[analysis] 일부 수정

NewWays-TechForImpactKAIST · Nov 27, 2023 · abb4816 · abb4816
1 parent 2eba982
commit abb4816
Show file tree

Hide file tree

Showing 2 changed files with 35 additions and 26 deletions.
diff --git a/analysis/age/hist_groups.py b/analysis/age/hist_groups.py
@@ -197,10 +197,12 @@ def cluster(df_original, n_clst, basedic):
     if basedic.method == "equal":
         statcoll.delete_many(basedic.__dict__)
     # 연도별로 데이터 찾아서 넣기!
-    years = [int(sgId//10000) for sgId in df_original["sgId"].unique()]
+    df_original["year"] = df_original["sgId"] // 10000
+    df_original = df_original[df_original["year"].isin([2010, 2014, 2018, 2022])]
+    years = df_original["year"].unique()
     for year in years:
-        basedic.year = year
-        df = df_original[df_original["sgId"] // 10000 == year]
+        basedic.year = int(year)
+        df = df_original[df_original["year"] == year]
         youngest_age = ("", 100)
         oldest_age = ("", 0)
         print(f"year {year}, {n_clst} clusters")

diff --git a/analysis/age/main.py b/analysis/age/main.py
@@ -28,34 +28,41 @@ def main(N=5, folder_name="To_be_filled"):
     ## 이 링크에 구현될 save_to_mongo함수 참고 : https://github.com/NewWays-TechForImpactKAIST/API-scrap-and-analysis//blob/bd817e9a15086d313d9615b2515a81e0dbd73850/API/utils.py#L34
     ## 1. 지역의회
     # cluster_by = input("구역을 나눌 기준을 입력해주세요 (sdName 즉 시/도 또는 wiwName 즉 기초단체단위): ")
-    cluster_by = "sdName"
+    cluster_by = "wiwName"
     assert cluster_by in ["sdName", "wiwName"]
     level = 1 if cluster_by == "sdName" else 2
     datadir = os.path.join(BASE_DIR, "_data", folder_name)
-    for d in os.listdir(datadir):
+    # for d in os.listdir(datadir):
         # xlsx 파일을 읽어옵니다.
-        if not d.endswith(".xlsx"):
-            continue
-        df = pd.read_excel(os.path.join(datadir, d))
-        # 필요한 열만 추출합니다.
-        if level == 1:
-            df = df[["sgId", "sdName", "name", "age", "gender"]]
-        else:
-            df = df[["sgId", "sdName", "wiwName", "name", "age", "gender"]]
-        df = df.sort_values(by="age")
-        is_elected = (
-            True
-            if "당선" in d
-            else False
-            if "후보" in d
-            else ValueError("엑셀파일 이름에 '당선'이든지 '후보'가 있어야 합니다.")
+        # if not d.endswith(".xlsx"):
+        #     continue
+    # df = pd.read_excel(os.path.join(datadir, d))
+    # d = "[당선][시도의원].xlsx"
+    d = "[당선][구시군의회의원].xlsx"
+    df_1 = pd.read_excel(os.path.join(datadir, d))
+    # d = "[당선][광역의원비례대표].xlsx"
+    d = "[당선][기초의원비례대표].xlsx"
+    df_2 = pd.read_excel(os.path.join(datadir, d))
+    df = pd.concat([df_1, df_2])
+    # 필요한 열만 추출합니다.
+    if level == 1:
+        df = df[["sgId", "sdName", "name", "age", "gender"]]
+    else:
+        df = df[["sgId", "sdName", "wiwName", "name", "age", "gender"]]
+    df = df.sort_values(by="age")
+    is_elected = (
+        True
+        if "당선" in d
+        else False
+        if "후보" in d
+        else ValueError("엑셀파일 이름에 '당선'이든지 '후보'가 있어야 합니다.")
+    )
+    councilorType = councilordict[d.split('[')[-1].split(']')[0]]
+    for method in ["kmeans", "equal"]:
+        basedic = BasicArgument(councilorType=councilorType, is_elected=is_elected, level=level, method=method)
+        cluster(
+            df, N, basedic
         )
-        councilorType = councilordict[d.split('[')[-1].split(']')[0]]
-        for method in ["kmeans", "equal"]:
-            basedic = BasicArgument(councilorType=councilorType, is_elected=is_elected, level=level, method=method)
-            cluster(
-                df, N, basedic
-            )
     ## 2. 광역의회