Upload 6 files

app.py

@@ -0,0 +1,405 @@
+import gradio as gr
+from openai import OpenAI
+from huggingface_hub import InferenceClient
+import os
+import sys
+import subprocess
+import time
+import requests
+import urllib.parse
+import pandas as pd
+from langchain_huggingface import HuggingFaceEmbeddings
+from langchain_community.vectorstores import FAISS
+from PIL import Image
+from dotenv import load_dotenv
+
+# ==========================================
+# Step 0: 環境安裝檢查 (僅保留 Playwright)
+# ==========================================
+def install_playwright():
+    """確保 Playwright 瀏覽器核心有安裝"""
+    try:
+        import playwright
+    except ImportError:
+        print("⚠️ 偵測到缺少 playwright，正在強制安裝...")
+        subprocess.check_call([sys.executable, "-m", "pip", "install", "playwright"])
+    
+    # 設定瀏覽器路徑
+    os.environ["PLAYWRIGHT_BROWSERS_PATH"] = os.path.join(os.getcwd(), "playwright_browsers")
+    
+    print("🔄 檢查 Chromium 瀏覽器...")
+    try:
+        # 檢查是否已安裝，若無則安裝
+        if not os.path.exists(os.environ["PLAYWRIGHT_BROWSERS_PATH"]):
+             subprocess.run([sys.executable, "-m", "playwright", "install", "chromium"], check=True)
+    except Exception as e:
+        print(f"⚠️ 瀏覽器安裝警告: {e}")
+
+# 執行安裝
+install_playwright()
+
+# 延遲匯入
+from playwright.sync_api import sync_playwright
+
+load_dotenv()
+
+# ==========================================
+# 模組載入 (emotion 模組)
+# ==========================================
+import emotion
+
+# ==========================================
+# 0. 環境變數 & 1. 資料初始化
+# ==========================================
+GROQ_API_KEY = os.getenv("GROQ_API_KEY")
+HF_TOKEN = os.getenv("HF_TOKEN")
+
+global_df = None
+global_mood_df = None
+global_retriever = None
+rag_initialized = False
+
+def init_static_data():
+    """初始化靜態的心情資料：讀取外部 CSV"""
+    global global_mood_df
+    if global_mood_df is None:
+        csv_path = 'mood_food_guide.csv'
+        if os.path.exists(csv_path):
+            try:
+                # 讀取 CSV，確保欄位正確
+                global_mood_df = pd.read_csv(csv_path)
+                print(f"✅ 成功載入心情指南：{csv_path}")
+            except Exception as e:
+                print(f"❌ 讀取 CSV 失敗: {e}")
+                global_mood_df = pd.DataFrame() # 建立空表防止後續報錯
+        else:
+            print(f"⚠️ 警告：找不到 {csv_path}，請確認檔案已上傳至 Space。")
+            global_mood_df = pd.DataFrame()
+
+# ==========================================
+# API 金鑰檢查工具函式
+# ==========================================
+def check_api_key_status(name, key):
+    """檢查 API Key 是否存在，並回傳部分內容以供辨識"""
+    if not key:
+        return "❌ 未設定 (Not Set)"
+    
+    # 遮罩處理，只顯示前後幾碼
+    if len(key) > 8:
+        masked = f"{key[:4]}...{key[-4:]}"
+    else:
+        masked = "******"
+    return f"✅ 已設定 ({masked})"
+
+# ==========================================
+# 2. Google Maps 爬蟲功能
+# ==========================================
+def sync_google_maps(url):
+    global global_df, global_retriever, rag_initialized
+    clean_url = url.strip() if url else ""
+    if not clean_url or "http" not in clean_url.lower():
+        yield "❌ 請輸入有效的 Google Maps 分享連結。"
+        return
+
+    try:
+        yield "🚀 [1/4] 啟動瀏覽器..."
+        with sync_playwright() as p:
+            browser = p.chromium.launch(headless=True)
+            page = browser.new_page()
+            yield "🌐 [2/4] 連線中..."
+            page.goto(clean_url, wait_until="domcontentloaded", timeout=60000)
+            yield "⏳ [3/4] 等待列表加載 (約 10 秒)..."
+            time.sleep(10) 
+            yield "📄 [4/4] 解析餐廳資訊..."
+            
+            titles = page.locator('div.fontHeadlineSmall').all_inner_texts()
+            details = page.locator('div.fontBodyMedium').all_inner_texts()
+            
+            restaurant_list = []
+            for i, name in enumerate(titles):
+                name = name.strip()
+                if name:
+                    addr = details[i].strip() if i < len(details) else ""
+                    search_query = f"{name} {addr}"
+                    restaurant_list.append({
+                        "Name": name,
+                        "Address": addr,
+                        "Category": "未分類",
+                        "RAG_Content": f"餐廳：{name}，資訊：{addr}",
+                        "URL": f"https://www.google.com/maps/search/?api=1&query={urllib.parse.quote(search_query)}"
+                    })
+            browser.close()
+
+            if not restaurant_list:
+                yield "⚠️ 找不到餐廳。請確認連結格式正確且已公開。"
+                return
+
+            global_df = pd.DataFrame(restaurant_list)
+            
+            try:
+                embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")
+                vectorstore = FAISS.from_texts(global_df['RAG_Content'].tolist(), embeddings, metadatas=global_df.to_dict('records'))
+                global_retriever = vectorstore.as_retriever(search_kwargs={"k": 3})
+            except Exception as e:
+                print(f"RAG 初始化警告: {e}")
+
+            rag_initialized = True
+            yield f"✅ 同步成功！已載入 {len(restaurant_list)} 間餐廳。"
+    except Exception as e:
+        yield f"❌ 系統錯誤：{str(e)}"
+
+# ==========================================
+# 3. 核心功能 (AI Agent)
+# ==========================================
+def get_restaurant_data(mood_score_str, food_choice):
+    init_static_data() # 確保心情指南已載入
+    
+    if global_df is None or global_df.empty: 
+        return None, True, "", "資料庫未載入，無建議"
+    
+    # 解析分數 (例如 "3 (普通)" -> 3)
+    try: score = int(str(mood_score_str).split(' ')[0])
+    except: score = 3
+    
+    # 從 CSV 資料中查找對應的推薦類別與原因
+    rec_categories = ""
+    mood_reason = "隨意探索"
+    
+    if global_mood_df is not None and not global_mood_df.empty:
+        mood_info = global_mood_df[global_mood_df['分數'] == score]
+        if not mood_info.empty:
+            rec_categories = mood_info.iloc[0]['推薦料理類別']
+            mood_reason = mood_info.iloc[0]['原因']
+    
+    candidates = global_df.copy()
+    food_keyword = "飯" if food_choice == "吃飯" else "麵" if food_choice == "吃麵" else ""
+    
+    is_random = False
+    if food_keyword:
+        filtered = candidates[candidates['Name'].str.contains(food_keyword, case=False, na=False) | candidates['RAG_Content'].str.contains(food_keyword, case=False, na=False)]
+        if not filtered.empty:
+            candidates = filtered
+        else:
+            is_random = True 
+    
+    if candidates.empty:
+        result = global_df.sample(1).iloc[0]; is_random = True
+    else:
+        result = candidates.sample(1).iloc[0]
+        
+    return result, is_random, rec_categories, mood_reason
+
+def generate_content_with_groq(restaurant_name, restaurant_detail, user_diary, mood_score, mood_guide_reason, debug_mode=False):
+    if not GROQ_API_KEY: return "⚠️ 請設定 GROQ_API_KEY", ""
+    
+    client = OpenAI(api_key=GROQ_API_KEY, base_url="https://api.groq.com/openai/v1")
+    
+    system_prompt = "你是一個幽默、懂吃且善解人意的 AI 朋友。請根據使用者的日記、心情以及「心情美食指南」來推薦餐廳。"
+    user_msg = f"""
+        【狀態】心情分數：{mood_score}，日記：{user_diary}
+        【心情美食指南建議】
+        因為分數是 {mood_score}，建議吃這類食物的原因是：「{mood_guide_reason}」。
+        
+        【推薦餐廳】
+        名稱：{restaurant_name}
+        資料：{restaurant_detail}
+        
+        任務：
+        請用繁體中文寫一段溫暖有趣的回覆：
+        1. 先回應他的日記與測驗人設。
+        2. 引用「心情美食指南」的原因，告訴他為什麼現在適合吃這家餐廳。
+        3. 介紹這家餐廳的特色。
+    """
+    
+    debug_log = ""
+    if debug_mode:
+        debug_log = f"### 🔧 Groq Prompt Debug\n**System:** {system_prompt}\n**User:** {user_msg}"
+
+    try:
+        response = client.chat.completions.create(model="llama-3.3-70b-versatile", messages=[{"role": "system", "content": system_prompt}, {"role": "user", "content": user_msg}])
+        return response.choices[0].message.content, debug_log
+    except Exception as e: 
+        return f"Groq Error: {str(e)}", debug_log
+
+def generate_image_huggingface(prompt):
+    if not HF_TOKEN: 
+        return None, "HF_TOKEN 未設定"
+    
+    try:
+        hf_client = InferenceClient(token=HF_TOKEN)
+        image = hf_client.text_to_image(
+            prompt=prompt, 
+            negative_prompt="blurry, low quality, distortion, text, watermark",
+            model="stabilityai/stable-diffusion-xl-base-1.0"
+        )
+        return image, None # 成功時，錯誤訊息為 None
+    except Exception as e:
+        # 回傳具體錯誤訊息
+        return None, str(e)
+
+def mood_agent_logic(score_input, food_input, diary_input, debug_mode):
+    if not rag_initialized:
+        yield "⚠️ 請先同步地圖清單！", None, "", gr.update()
+        return
+    
+    restaurant, is_random, rec_categories, mood_reason = get_restaurant_data(score_input, food_input)
+    if restaurant is None: 
+        yield "資料庫讀取錯誤", None, "", gr.update()
+        return
+    
+    name = restaurant['Name']; url = restaurant['URL']
+    info = restaurant.get('RAG_Content', '')
+    rag_info = str(restaurant.get('RAG_Content', ''))
+    
+    img_prompt = f"Delicious food photography of {name}, {info}. high quality, photorealistic, 8k, cinematic lighting, appetizing, restaurant atmosphere, 50mm lens"
+    
+    ai_text, groq_debug_log = generate_content_with_groq(name, rag_info, diary_input, score_input, mood_reason, debug_mode)
+    
+    prefix = ""
+    if is_random and food_input != "隨便":
+        prefix = f"> 💡 **溫馨提示**：清單中暫無『{food_input}』，已從現有名單挑選最適合的店！\n\n"
+
+    # 組合完整的除錯資訊，包含 API Key 狀態與錯誤訊息
+    full_debug_log = ""
+    if debug_mode:
+        # 1. 檢查 Key 狀態
+        groq_status = check_api_key_status("GROQ_API_KEY", GROQ_API_KEY)
+        hf_status = check_api_key_status("HF_TOKEN", HF_TOKEN)
+        
+        api_debug_block = f"""
+### 🔑 API 金鑰與系統狀態
+- **GROQ_API_KEY**: {groq_status}
+- **HF_TOKEN**: {hf_status}
+- **RAG 狀態**: {"✅ 已初始化" if rag_initialized else "❌ 未初始化"}
+"""
+        full_debug_log = api_debug_block + "\n" + groq_debug_log + f"\n\n### 🎨 Image Prompt Debug\n{img_prompt}"
+
+    debug_output_update = gr.update(value=full_debug_log, visible=debug_mode)
+    
+    final_response = f"{prefix}### 🍽️ 推薦：{name}\n\n{ai_text}"
+    map_html = f'<div style="text-align:center"><a href="{url}" target="_blank" style="background:#4CAF50;color:white;padding:8px 16px;border-radius:20px;text-decoration:none">🗺️ Google Map 導航</a></div>'
+    
+    # 先回傳文字
+    yield final_response, None, map_html, debug_output_update
+    
+    # 呼叫圖片生成並捕捉錯誤
+    image_output, img_error = generate_image_huggingface(img_prompt)
+    
+    # 如果有圖片錯誤且在除錯模式，追加錯誤訊息到 log
+    if img_error and debug_mode:
+        full_debug_log += f"\n\n⚠️ **Hugging Face 圖片生成失敗:**\n{img_error}"
+        debug_output_update = gr.update(value=full_debug_log)
+        
+    yield final_response, image_output, map_html, debug_output_update
+    
+# ==========================================
+# 4. 橋接邏輯
+# ==========================================
+def _score_to_radio_value(score):
+    mapping = {1: "1 (心情差)", 2: "2 (不太好)", 3: "3 (普通)", 4: "4 (不錯)", 5: "5 (超棒)"}
+    try: score = int(score)
+    except: score = 3
+    return mapping.get(score, "3 (普通)")
+
+def bridge_start_click(st):
+    try:
+        res = emotion.on_restart(st)
+        return res[1], res[0], gr.update(visible=True), res[5], res[4], gr.update(visible=False)
+    except: return gr.update(), gr.update(), gr.update(), gr.update(), st, gr.update()
+
+def bridge_stop_click(st):
+    try:
+        res = emotion.on_stop(st)
+        return res[1], res[0], gr.update(visible=False), res[5], res[4]
+    except: return gr.update(), gr.update(), gr.update(), gr.update(), st
+
+def bridge_predict_frame(frame, st):
+    try:
+        res = emotion.on_stream(frame, st)
+        out_cam = res[0]; out_result = res[1]; out_st = res[4]; out_btn_start = res[5]
+        score_update = gr.update(); btn_go_visible = gr.update(visible=False)
+        out_btn_stop = gr.update()
+
+        if hasattr(out_st, 'finished') and out_st.finished and hasattr(out_st, 'final_score'):
+            new_val = _score_to_radio_value(out_st.final_score)
+            score_update = gr.update(value=new_val)
+            out_btn_stop = gr.update(visible=False)
+            btn_go_visible = gr.update(visible=True)
+            
+        return out_cam, out_result, out_st, out_btn_stop, out_btn_start, score_update, btn_go_visible
+    except Exception as e:
+        return frame, gr.update(), st, gr.update(), gr.update(), gr.update(), gr.update()
+
+def bridge_predict_upload(img, st):
+    try:
+        res = emotion.on_upload(img, st)
+        out_result, out_st = res[0], res[2]
+        score_update = gr.update()
+        if hasattr(out_st, 'finished') and out_st.finished and hasattr(out_st, 'final_score'):
+            new_val = _score_to_radio_value(out_st.final_score)
+            score_update = gr.update(value=new_val)
+        return out_result, score_update, gr.Tabs(selected=1), out_st
+    except: return gr.update(), gr.update(), gr.Tabs(), st
+
+# ==========================================
+# 5. Gradio 介面
+# ==========================================
+css_ = "#app_container { max-width: 960px; margin: 0 auto; }" 
+if hasattr(emotion, 'css'): css_ += "\n" + emotion.css
+
+with gr.Blocks(title="AI 心情食堂", css=css_) as demo:
+    st_state = gr.State(emotion.AppState())
+
+    with gr.Tabs() as tabs:
+        # Tab 1
+        with gr.TabItem("😊 情緒辨識 (Step 1)", id=0):
+            with gr.Column(elem_id="app_container"):
+                gr.Markdown("### 第一步：測測你的心情能量\n讓 AI 看看你的表情，自動幫你決定心情分數！")
+                with gr.Row():
+                    btn_start = gr.Button("📸 開啟攝影機辨識", variant="primary")
+                    btn_stop  = gr.Button("⏹️ 停止", variant="secondary", visible=False)
+                cam = gr.Image(sources=["webcam"], streaming=True, type="numpy", label="攝影機畫面", visible=False)
+                result_markdown = gr.Markdown(emotion._hint_html("請按「開啟攝影機辨識」並允許瀏覽器使用相機。"))
+                btn_go_dining = gr.Button("🚀 確定心情，來找餐廳！", variant="primary", visible=False, size="lg")
+
+        # Tab 2
+        with gr.TabItem("🍽️ AI 心情食堂 (Step 2)", id=1):
+            with gr.Column():
+                gr.Markdown(f"## 🔗 載入你的口袋名單")
+                with gr.Row():
+                    map_url = gr.Textbox(label="Google Maps Saved Lists 連結", placeholder="請貼上清單的分享連結...", scale=3)
+                    sync_btn = gr.Button("🔄 同步清單", variant="secondary", scale=1)
+                sync_msg = gr.Markdown("ℹ️ 尚未同步資料庫")
+                gr.Markdown("---")
+                gr.Markdown(f"## 🍱 今天想吃點什麼？")
+                with gr.Row():
+                    with gr.Column(scale=1):
+                        score_input = gr.Radio(["1 (心情差)", "2 (不太好)", "3 (普通)", "4 (不錯)", "5 (超棒)"], label="1. 心情分數 (由 Tab 1 自動填入)", value="3 (普通)")
+                        food_input = gr.Radio(["吃飯", "吃麵", "隨便"], label="2. 想吃什麼", value="隨便")
+                        diary_input = gr.Textbox(lines=4, label="3. 心情日記", placeholder="寫下今天發生的事...")
+                        
+                        debug_mode_btn = gr.Checkbox(label="🔧 開啟除錯模式", value=False)
+                        submit_btn = gr.Button("🍱 送出給 Agent", variant="primary")
+                        debug_output = gr.Markdown(label="除錯資訊 (Debug Log)", visible=False)
+                    
+                    with gr.Column(scale=1):
+                        agent_output = gr.Markdown(label="AI 回應")
+                        image_output = gr.Image(label="AI 推薦美食圖", type="pil", width=400)
+                        map_output = gr.HTML(label="地圖導航")
+
+    # 事件
+    sync_btn.click(fn=sync_google_maps, inputs=[map_url], outputs=[sync_msg])
+    btn_start.click(fn=bridge_start_click, inputs=[st_state], outputs=[result_markdown, cam, btn_stop, btn_start, st_state, btn_go_dining])
+    btn_stop.click(fn=bridge_stop_click, inputs=[st_state], outputs=[result_markdown, cam, btn_stop, btn_start, st_state])
+    cam.stream(fn=bridge_predict_frame, inputs=[cam, st_state], outputs=[cam, result_markdown, st_state, btn_stop, btn_start, score_input, btn_go_dining])
+    btn_go_dining.click(fn=lambda: gr.Tabs(selected=1), inputs=None, outputs=tabs)
+    
+    # debug_mode
+    submit_btn.click(
+        fn=mood_agent_logic, 
+        inputs=[score_input, food_input, diary_input, debug_mode_btn], 
+        outputs=[agent_output, image_output, map_output, debug_output]
+    )
+
+if __name__ == "__main__":
+    demo.launch(ssr_mode=False)

blaze_face_short_range.tflite

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4578f35940bf5a1a655214a1cce5cab13eba73c1297cd78e1a04c2380b0152f
+size 229746

emotion.py

@@ -0,0 +1,623 @@
+import time
+import threading
+from dataclasses import dataclass, field
+from typing import Optional, List, Tuple, Dict
+
+import gradio as gr
+import numpy as np
+import cv2
+
+import mediapipe as mp
+from mediapipe.tasks import python
+from mediapipe.tasks.python import vision
+
+from deepface import DeepFace
+
+
+# =========================
+# 參數（可微調）
+# =========================
+ANALYZE_EVERY_SEC = 1.5          # ⭐ 建議 1.5~2.0：HF CPU 比較順
+EMA_ALPHA = 0.45
+CONF_TH = 40.0                   # DeepFace emotion confidence 門檻
+SWITCH_CONFIRM_SEC = 2.5         # 連續穩定多久才算完成
+ZOOM_FACTOR = 1.5                # ✅ 保留：只用在「顯示」
+UI_TARGET_W = 720                # ✅ 回傳畫面縮小寬度，提升順暢度
+
+# MediaPipe face detector 門檻（太高很容易漏）
+MP_MIN_DET_CONF = 0.3
+
+# 模型路徑（確認 repo 中模型真的在這裡）
+MP_MODEL_PATH = "blaze_face_short_range.tflite"
+
+
+# =========================
+# 情緒中文 + 分數映射（7 -> 5）
+# =========================
+EMO_ZH = {
+    "angry": "生氣",
+    "disgust": "厭惡",
+    "fear": "害怕",
+    "happy": "開心",
+    "sad": "難過",
+    "surprise": "驚訝",
+    "neutral": "平靜",
+    "unknown": "未知",
+}
+
+EMO_TO_SCORE = {
+    "angry": 1,
+    "disgust": 1,
+    "fear": 1,
+    "sad": 2,
+    "neutral": 3,
+    "surprise": 4,
+    "happy": 5,
+    "unknown": 3,
+}
+
+SCORE_LABELS = {
+    1: "1（心情差）",
+    2: "2（不太好）",
+    3: "3（普通）",
+    4: "4（不錯）",
+    5: "5（超棒）",
+}
+SCORE_CHOICES = [SCORE_LABELS[i] for i in [1, 2, 3, 4, 5]]
+
+
+# =========================
+# 狀態
+# =========================
+@dataclass
+class AppState:
+    mode: str = "攝影機辨識"
+
+    running: bool = True
+    finished: bool = False
+    last_analyze_t: float = 0.0
+
+    # smoothing
+    ema_scores: Optional[np.ndarray] = None
+    emo_keys: List[str] = field(default_factory=list)
+
+    # stable/candidate
+    stable_emo: str = "unknown"
+    candidate_emo: Optional[str] = None
+    candidate_since: Optional[float] = None
+
+    # final
+    final_emo: Optional[str] = None
+    final_conf: Optional[float] = None
+    final_score: int = 3
+
+
+# =========================
+# UI HTML
+# =========================
+def _hint_html(msg: str) -> str:
+    return f"""
+    <div style="border-radius:14px;padding:14px;border:1px dashed #DADADA;background:#FAFAFA;color:#000000;">
+      {msg}
+    </div>
+    """
+
+def _camera_hint_before_start() -> str:
+    return _hint_html(
+        "請先按「Click to access webcam / 允許」開啟攝影機；"
+        "接著，請按「錄製（Record）」開始。"
+    )
+
+def _camera_hint_need_face() -> str:
+    return _hint_html(
+        "⚠️ 目前未偵測到人臉，請將臉部移至畫面中央、靠近鏡頭，並避免瀏海遮住眉眼，才可進行情緒辨識。"
+    )
+
+def _camera_hint_done() -> str:
+    return _hint_html("已完成辨識，可按「重新攝影機辨識」再次進行辨識。")
+
+def _camera_hint_stopped() -> str:
+    return _hint_html("已停止辨識。可按「重新攝影機辨識」再試一次。")
+
+def _score_radio_value(score: int) -> str:
+    score = int(score)
+    score = min(5, max(1, score))
+    return SCORE_LABELS[score]
+
+def _result_card_html(emo_key: str, conf: Optional[float]) -> str:
+    zh = EMO_ZH.get(emo_key, emo_key)
+    score = EMO_TO_SCORE.get(emo_key, 3)
+    conf_txt = "" if conf is None else f"{conf:.1f}%"
+    score_desc = SCORE_LABELS[score].split("（")[1].rstrip("）")
+
+    return f"""
+    <div style="
+        border-radius:16px;
+        padding:20px;
+        border:1px solid #E6E6E6;
+        background:#FFFFFF;
+        color:#000000;
+    ">
+      <div style="font-size:14px;color:#000000;margin-bottom:8px;">
+        辨識結果
+      </div>
+
+      <div style="
+        font-size:42px;
+        font-weight:800;
+        line-height:1.1;
+        margin-bottom:14px;
+        color:#000000;
+      ">
+        {zh}
+      </div>
+
+      <div style="font-size:16px;color:#000000;margin-bottom:6px;">
+        心情分數：{score}（{score_desc}）
+      </div>
+
+      <div style="font-size:14px;color:#000000;">
+        信心值：{conf_txt}
+      </div>
+    </div>
+    """
+
+
+# =========================
+# 影像處理（保命 + 顯示優化）
+# =========================
+def _to_rgb_uint8(frame: np.ndarray) -> np.ndarray:
+    """統一成 MediaPipe Tasks 最穩格式：RGB、uint8、0~255、contiguous。"""
+    if frame is None:
+        return frame
+
+    # RGBA -> RGB
+    if frame.ndim == 3 and frame.shape[2] == 4:
+        frame = frame[:, :, :3]
+
+    # 灰階 -> RGB
+    if frame.ndim == 2:
+        frame = np.stack([frame, frame, frame], axis=-1)
+
+    # dtype / range 修正
+    if frame.dtype != np.uint8:
+        mx = float(np.max(frame)) if frame.size else 0.0
+        if mx <= 1.5:  # 多半是 0~1
+            frame = (frame * 255.0).clip(0, 255).astype(np.uint8)
+        else:          # 多半是 0~255 但 float
+            frame = np.clip(frame, 0, 255).astype(np.uint8)
+
+    return np.ascontiguousarray(frame)
+
+def _zoom_center(frame: np.ndarray, zoom: float) -> np.ndarray:
+    if frame is None or zoom <= 1.0:
+        return frame
+    h, w = frame.shape[:2]
+    new_w = int(w / zoom)
+    new_h = int(h / zoom)
+    start_x = (w - new_w) // 2
+    start_y = (h - new_h) // 2
+    cropped = frame[start_y : start_y + new_h, start_x : start_x + new_w]
+    return np.ascontiguousarray(cropped)
+
+def _resize_for_ui(frame_rgb: np.ndarray, target_w: int = UI_TARGET_W) -> np.ndarray:
+    """回傳給 Gradio 的畫面縮到固定寬度，減少傳輸/渲染負擔。"""
+    if frame_rgb is None:
+        return frame_rgb
+    h, w = frame_rgb.shape[:2]
+    if w <= target_w:
+        return frame_rgb
+    scale = target_w / w
+    new_h = max(1, int(h * scale))
+    return cv2.resize(frame_rgb, (target_w, new_h), interpolation=cv2.INTER_AREA)
+
+
+# =========================
+# MediaPipe Tasks Face Detector
+# =========================
+BaseOptions = python.BaseOptions
+FaceDetector = vision.FaceDetector
+FaceDetectorOptions = vision.FaceDetectorOptions
+VisionRunningMode = vision.RunningMode
+
+_face_detector = FaceDetector.create_from_options(
+    FaceDetectorOptions(
+        base_options=BaseOptions(model_asset_path=MP_MODEL_PATH),
+        running_mode=VisionRunningMode.IMAGE,
+        min_detection_confidence=MP_MIN_DET_CONF,
+    )
+)
+
+def mp_tasks_detect_and_draw(frame_rgb: np.ndarray):
+    """
+    - 最大臉策略（bbox 面積最大）
+    - 有臉：畫白框 + 紅點 keypoints
+    - 回 bbox_px (x, y, w, h) 供 ROI 裁切
+    """
+    frame_rgb = _to_rgb_uint8(frame_rgb)
+    h, w = frame_rgb.shape[:2]
+
+    mp_image = mp.Image(image_format=mp.ImageFormat.SRGB, data=frame_rgb)
+    result = _face_detector.detect(mp_image)
+
+    if not result.detections:
+        return frame_rgb, None
+
+    best_det = None
+    best_area = -1.0
+    best_bbox = None
+
+    for det in result.detections:
+        bbox = det.bounding_box
+        area = float(bbox.width) * float(bbox.height)
+        if area > best_area:
+            best_area = area
+            best_det = det
+            best_bbox = bbox
+
+    if best_det is None or best_bbox is None:
+        return frame_rgb, None
+
+    lx = int(best_bbox.origin_x)
+    ly = int(best_bbox.origin_y)
+    bw = int(best_bbox.width)
+    bh = int(best_bbox.height)
+
+    lx = max(0, min(w - 1, lx))
+    ly = max(0, min(h - 1, ly))
+    bw = max(1, min(w - lx, bw))
+    bh = max(1, min(h - ly, bh))
+
+    bgr = cv2.cvtColor(frame_rgb, cv2.COLOR_RGB2BGR)
+    cv2.rectangle(bgr, (lx, ly), (lx + bw, ly + bh), (255, 255, 255), 3)
+
+    for kp in best_det.keypoints:
+        cx = int(kp.x * w)
+        cy = int(kp.y * h)
+        cv2.circle(bgr, (cx, cy), 6, (0, 0, 255), -1)
+
+    drawn = cv2.cvtColor(bgr, cv2.COLOR_BGR2RGB)
+    return drawn, (lx, ly, bw, bh)
+
+
+# =========================
+# 背景推論（避免卡頓） + ✅ 沒臉時清空
+# =========================
+_ANALYZE_LOCK = threading.Lock()
+_ANALYZE_THREAD: Optional[threading.Thread] = None
+_PENDING_FACE: Optional[np.ndarray] = None
+_LAST_ANALYZE: Dict[str, Optional[float]] = {"ts": 0.0, "conf": None, "emo": "unknown"}
+
+def _clear_last_analyze():
+    """✅ 沒有人臉時，清空上一筆情緒，避免殘留顯示（你影片遇到的 bug）"""
+    global _LAST_ANALYZE
+    _LAST_ANALYZE = {"ts": 0.0, "conf": None, "emo": "unknown"}
+
+def _analyze_worker():
+    global _PENDING_FACE, _ANALYZE_THREAD, _LAST_ANALYZE
+
+    while True:
+        with _ANALYZE_LOCK:
+            face = _PENDING_FACE
+            _PENDING_FACE = None
+
+        if face is None:
+            break
+
+        try:
+            r = DeepFace.analyze(
+                img_path=face,
+                actions=["emotion"],
+                enforce_detection=False,
+                detector_backend="skip",  # ⭐ 你已裁 ROI，別再偵測一次
+            )
+            if isinstance(r, list):
+                r = r[0]
+            emo_dict = r.get("emotion", None)
+
+            if isinstance(emo_dict, dict) and len(emo_dict) > 0:
+                top_key = max(emo_dict, key=lambda k: float(emo_dict[k]))
+                top_conf = float(emo_dict[top_key])
+                _LAST_ANALYZE = {"ts": time.time(), "emo": top_key, "conf": top_conf}
+        except Exception:
+            pass
+
+    _ANALYZE_THREAD = None
+
+def _enqueue_analyze(face_roi: np.ndarray):
+    global _PENDING_FACE, _ANALYZE_THREAD
+
+    with _ANALYZE_LOCK:
+        _PENDING_FACE = face_roi
+
+    if _ANALYZE_THREAD is None:
+        _ANALYZE_THREAD = threading.Thread(target=_analyze_worker, daemon=True)
+        _ANALYZE_THREAD.start()
+
+
+# =========================
+# state helper
+# =========================
+def _reset_state_for_camera(st: AppState) -> AppState:
+    st.running = True
+    st.finished = False
+    st.last_analyze_t = 0.0
+
+    st.ema_scores = None
+    st.emo_keys = []
+
+    st.stable_emo = "unknown"
+    st.candidate_emo = None
+    st.candidate_since = None
+
+    st.final_emo = None
+    st.final_conf = None
+    st.final_score = 3
+
+    _clear_last_analyze()
+    return st
+
+
+# =========================
+# streaming callback
+# =========================
+def on_stream(frame_rgb: np.ndarray, st: AppState):
+    # stopped / finished
+    if st.finished or (not st.running):
+        cam_visible = False
+
+        if st.final_emo:
+            result_html = _result_card_html(st.final_emo, st.final_conf)
+            mood_val = _score_radio_value(EMO_TO_SCORE.get(st.final_emo, 3))
+            hint_html = _camera_hint_done()
+        else:
+            result_html = _hint_html("尚未完成辨識。")
+            mood_val = _score_radio_value(3)
+            hint_html = _camera_hint_stopped()
+
+        return (
+            gr.update(value=None, visible=cam_visible),
+            gr.update(value=result_html),
+            gr.update(value=mood_val),
+            gr.update(value=hint_html),
+            st,
+            gr.update(value="重新攝影機辨識"),
+        )
+
+    # no frame yet
+    if frame_rgb is None:
+        return (
+            gr.update(value=None, visible=True),
+            gr.update(value=_hint_html("等待影像…（尚未開始串流）")),
+            gr.update(value=_score_radio_value(3)),
+            gr.update(value=_camera_hint_before_start()),
+            st,
+            gr.update(value="重新攝影機辨識"),
+        )
+
+    frame_rgb = _to_rgb_uint8(frame_rgb)
+
+    # ✅ 偵測用原圖
+    frame_drawn, bbox_px = mp_tasks_detect_and_draw(frame_rgb)
+
+    # ✅ 顯示用 zoom
+    frame_show = _zoom_center(frame_drawn if bbox_px is not None else frame_rgb, ZOOM_FACTOR)
+    frame_show = _resize_for_ui(frame_show, UI_TARGET_W)
+
+    # ✅ 沒有臉：清空殘留情緒 + 重置候選狀態
+    if bbox_px is None:
+        _clear_last_analyze()
+
+        st.stable_emo = "unknown"
+        st.candidate_emo = None
+        st.candidate_since = None
+        st.ema_scores = None
+        st.emo_keys = []
+
+        return (
+            gr.update(value=frame_show, visible=True),
+            gr.update(value=_hint_html("尚未偵測到人臉")),
+            gr.update(value=_score_radio_value(3)),
+            gr.update(value=_camera_hint_need_face()),
+            st,
+            gr.update(value="重新攝影機辨識"),
+        )
+
+    now = time.time()
+
+    # 到分析時間：只排隊給背景 thread
+    if now - st.last_analyze_t >= ANALYZE_EVERY_SEC:
+        st.last_analyze_t = now
+
+        x, y, bw, bh = bbox_px
+        pad_x = int(bw * 0.10)
+        pad_y = int(bh * 0.10)
+        H, W = frame_rgb.shape[:2]
+        x1 = max(0, x - pad_x)
+        y1 = max(0, y - pad_y)
+        x2 = min(W, x + bw + pad_x)
+        y2 = min(H, y + bh + pad_y)
+        face_roi = frame_rgb[y1:y2, x1:x2].astype(np.uint8)
+
+        _enqueue_analyze(face_roi)
+
+    # ✅ Gate：只有「有臉」才讀取 _LAST_ANALYZE（避免殘留）
+    top_key = _LAST_ANALYZE.get("emo", "unknown") or "unknown"
+    top_conf = _LAST_ANALYZE.get("conf", None)
+
+    # 尚未推論到結果：先顯示 unknown
+    if top_conf is None:
+        show = st.stable_emo or "unknown"
+        return (
+            gr.update(value=frame_show, visible=True),
+            gr.update(value=_result_card_html(show, None)),
+            gr.update(value=_score_radio_value(EMO_TO_SCORE.get(show, 3))),
+            gr.update(value=_camera_hint_before_start()),
+            st,
+            gr.update(value="重新攝影機辨識"),
+        )
+
+    # 低信心就不切換
+    if top_conf < CONF_TH:
+        show = st.stable_emo or "unknown"
+        return (
+            gr.update(value=frame_show, visible=True),
+            gr.update(value=_result_card_html(show, None)),
+            gr.update(value=_score_radio_value(EMO_TO_SCORE.get(show, 3))),
+            gr.update(value=_camera_hint_before_start()),
+            st,
+            gr.update(value="重新攝影機辨識"),
+        )
+
+    # === EMA smoothing（用 7 類別向量）===
+    all_keys = ["angry", "disgust", "fear", "happy", "sad", "surprise", "neutral"]
+    vec = np.zeros((len(all_keys),), dtype=np.float32)
+    if top_key in all_keys:
+        vec[all_keys.index(top_key)] = float(top_conf)
+
+    if st.ema_scores is None or st.emo_keys != all_keys:
+        st.ema_scores = vec
+        st.emo_keys = all_keys
+    else:
+        st.ema_scores = EMA_ALPHA * vec + (1.0 - EMA_ALPHA) * st.ema_scores
+
+    stable_idx = int(np.argmax(st.ema_scores))
+    stable_key = st.emo_keys[stable_idx] if st.emo_keys else "unknown"
+    st.stable_emo = stable_key
+
+    # 延遲確認，避免跳動
+    if st.candidate_emo != stable_key:
+        st.candidate_emo = stable_key
+        st.candidate_since = now
+    else:
+        if st.candidate_since is None:
+            st.candidate_since = now
+
+    confirmed = (
+        st.candidate_since is not None
+        and (now - st.candidate_since) >= SWITCH_CONFIRM_SEC
+    )
+
+    # ✅ 完成一次辨識後停止
+    if confirmed and st.final_emo is None:
+        st.final_emo = stable_key
+        st.final_conf = float(top_conf) if top_conf is not None else None
+        st.final_score = EMO_TO_SCORE.get(stable_key, 3)
+
+        st.running = False
+        st.finished = True
+
+        return (
+            gr.update(value=None, visible=False),
+            gr.update(value=_result_card_html(stable_key, st.final_conf)),
+            gr.update(value=_score_radio_value(EMO_TO_SCORE.get(stable_key, 3))),
+            gr.update(value=_camera_hint_done()),
+            st,
+            gr.update(value="重新攝影機辨識"),
+        )
+
+    return (
+        gr.update(value=frame_show, visible=True),
+        gr.update(value=_result_card_html(stable_key, float(top_conf))),
+        gr.update(value=_score_radio_value(EMO_TO_SCORE.get(stable_key, 3))),
+        gr.update(value=_camera_hint_before_start()),
+        st,
+        gr.update(value="重新攝影機辨識"),
+    )
+
+
+# =========================
+# Buttons
+# =========================
+def on_restart(st: AppState):
+    st = _reset_state_for_camera(st)
+    return (
+        gr.update(value=None, visible=True),
+        gr.update(value=_hint_html("等待影像…（尚未開始串流）")),
+        gr.update(value=_score_radio_value(3)),
+        gr.update(value=_camera_hint_before_start()),
+        st,
+        gr.update(value="重新攝影機辨識"),
+    )
+
+def on_stop(st: AppState):
+    st.running = False
+    st.finished = True
+
+    if st.final_emo is None:
+        return (
+            gr.update(value=None, visible=False),
+            gr.update(value=_hint_html("已停止辨識。")),
+            gr.update(value=_score_radio_value(3)),
+            gr.update(value=_camera_hint_stopped()),
+            st,
+            gr.update(value="重新攝影機辨識"),
+        )
+
+    return (
+        gr.update(value=None, visible=False),
+        gr.update(value=_result_card_html(st.final_emo, st.final_conf)),
+        gr.update(value=_score_radio_value(EMO_TO_SCORE.get(st.final_emo, 3))),
+        gr.update(value=_camera_hint_done()),
+        st,
+        gr.update(value="重新攝影機辨識"),
+    )
+
+
+# =========================
+# Gradio UI
+# =========================
+css = ""
+
+if __name__ == "__main__":
+    with gr.Blocks(title="Emotion Detector", css=css) as demo:
+        st = gr.State(AppState())
+
+        gr.Markdown("## 情緒辨識")
+        hint = gr.HTML(_camera_hint_before_start())
+
+        cam = gr.Image(
+            sources=["webcam"],
+            streaming=True,
+            type="numpy",
+            label="攝影機畫面",
+            visible=True,
+        )
+
+        with gr.Row():
+            btn_restart = gr.Button("重新攝影機辨識", variant="primary")
+            btn_stop = gr.Button("停止", variant="secondary")
+
+        gr.Markdown("### 辨識結果")
+        result = gr.HTML(_hint_html("等待影像…（尚未開始串流）"))
+
+        gr.Markdown("### 心情分數（系統判定）")
+        mood = gr.Radio(
+            choices=SCORE_CHOICES,
+            value=_score_radio_value(3),
+            label="心情分數",
+            interactive=False,
+        )
+
+        cam.stream(
+            fn=on_stream,
+            inputs=[cam, st],
+            outputs=[cam, result, mood, hint, st, btn_restart],
+            show_progress="minimal",
+        )
+
+        btn_restart.click(
+            fn=on_restart,
+            inputs=[st],
+            outputs=[cam, result, mood, hint, st, btn_restart],
+            show_progress="minimal",
+        )
+
+        btn_stop.click(
+            fn=on_stop,
+            inputs=[st],
+            outputs=[cam, result, mood, hint, st, btn_restart],
+            show_progress="minimal",
+        )
+
+    demo.launch(server_name="0.0.0.0", server_port=7860, share=False)

mood_food_guide.csv

@@ -0,0 +1,6 @@
+分數,心情狀態,需求,推薦料理類別,原因
+1,糟透了,"極度需要安慰、溫暖、擁抱感，或者是快速的多巴胺",日式料理(拉麵/壽喜燒/關東煮)、甜點/點心,"當心情跌落谷底，溫暖軟糯的碳水化合物像擁抱；糖分能快速上升血糖帶來短暫快樂，是「人生好難」的解藥。"
+2,糟糕,"發洩壓力、需要刺激、想要大口吃肉或重口味轉移注意力",韓式料理(炸雞/年糕/部隊鍋)、泰式料理,"辣味帶來的痛覺能刺激腦內啡分泌，酸辣重口能瞬間醒腦，暫時沖刷煩惱。"
+3,普通,"維持日常機能、追求安穩、健康、不想踩雷也不想太負擔",台式/中式(滷肉飯/水餃/乾麵)、健康餐(水煮/波奇碗),"這是刻在基因裡的安全感味道；或是選擇清淡原型食物，作為一種「自我保養」。"
+4,不錯,"有些許興致、願意嘗試豐富的香料、適合輕鬆的社交",港式料理、印度料理/馬來西亞料理,"適合花時間慢慢吃的點心，或欣賞香料層次豐富的異國風味，適合心情好且感官敏銳時享用。"
+5,太棒了,"慶祝、獎勵自己、高熱量、高爽度、大份量分享",美式/西式(牛排/漢堡/披薩/義大利麵),"標準的「慶祝食物」，伴隨著高熱量和飽足感，是大腦認定最高級的「獎勵」。"

packages.txt

@@ -0,0 +1,16 @@
+libnss3
+libnspr4
+libatk1.0-0t64
+libatk-bridge2.0-0t64
+libcups2t64
+libdrm2
+libxkbcommon0
+libxcomposite1
+libxdamage1
+libxext6
+libxfixes3
+libxrandr2
+libgbm1
+libasound2t64
+libpango-1.0-0
+libcairo2

requirements.txt

@@ -0,0 +1,49 @@
+# =========================================
+# 1. 核心與資料處理 (Core & Data)
+# =========================================
+python-dotenv
+requests
+numpy
+pandas
+
+# =========================================
+# 2. 使用者介面 (UI Framework)
+# =========================================
+# 採用您指定的版本，確保組員的 emotion.py 樣式相容
+gradio==4.44.1
+
+# =========================================
+# 3. 網路爬蟲 (Web Scraping)
+# =========================================
+playwright
+
+# =========================================
+# 4. RAG 與 LLM 應用 (AI Logic)
+# =========================================
+openai
+huggingface_hub
+langchain
+langchain-community
+langchain-huggingface
+sentence-transformers
+# 向量資料庫 (務必使用 CPU 版)
+faiss-cpu
+
+# =========================================
+# 5. 情緒辨識與影像處理 (Computer Vision)
+# =========================================
+# emotion.py 依賴的核心庫
+deepface==0.0.92
+mtcnn==0.1.1
+pillow==10.4.0
+mediapipe
+
+# 雲端環境必須使用 headless 版本，避免 GUI 錯誤
+opencv-python-headless==4.10.0.84
+
+# =========================================
+# 6. 深度學習底層 (Deep Learning Backend)
+# =========================================
+tensorflow==2.15.1
+tf-keras==2.15.1
+keras==2.15.0