hossam 0.4.17__py3-none-any.whl → 0.4.18__py3-none-any.whl

hossam/hs_cluster.py

@@ -23,6 +23,7 @@ from sklearn.metrics import silhouette_score, adjusted_rand_score
 # hossam 패키지 참조
 # ===================================================================
 from . import hs_plot
+from .hs_util import is_2d
 
 RANDOM_STATE = 52
 
@@ -32,10 +33,11 @@ RANDOM_STATE = 52
 # ===================================================================
 def kmeans_fit(
     data: DataFrame,
-    n_clusters: int,
+    n_clusters: int | None = None,
+    k_range: list | tuple = [2, 11],
     random_state: int = RANDOM_STATE,
     plot: bool = False,
-    fields: list[list[str]] | None = None,
+    fields: list[str] | tuple[str] | tuple[tuple[str]] | list[list[str]] | None = None,
     **params,
 ) -> tuple[KMeans, DataFrame, float]:
     """
@@ -43,7 +45,7 @@ def kmeans_fit(
 
     Args:
         data (DataFrame): 군집화할 데이터프레임.
-        n_clusters (int): 군집 개수.
+        n_clusters (int | None): 군집 개수.
         random_state (int, optional): 랜덤 시드. 기본값은 RANDOM_STATE.
         plot (bool, optional): True면 결과를 시각화함. 기본값 False.
         fields (list[list[str]] | None, optional): 시각화할 필드 쌍 리스트. 기본값 None이면 수치형 컬럼의 모든 조합 사용.
@@ -55,18 +57,36 @@ def kmeans_fit(
         float: 실루엣 점수
     """
     df = data.copy()
+
+    if n_clusters is None:
+        n_clusters = kmeans_best_k(data=df, k_range=k_range, random_state=random_state, plot=False)
+        print(f"Best k found: {n_clusters}")
+
     kmeans = KMeans(n_clusters=n_clusters, random_state=random_state, **params)
     kmeans.fit(data)
     df["cluster"] = kmeans.predict(df)
     score = float(silhouette_score(X=data, labels=df["cluster"]))
 
     if plot:
-        cluster_plot(
-            estimator=kmeans,
-            data=data,
-            fields=fields,
-            title=f"K-Means Clustering (k={n_clusters})",
-        )
+
+        if not is_2d(fields):
+            fields = [fields]   # type: ignore
+
+        # cluster_plot(
+        #     estimator=kmeans,
+        #     data=data,
+        #     fields=fields,
+        #     title=f"K-Means Clustering (k={n_clusters})",
+        # )
+        for f in fields:  # type: ignore
+            hs_plot.visualize_silhouette(
+                estimator=kmeans,
+                data=data,
+                xname=f[0],     # type: ignore
+                yname=f[1],     # type: ignore
+                title=f"K-Means Clustering (k={n_clusters})",
+                outline=True,
+            )
 
     return kmeans, df, score
 
@@ -544,6 +564,9 @@ def persona(
         persona_dict[("", f"count")] = len(group)
 
         for field in fields:
+            if field == cluster:
+                continue
+            
             # 명목형일 경우 최빈값 사용
             if df[field].dtype == "object" or df[field].dtype.name == "category":
                 persona_dict[(field, "mode")] = group[field].mode()[0]
@@ -779,7 +802,9 @@ def dbscan_eps(
 
     return best_eps, eps_grid
 
-
+# ===================================================================
+# DBSCAN 군집화 모델을 적합하고 최적의 eps 값을 탐지하는 함수.
+# ===================================================================
 def dbscan_fit(
     data: DataFrame,
     eps: float | list | np.ndarray | None = None,
@@ -789,6 +814,25 @@ def dbscan_fit(
     plot: bool = True,
     **params,
 ) -> tuple[DBSCAN, DataFrame, DataFrame]:
+    """
+    DBSCAN 군집화 모델을 적합하고 최적의 eps 값을 탐지하는 함수.
+
+    Args:
+        data (DataFrame): 군집화할 데이터프레임.
+        eps (float | list | np.ndarray | None, optional): eps 값 또는 리스트.
+            None이면 최적의 eps 값을 탐지함. 기본값 None.
+        min_samples (int, optional): 핵심점이 되기 위한 최소 샘플수. 기본값 5.
+        ari_threshold (float, optional): 안정 구간 탐지를 위한 ARI 임계값. 기본값 0.9.
+        noise_diff_threshold (float, optional): 안정 구간 탐지를 위한 노이즈 비율 변화 임계값. 기본값 0.05.
+        plot (bool, optional): True면 결과를 시각화함. 기본값 True.
+        **params: DBSCAN에 전달할 추가 파라미터.
+
+    Returns:
+        tuple: (estimator, cluster_df, result_df)
+            - estimator: 적합된 DBSCAN 모델 또는 모델 리스트(최적 eps가 여러 개인 경우).
+            - cluster_df: 클러스터 및 벡터 유형이 포함된 데이터 프레임 또는 데이터 프레임 리스트(최적 eps가 여러 개인 경우).
+            - result_df: eps 값에 따른 군집화 요약 통계 데이터 프레임.
+    """
 
     # eps 값이 지정되지 않은 경우 최적의 eps 탐지
     if eps is None:
@@ -874,18 +918,52 @@ def dbscan_fit(
                 # result_dfs에서 recommand가 best에 해당하는 인덱스와 같은 위치의 추정기만 추출
                 best_indexes = list(result_dfs[result_dfs["recommand"] == "best"].index)  # type: ignore
 
-                for i in range(len(estimators) - 1, -1, -1):
-                    if i not in best_indexes:
-                        del estimators[i]
-                        del cluster_dfs[i]
+                # for i in range(len(estimators) - 1, -1, -1):
+                #     if i not in best_indexes:
+                #         del estimators[i]
+                #         del cluster_dfs[i]
 
             pbar.update(1)
 
-    return (
-        estimators[0] if len(estimators) == 1 else estimators,  # type: ignore
-        cluster_dfs[0] if len(cluster_dfs) == 1 else cluster_dfs,
-        result_dfs,  # type: ignore
-    )
+    # best 모델 선정: recommand=='best'인 인덱스의 estimator/cluster_df만 반환
+    if len(estimators) == 1:
+
+        if plot:
+            hs_plot.scatterplot(
+                df=cluster_dfs[0],
+                xname=cluster_dfs[0].columns[0],
+                yname=cluster_dfs[0].columns[1],
+                hue="cluster",
+                vector="vector",
+                title=f"DBSCAN Clustering (eps={estimators[0].eps}, min_samples={estimators[0].min_samples})",
+                outline=True
+            )
+
+        return estimators[0], cluster_dfs[0], result_dfs # type: ignore
+    
+    # recommand=='best'인 인덱스 추출 (여러 개면 첫 번째)
+    best_indexes = list(result_dfs[result_dfs["recommand"] == "best"].index) # type: ignore
+    if not best_indexes:
+        # fallback: 첫 번째
+        best_index = 0
+    else:
+        best_index = best_indexes[0]
+
+    best_estimator = estimators[best_index]
+    best_cluster_df = cluster_dfs[best_index]
+
+    if plot:
+        hs_plot.scatterplot(
+            df=best_cluster_df,
+            xname=best_cluster_df.columns[0],
+            yname=best_cluster_df.columns[1],
+            hue="cluster",
+            vector="vector",
+            title=f"DBSCAN Clustering (eps={best_estimator.eps}, min_samples={best_estimator.min_samples})",
+            outline=True
+        )
+
+    return best_estimator, best_cluster_df, result_dfs # type: ignore
 
 
 # ===================================================================
@@ -950,8 +1028,8 @@ def agg_fit(
 
     Returns:
         tuple: (estimator(s), df(s), score_df)
-            - estimator(s): 적합된 AgglomerativeClustering 모델 또는 모델 리스트.
-            - df(s): 클러스터 결과가 포함된 데이터 프레임 또는 데이터 프레임 리스트.
+            - estimator(s): 적합된 AgglomerativeClustering 모델 또는 모델 리스트 (n_clusters가 리스트일 때 리턴도 리스트로 처리됨).
+            - df(s): 클러스터 결과가 포함된 데이터 프레임 또는 데이터 프레임 리스트(n_cluseters가 리스트일 때 리턴되 리스트로 처리됨).
             - score_df: 각 군집 개수에 대한 실루엣 점수 데이터프레임.
 
     Examples:

hossam/hs_plot.py

@@ -309,7 +309,7 @@ def lineplot(
 # 상자그림(boxplot)을 그린다
 # ===================================================================
 def boxplot(
-    df: DataFrame,
+    df: DataFrame | None = None,
     xname: str | None = None,
     yname: str | None = None,
     title: str | None = None,
@@ -331,7 +331,7 @@ def boxplot(
     """상자그림(boxplot)을 그린다.
 
     Args:
-        df (DataFrame): 시각화할 데이터.
+        df (DataFrame|None): 시각화할 데이터.
         xname (str|None): x축 범주 컬럼명.
         yname (str|None): y축 값 컬럼명.
         title (str|None): 그래프 제목.
@@ -359,13 +359,20 @@ def boxplot(
         fig, ax = get_default_ax(width, height, 1, 1, dpi)  # type: ignore
         outparams = True
 
-    if xname is not None and yname is not None:
+    if xname is not None or yname is not None:
+        if xname is not None and yname is None:
+            orient = "h"
+        elif xname is None and yname is not None:
+            orient = "v"
+
+
         boxplot_kwargs = {
             "data": df,
             "x": xname,
             "y": yname,
             "orient": orient,
             "ax": ax,
+            "linewidth": linewidth,
         }
 
         # hue 파라미터 확인 (params에 있을 수 있음)
@@ -377,12 +384,12 @@ def boxplot(
             boxplot_kwargs["color"] = sb.color_palette(palette)[0]
 
         boxplot_kwargs.update(params)
-        sb.boxplot(**boxplot_kwargs, linewidth=linewidth)
+        sb.boxplot(**boxplot_kwargs)
 
         # 통계 검정 추가
         if stat_test is not None:
             if stat_pairs is None:
-                stat_pairs = [df[xname].dropna().unique().tolist()]
+                stat_pairs = [df[xname].dropna().unique().tolist()] # type: ignore
 
             annotator = Annotator(
                 ax, data=df, x=xname, y=yname, pairs=stat_pairs, orient=orient
@@ -847,15 +854,15 @@ def scatterplot(
     else:
         # 핵심벡터
         scatterplot_kwargs["edgecolor"] = "#ffffff"
-        sb.scatterplot(data=df[df[vector] == "core"], **scatterplot_kwargs)
+        sb.scatterplot(data=df[df[vector] == "core"], **scatterplot_kwargs) # type: ignore
 
         # 외곽백터
         scatterplot_kwargs["edgecolor"] = "#000000"
         scatterplot_kwargs["s"] = 25
         scatterplot_kwargs["marker"] = "^"
         scatterplot_kwargs["linewidth"] = 0.8
-        sb.scatterplot(data=df[df[vector] == "border"], **scatterplot_kwargs)
-
+        sb.scatterplot(data=df[df[vector] == "border"], **scatterplot_kwargs) # type: ignore
+ 
         # 노이즈벡터
         scatterplot_kwargs["edgecolor"] = None
         scatterplot_kwargs["s"] = 25
@@ -3022,10 +3029,15 @@ def pca_plot(
         if field_group is not None:
             title += " - " + ", ".join(field_group)
 
+        tdf = DataFrame({
+            field_group[0]: xs * scalex,
+            field_group[1]: ys * scaley,
+        })
+
         scatterplot(
-            df=None,
-            xname=xs * scalex,
-            yname=ys * scaley,
+            df=tdf,
+            xname=field_group[0],
+            yname=field_group[1],
             hue=pca_df[hue] if hue is not None else None,
             outline=False,
             palette=palette,

hossam/hs_stats.py

@@ -244,6 +244,41 @@ def outlier_table(data: DataFrame, *fields: str):
         outlier_count = ((data[f] < down) | (data[f] > up)).sum()
         outlier_rate = (outlier_count / len(data)) * 100
 
+        # 왜도
+        skew = data[f].skew()
+
+        # 이상치 개수 및 비율
+        outlier_count = ((data[f] < down) | (data[f] > up)).sum()
+        outlier_rate = (outlier_count / len(data)) * 100
+
+        # 분포 특성 판정 (왜도 기준)
+        abs_skew = abs(skew)    # type: ignore
+        if abs_skew < 0.5:      # type: ignore
+            dist = "거의 대칭"
+        elif abs_skew < 1.0:    # type: ignore
+            if skew > 0:        # type: ignore
+                dist = "약한 우측 꼬리"
+            else:
+                dist = "약한 좌측 꼬리"
+        elif abs_skew < 2.0:    # type: ignore
+            if skew > 0:        # type: ignore
+                dist = "중간 우측 꼬리"
+            else:
+                dist = "중간 좌측 꼬리"
+        else:
+            if skew > 0:        # type: ignore
+                dist = "극단 우측 꼬리"
+            else:
+                dist = "극단 좌측 꼬리"
+
+        # 로그변환 필요성 판정
+        if abs_skew < 0.5:      # type: ignore
+            log_need = "낮음"
+        elif abs_skew < 1.0:    # type: ignore
+            log_need = "중간"
+        else:
+            log_need = "높음"
+
         iq = {
             "field": f,
             "q1": q1,
@@ -254,9 +289,11 @@ def outlier_table(data: DataFrame, *fields: str):
             "down": down,
             "min": min_value,
             "max": max_value,
-            "skew": skew,
             "outlier_count": outlier_count,
-            "outlier_rate": outlier_rate
+            "outlier_rate": outlier_rate,
+            "skew": skew,
+            "dist": dist,
+            "log_need": log_need
         }
 
         result.append(iq)

{hossam-0.4.17.dist-info → hossam-0.4.18.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: hossam
-Version: 0.4.17
+Version: 0.4.18
 Summary: Hossam Data Helper
 Author-email: Lee Kwang-Ho <leekh4232@gmail.com>
 License-Expression: MIT

hossam-0.4.18.dist-info/RECORD

@@ -0,0 +1,18 @@
+hossam/NotoSansKR-Regular.ttf,sha256=0SCufUQwcVWrWTu75j4Lt_V2bgBJIBXl1p8iAJJYkVY,6185516
+hossam/__init__.py,sha256=_v_Gj01j6nNtDfcxS9B2MB8ZPQ_PVnY9jTkvgRJdVxc,4492
+hossam/hs_classroom.py,sha256=Sb1thy49LKn2zU90aiOVwHWhyWSMHLZbZX7eXmQlquc,27523
+hossam/hs_cluster copy.py,sha256=1pylX0lV_FR1ZhcYN1sosMcAxj5WGLjwgHHjkkCPsCY,39761
+hossam/hs_cluster.py,sha256=wrDPsF14tHwRuwBEUz765j_I87JH5HFtZZoUuGf-W-4,41190
+hossam/hs_gis.py,sha256=DVmndBK-_7GMK3J1_on3ieEQk1S0MfUZ8_wlX-cDdZQ,11581
+hossam/hs_plot.py,sha256=KW9NtD-ecFvas5h58ayEPzV0Eu2Dtt6twhj4z4LhPzA,103127
+hossam/hs_prep.py,sha256=lsK_BxuOXNdTTNED-nkw3EY2vt5nQGMvkWvVLqNkwM0,43255
+hossam/hs_stats.py,sha256=1h3RPOVJ_5EGO5_1yHTE8zpuJyy85xOLPqWmR-xqqmo,121161
+hossam/hs_study.py,sha256=ZzL76_V0IHnk_YUTbWncIIBruOj2Sz3xs91snS6cpu0,2776
+hossam/hs_timeserise.py,sha256=NzGV4bJmVQr3qUFySOP25qENItmloYjgh3VgwSbSmXc,43163
+hossam/hs_util.py,sha256=LJvoTFSLX_uADJrC1ONGM93DX2HS2jcNBnPariLPZko,16704
+hossam/leekh.png,sha256=1PB5NQ24SDoHA5KMiBBsWpSa3iniFcwFTuGwuOsTHfI,6395
+hossam-0.4.18.dist-info/licenses/LICENSE,sha256=nIqzhlcFY_2D6QtFsYjwU7BWkafo-rUJOQpDZ-DsauI,941
+hossam-0.4.18.dist-info/METADATA,sha256=FUrD5dnNDWgA8dChQq8gmFg1lQ47YKiIdqhItMl6Vi8,3803
+hossam-0.4.18.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
+hossam-0.4.18.dist-info/top_level.txt,sha256=_-7bwjhthHplWhywEaHIJX2yL11CQCaLjCNSBlk6wiQ,7
+hossam-0.4.18.dist-info/RECORD,,

hossam-0.4.17.dist-info/RECORD

@@ -1,17 +0,0 @@
-hossam/NotoSansKR-Regular.ttf,sha256=0SCufUQwcVWrWTu75j4Lt_V2bgBJIBXl1p8iAJJYkVY,6185516
-hossam/__init__.py,sha256=lJ-_g2HAmFnixOmKjCv7_cMSdiYwbM6SNlHEtptUlUI,4045
-hossam/hs_classroom.py,sha256=Sb1thy49LKn2zU90aiOVwHWhyWSMHLZbZX7eXmQlquc,27523
-hossam/hs_cluster.py,sha256=yFlEaLz-cEueurw5nvKuUogdwepOEU3jN7woaYeN-cM,37581
-hossam/hs_gis.py,sha256=DVmndBK-_7GMK3J1_on3ieEQk1S0MfUZ8_wlX-cDdZQ,11581
-hossam/hs_plot.py,sha256=nuN8DoQbMRLMxFRO1VD1zhg2yA1NHqFGrjuvS4FLahY,102773
-hossam/hs_prep.py,sha256=lsK_BxuOXNdTTNED-nkw3EY2vt5nQGMvkWvVLqNkwM0,43255
-hossam/hs_stats.py,sha256=MDS3rvaXDP8aYwcE36JTetWiZgE4fkXnNo0vwlXu-pA,119890
-hossam/hs_study.py,sha256=ZzL76_V0IHnk_YUTbWncIIBruOj2Sz3xs91snS6cpu0,2776
-hossam/hs_timeserise.py,sha256=NzGV4bJmVQr3qUFySOP25qENItmloYjgh3VgwSbSmXc,43163
-hossam/hs_util.py,sha256=LJvoTFSLX_uADJrC1ONGM93DX2HS2jcNBnPariLPZko,16704
-hossam/leekh.png,sha256=1PB5NQ24SDoHA5KMiBBsWpSa3iniFcwFTuGwuOsTHfI,6395
-hossam-0.4.17.dist-info/licenses/LICENSE,sha256=nIqzhlcFY_2D6QtFsYjwU7BWkafo-rUJOQpDZ-DsauI,941
-hossam-0.4.17.dist-info/METADATA,sha256=camh3iJFBxxmzX1tHg71MWwNcqcY3a_GSKbNhf8iR7E,3803
-hossam-0.4.17.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
-hossam-0.4.17.dist-info/top_level.txt,sha256=_-7bwjhthHplWhywEaHIJX2yL11CQCaLjCNSBlk6wiQ,7
-hossam-0.4.17.dist-info/RECORD,,