
+    final = []
+    for r in result:
+        marks = (r[1]/(r[1]+r[2])) * 100
+        name = (r[0].split('@'))[0].split('.')
+        st_name = name[0] + " " + name[1] + " : "
+        final.append((st_name,marks))
+ 
+    final.sort(key=lambda y: y[1],reverse=True)
+
+    values = {
+        "result" : final[:5],
+    }
+ 
+    return values
+
+#function to predict cetogory
+@app.route('/cetogoryPrediction', methods=['GET']) 
+def cetogoryPrediction():
+    content = request.args.get('text')  
+    
+    new_doc = [content]
+
+    loaded_model = pickle.load(open(model, 'rb'))
+    loaded_vec = CountVectorizer(vocabulary=pickle.load(open(vec_count, "rb"))) 
+    loaded_tfidf = pickle.load(open(tfidf,"rb"))
+
+    x_new_counts = loaded_vec.transform(new_doc)
+    x_new_tfidf = loaded_tfidf.transform(x_new_counts)
+    predict = loaded_model.predict(x_new_tfidf)
+    values = {
+        "result" : predict,
+    }
+
+    return values
+
+if __name__ == '__main__':
+   loadResult()
\ No newline at end of file

categorize.py

+# from flask import Flask
+# from flask import request
+# from flask_cors import CORS
+import os,sys
+from operator import attrgetter
+import pickle
+from sklearn.feature_extraction.text import CountVectorizer
+#from sklearn.feature_extraction.text import TfidfTransformer
+import shutil
+
+current_path =  os.path.abspath(os.path.join(os.path.dirname(__file__)))
+model = current_path+'\\finalized_model.pickle'
+vec_count = current_path+'\\vec_count.pkl'
+tfidf = current_path+'\\tfidf.pkl'
+input_path = current_path+'\input.txt'
+doc_path = current_path+'\Docs'
+notice_path = current_path+'\Docs\\notices'
+grade10_path = current_path+'\Docs\\grade10'
+grade11_path = current_path+'\Docs\\grade11'
+
+
+path = str(sys.argv[1])
+content = str(sys.argv[2])
+
+new_doc = [content]
+
+loaded_model = pickle.load(open(model, 'rb'))
+loaded_vec = CountVectorizer(vocabulary=pickle.load(open(vec_count, "rb"))) 
+loaded_tfidf = pickle.load(open(tfidf,"rb"))
+
+x_new_counts = loaded_vec.transform(new_doc)
+x_new_tfidf = loaded_tfidf.transform(x_new_counts)
+predict = loaded_model.predict(x_new_tfidf)
+print(predict[0])
+
+if(predict[0] == "Notice"):
+    shutil.move(doc_path+"\\"+path, notice_path+"\\"+path)
+elif(predict[0] == "grade10"):
+    shutil.move(doc_path+"\\"+path, grade10_path+"\\"+path)
+elif(predict[0] == "grade11"):
+    shutil.move(doc_path+"\\"+path, grade11_path+"\\"+path)
+print("success")
\ No newline at end of file