Update app.py

app.py

@@ -298,68 +298,59 @@ def get_models_data(progress=gr.Progress()):
         """모델 ID를 정규화"""
         return model_id.strip().lower()
         
-    url = "https://huggingface.co/api/models"  # 일반 API 사용
-    
     try:
         progress(0, desc="Fetching models data...")
-        params = {
-            'full': 'true',
-            'limit': 3000,
-            'sort': 'lastModified',  # 최신 수정순으로 정렬
-            'direction': -1
-        }
-        
-        headers = {'Accept': 'application/json'}
-        
-        response = requests.get(url, params=params, headers=headers)
-        if response.status_code != 200:
-            print(f"API 요청 실패: {response.status_code}")
-            print(f"Response: {response.text}")
-            return create_error_plot(), "<div>모델 데이터를 가져오는데 실패했습니다.</div>", pd.DataFrame()
-        
-        models = response.json()
         
-        # 전체 순위 정보 저장
-        model_ranks = {}
-        model_data = {}  # 모든 모델의 상세 데이터 저장
-        
-        for idx, model in enumerate(models, 1):
-            model_id = normalize_model_id(model.get('id', ''))
-            model_data[model_id] = {
-                'rank': idx,
-                'downloads': model.get('downloads', 0),
-                'likes': model.get('likes', 0),
-                'title': model.get('title', 'No Title')
-            }
-        
-        # target_models 중 순위권 내 모델 필터링
+        # 각 모델의 상세 정보를 개별적으로 가져오기
         filtered_models = []
-        for target_id in target_models.keys():
-            normalized_target_id = normalize_model_id(target_id)
-            if normalized_target_id in model_data:
-                model_info = {
-                    'id': target_id,
-                    'rank': model_data[normalized_target_id]['rank'],
-                    'downloads': model_data[normalized_target_id]['downloads'],
-                    'likes': model_data[normalized_target_id]['likes'],
-                    'title': model_data[normalized_target_id]['title']
-                }
-            else:
+        total_models = len(target_models)
+        
+        for idx, (model_id, model_url) in enumerate(target_models.items()):
+            progress((idx + 1) / total_models, desc=f"Fetching model {idx + 1}/{total_models}...")
+            
+            try:
+                # 개별 모델 API 호출
+                model_url_api = f"https://huggingface.co/api/models/{model_id}"
+                response = requests.get(model_url_api, headers={'Accept': 'application/json'})
+                
+                if response.status_code == 200:
+                    model_data = response.json()
+                    model_info = {
+                        'id': model_id,
+                        'downloads': model_data.get('downloads', 0),
+                        'likes': model_data.get('likes', 0),
+                        'title': model_data.get('title', 'No Title')
+                    }
+                    filtered_models.append(model_info)
+                    print(f"Model {model_id}: Downloads={model_info['downloads']}, Likes={model_info['likes']}")
+                else:
+                    print(f"Failed to fetch data for {model_id}: {response.status_code}")
+                    model_info = {
+                        'id': model_id,
+                        'downloads': 0,
+                        'likes': 0,
+                        'title': 'No Title'
+                    }
+                    filtered_models.append(model_info)
+            except Exception as e:
+                print(f"Error fetching data for {model_id}: {str(e)}")
                 model_info = {
-                    'id': target_id,
-                    'rank': 'Not in top 3000',
-                    'downloads': 'N/A',
-                    'likes': 'N/A',
+                    'id': model_id,
+                    'downloads': 0,
+                    'likes': 0,
                     'title': 'No Title'
                 }
-            filtered_models.append(model_info)
+                filtered_models.append(model_info)
         
-        # 순위로 정렬
-        filtered_models.sort(key=lambda x: float('inf') if x['rank'] == 'Not in top 3000' else x['rank'])
-
+        # 다운로드 수로 정렬
+        filtered_models.sort(key=lambda x: x['downloads'], reverse=True)
+        
+        # 순위 할당
+        for idx, model in enumerate(filtered_models, 1):
+            model['rank'] = idx
         
         if not filtered_models:
-            return create_error_plot(), "<div>선택된 모델의 데이터를 찾을 수 없습니다.</div>", pd.DataFrame()
+            return create_error_plot(), "<div>모델 데이터를 찾을 수 없습니다.</div>", pd.DataFrame()
         
         progress(0.3, desc="Creating visualization...")
         
@@ -372,14 +363,11 @@ def get_models_data(progress=gr.Progress()):
         likes = [model['likes'] for model in filtered_models]
         downloads = [model['downloads'] for model in filtered_models]
         
-        # Y축 값을 반전 (숫자 순위만)
-        y_values = [3001 - r if isinstance(r, int) else 0 for r in ranks]
-        
         # 막대 그래프 생성
         fig.add_trace(go.Bar(
             x=ids,
-            y=y_values,
-            text=[f"Rank: {r}<br>Likes: {format(l, ',') if isinstance(l, (int, float)) else l}<br>Downloads: {format(d, ',') if isinstance(d, (int, float)) else d}" 
+            y=ranks,
+            text=[f"Rank: {r}<br>Likes: {format(l, ',')}<br>Downloads: {format(d, ',')}" 
                   for r, l, d in zip(ranks, likes, downloads)],
             textposition='auto',
             marker_color='rgb(158,202,225)',
@@ -388,19 +376,15 @@ def get_models_data(progress=gr.Progress()):
         
         fig.update_layout(
             title={
-                'text': 'Hugging Face Models Global Download Rankings (Top 3000)',
+                'text': 'Hugging Face Models Rankings by Downloads',
                 'y':0.95,
                 'x':0.5,
                 'xanchor': 'center',
                 'yanchor': 'top'
             },
             xaxis_title='Model ID',
-            yaxis_title='Global Rank',
-            yaxis=dict(
-                ticktext=[str(i) for i in range(1, 3001, 150)],
-                tickvals=[3001 - i for i in range(1, 3001, 150)],
-                range=[0, 3000]
-            ),
+            yaxis_title='Rank',
+            yaxis_autorange='reversed',  # 순위를 위에서 아래로
             height=800,
             showlegend=False,
             template='plotly_white',
@@ -412,22 +396,17 @@ def get_models_data(progress=gr.Progress()):
         # HTML 카드 생성
         html_content = """
         <div style='padding: 20px; background: #f5f5f5;'>
-            <h2 style='color: #2c3e50;'>Models 글로벌 인기 순위 (Top 3000)</h2>
+            <h2 style='color: #2c3e50;'>Models Rankings by Downloads</h2>
             <div style='display: grid; grid-template-columns: repeat(auto-fill, minmax(300px, 1fr)); gap: 20px;'>
         """
         
-        # 순위권 내 모델 카드 생성
         for model in filtered_models:
             model_id = model['id']
             rank = model['rank']
-            likes = model.get('likes', 0)
-            downloads = model.get('downloads', 0)
+            likes = model['likes']
+            downloads = model['downloads']
             title = model.get('title', 'No Title')
             
-            # 숫자 포맷팅 처리
-            likes_display = format(likes, ',') if isinstance(likes, (int, float)) else likes
-            downloads_display = format(downloads, ',') if isinstance(downloads, (int, float)) else downloads
-            
             html_content += f"""
             <div style='
                 background: white;
@@ -438,8 +417,8 @@ def get_models_data(progress=gr.Progress()):
             '>
                 <h3 style='color: #34495e;'>Rank #{rank} - {model_id}</h3>
                 <p style='color: #2c3e50;'>{title}</p>
-                <p style='color: #7f8c8d;'>👍 Likes: {likes_display}</p>
-                <p style='color: #7f8c8d;'>⬇️ Downloads: {downloads_display}</p>
+                <p style='color: #7f8c8d;'>👍 Likes: {format(likes, ',')}</p>
+                <p style='color: #7f8c8d;'>⬇️ Downloads: {format(downloads, ',')}</p>
                 <a href='{target_models[model_id]}' 
                    target='_blank' 
                    style='
@@ -459,24 +438,14 @@ def get_models_data(progress=gr.Progress()):
         html_content += "</div></div>"
         
         # 데이터프레임 생성
-        df_data = []
-        # 모든 모델 정보를 데이터프레임에 추가
-        for model in filtered_models:
-            likes = model.get('likes', 0)
-            downloads = model.get('downloads', 0)
-            
-            # 숫자 포맷팅 처리
-            likes_display = format(likes, ',') if isinstance(likes, (int, float)) else likes
-            downloads_display = format(downloads, ',') if isinstance(downloads, (int, float)) else downloads
-            
-            df_data.append({
-                'Global Rank': model['rank'],
-                'Model ID': model['id'],
-                'Title': model.get('title', 'No Title'),
-                'Likes': likes_display,
-                'Downloads': downloads_display,
-                'URL': target_models[model['id']]
-            })
+        df_data = [{
+            'Rank': model['rank'],
+            'Model ID': model['id'],
+            'Title': model.get('title', 'No Title'),
+            'Likes': format(model['likes'], ','),
+            'Downloads': format(model['downloads'], ','),
+            'URL': target_models[model['id']]
+        } for model in filtered_models]
         
         df = pd.DataFrame(df_data)