新增生成赋能题组功能

工具v4/database_tools_2.py

@@ -184,7 +184,7 @@ def treat_dict(database): #对整个题库字典中的内容部分进行预处
     # mycursor.execute("SELECT ID,SAME_ID FROM same;")
     # same_list = [(ret[0],ret[1]) for ret in mycursor.fetchall()]
     for id in p_dict:
-        treated_dict[id] = {}
+        # treated_dict[id] = {}
         treated_dict[id] = pre_treating(p_dict[id])
     mydb.close()
     return treated_dict #返回处理后的字典, 含内容字段及相同题目字段
@@ -3078,5 +3078,84 @@ def generateProDict(cursor): #从数据库的cursor中得到problems字典的con
     return prodict
 
 
+
+def generate_diff_dict(pro_dict): #根据pro_dict的使用数据生成diff_dict难度词典
+    diff_dict = {}
+    for id in pro_dict:
+        usages = pro_dict[id]["usages"]
+        if usages == []:
+            diff_dict[id] = {"min":-1,"average":-1,"median":-1,"max":-1}
+        else:
+            diff_list = []
+            for usage_raw in usages:
+                usage = parseUsage(usage_raw)['difficulties']
+                diff_list.append(np.mean(usage))
+            diff_dict[id] = {"min":min(diff_list),"average":np.mean(diff_list),"median":np.median(diff_list),"max":max(diff_list)}
+    return diff_dict
+
+
+
+def generate_adaptive_problemIDs(pro_dict,units_list,genre_tuple,diffinterval,method="min",excludeids = None,multianswers = False, sim_threshold = 0.8): #生成弗能题组, genre_tuple是二元组(填空题数, 选择题数), diffinterval是二元组(难度最小值,难度最大值)
+    diff_dict = generate_diff_dict(pro_dict)
+    treated_dict = {id:pre_treating(pro_dict[id]["content"]) for id in pro_dict}
+    if excludeids is None:
+        excludeids = ()
+    if len(units_list) == 0:
+        units_list = ["第一单元","第二单元","第三单元","第四单元","第五单元","第六单元","第七单元","第八单元","第九单元"]
+    # treated_dict = {id:pre_treating(pro_dict[id]["content"]) for id in pro_dict}
+    raw_output_list = []
+    same_list = []
+    for id in pro_dict: #生成raw_output_list, 满足条件的所有题目(无论题目个数, 题目类型)
+        tags = pro_dict[id]["tags"]
+        diff_info = diff_dict[id]
+        if len(set(units_list) & set(tags)) > 0 and diffinterval[0] <= diff_info[method] <= diffinterval[1] and not id in excludeids and not id in same_list:
+            raw_output_list.append(id)
+            same_list += pro_dict[id]["same"].copy()
+    if multianswers == True: #排除多个作答项的题目
+        output_list = raw_output_list.copy()
+    else:
+        output_list = []
+        for id in raw_output_list:
+            subproblemsfromusage = [parseUsage(u)["subproblems"] for u in pro_dict[id]["usages"]]
+            if max(subproblemsfromusage) == 1:
+                output_list.append(id)
+    tiankong_list_raw = []
+    xuanze_list_raw = []
+    for id in output_list: #分开生成填空题列表和选择题列表
+        if pro_dict[id]["genre"] == "填空题":
+            tiankong_list_raw.append(id)
+        if pro_dict[id]["genre"] == "选择题":
+            xuanze_list_raw.append(id)
+    print(f"满足条件的填空有 {len(tiankong_list_raw)} 题, 选择有 {len(xuanze_list_raw)} 题.")
+    tiankong_list = []
+    while len(tiankong_list_raw) + len(tiankong_list) > genre_tuple[0] and not len(tiankong_list) == genre_tuple[0]:
+        poping_index = np.random.randint(len(tiankong_list_raw))
+        new_id = tiankong_list_raw.pop(poping_index)
+        newid_applicable = True
+        for id in tiankong_list:
+            if Levenshtein.jaro(treated_dict[id],treated_dict[new_id]) > sim_threshold:
+                newid_applicable = False
+                break
+        if newid_applicable:
+            tiankong_list.append(new_id)
+    if len(tiankong_list_raw) + len(tiankong_list) == genre_tuple[0]:
+        tiankong_list = tiankong_list + tiankong_list_raw
+    xuanze_list = []
+    while len(xuanze_list_raw) + len(xuanze_list) > genre_tuple[1] and not len(xuanze_list) == genre_tuple[1]:
+        poping_index = np.random.randint(len(xuanze_list_raw))
+        new_id = xuanze_list_raw.pop(poping_index)
+        newid_applicable = True
+        for id in tiankong_list + xuanze_list:
+            if Levenshtein.jaro(treated_dict[id],treated_dict[new_id]) > sim_threshold:
+                newid_applicable = False
+                break
+        if newid_applicable:
+            xuanze_list.append(new_id)
+    if len(xuanze_list_raw) + len(xuanze_list) == genre_tuple[1]:
+        xuanze_list = xuanze_list + xuanze_list_raw
+
+    return tiankong_list,xuanze_list
+
+
 if __name__ == "__main__":
     print("数据库工具, import用.")