Update app.py

app.py

@@ -1,136 +1,128 @@
 import gradio as gr
 import re
+import matplotlib.pyplot as plt
+import numpy as np
 from transformers import pipeline
 
 # ---------------------------
-# Transformer 翻译模型
+# Translator Model
 # ---------------------------
 translator = pipeline("translation", model="Helsinki-NLP/opus-mt-zh-en")
 
-
 def translate_text(cn_text):
     try:
-        result = translator(cn_text)[0]["translation_text"]
-        return result
+        return translator(cn_text)[0]["translation_text"]
     except:
         return "Translation model error."
 
 
 # ---------------------------
-# 礼貌标记自动抽取（英文）
+# Politeness Markers Extraction
 # ---------------------------
-def extract_politeness_markers(en_text):
-    markers = ["please", "could", "would", "possibly", "may", "kindly"]
-    found = [m for m in markers if m in en_text.lower()]
-    return ", ".join(found) if found else "无明显显性礼貌标记"
+POLITENESS_FEATURES = {
+    "modals": ["could", "would", "may", "might"],
+    "softener": ["please", "kindly"],
+    "hedging": ["possibly", "perhaps", "a bit", "a little"],
+    "formal": ["professor", "sir", "madam"],
+    "passive": ["is requested", "may cause", "is appreciated"],
+    "appreciation": ["thank", "appreciate", "grateful"]
+}
+
+def extract_politeness_scores(en_text):
+    text = en_text.lower()
+    scores = {}
+
+    for feat, markers in POLITENESS_FEATURES.items():
+        score = sum(text.count(m) for m in markers)
+        score = min(score, 5)  # cap at 5 for radar chart
+        scores[feat] = score
+
+    return scores
 
 
 # ---------------------------
-# TIRA 判定逻辑（强化版）
-# 基于中文意图 + 英文特征双通道判定
+# TIRA Classification
 # ---------------------------
 def classify_tira(cn_text, en_text):
     cn = cn_text.strip()
-    en = en_text.lower().strip()
+    en = en_text.lower()
 
-    # ---- 中文触发条件 ----
     cn_amplify = ["能否", "是否方便", "请您帮忙", "抽空", "劳驾", "麻烦帮我"]
     cn_attenuate = ["尽快", "必须", "需要您", "立即"]
     cn_redirect = ["给您添麻烦", "不好意思打扰", "不胜感激"]
 
-    # ---- 英文特征 ----
-    en_amplify_markers = ["please", "possibly", "could you", "would you"]
-    en_attenuate_markers = ["directly", "please do", "must", "asap"]
-    en_passive_nominal = ["inconvenience", "may cause", "it is requested"]
-
-    # ---- 双通道判断 ----
+    en_amplify = ["please", "could you", "possibly", "would you"]
+    en_attenuate = ["directly", "must", "asap"]
+    en_redirect = ["inconvenience", "may cause"]
 
-    # Redirect — 特殊中文敬辞 or 英文被动/名词化
-    if any(p in cn for p in cn_redirect) or any(p in en for p in en_passive_nominal):
-        return "Redirect（转向）", "中文情感敬辞或英文被动/名词化导致策略位移 → 方向不确定。"
+    if any(p in cn for p in cn_redirect) or any(p in en for p in en_redirect):
+        return "Redirect（转向）", "中文敬辞或英文名词化导致策略转向 → 方向不定。"
 
-    # Amplify — 英文显性礼貌标记上升
-    if any(p in cn for p in cn_amplify) or any(p in en for p in en_amplify_markers):
-        return "Amplify（放大）", "英文加入显性礼貌 markers，如 could / please / possibly → PDI 上移。"
+    if any(p in cn for p in cn_amplify) or any(p in en for p in en_amplify):
+        return "Amplify（放大）", "英文显性礼貌增强 → PDI 上移。"
 
-    # Attenuate — 英文更直接
-    if any(p in cn for p in cn_attenuate) or any(p in en for p in en_attenuate_markers):
-        return "Attenuate（削弱）", "英文更直接或缺乏缓和手段 → PDI 下移。"
+    if any(p in cn for p in cn_attenuate) or any(p in en for p in en_attenuate):
+        return "Attenuate（削弱）", "英文更直接 → PDI 下移。"
 
-    # Retain — 默认
-    return "Retain（保留）", "英文基本保持中文的礼貌策略 → PDI 稳定。"
+    return "Retain（保留）", "礼貌构式基本被稳定保留。"
 
 
 # ---------------------------
-# 主函数
+# Radar Chart Drawing
 # ---------------------------
-def predict_tira(cn_text):
-    # 1. 翻译
-    en_trans = translate_text(cn_text)
+def draw_radar(scores_dict):
+    labels = list(scores_dict.keys())
+    scores = list(scores_dict.values())
+    
+    angles = np.linspace(0, 2 * np.pi, len(labels), endpoint=False).tolist()
+    scores += scores[:1]
+    angles += angles[:1]
+
+    fig, ax = plt.subplots(figsize=(5, 5), subplot_kw=dict(polar=True))
 
-    # 2. 抽取礼貌
-    markers = extract_politeness_markers(en_trans)
+    ax.fill(angles, scores, color="#1B3B6F80", alpha=0.4)
+    ax.plot(angles, scores, color="#1B3B6F", linewidth=2)
 
-    # 3. TIRA 分类
-    tira_type, analysis = classify_tira(cn_text, en_trans)
+    ax.set_xticks(angles[:-1])
+    ax.set_xticklabels([label.capitalize() for label in labels], fontsize=10)
+    ax.set_yticks(range(1, 6))
+    ax.set_yticklabels([str(i) for i in range(1, 6)], fontsize=8)
+    ax.set_ylim(0, 5)
+    ax.grid(True)
 
-    # 4. 风险
-    risk_map = {
+    return fig
+
+
+# ---------------------------
+# Main Function
+# ---------------------------
+def full_process(cn_text):
+    en = translate_text(cn_text)
+    scores = extract_politeness_scores(en)
+    tira, analysis = classify_tira(cn_text, en)
+
+    risk = {
         "Retain": "低风险（关系保持稳定）",
-        "Amplify": "中风险（英文更显远礼貌）",
-        "Attenuate": "中偏高风险（可能显得更直接）",
-        "Redirect": "不稳定（误解风险较高）"
+        "Amplify": "中风险（可能显得更有距离）",
+        "Attenuate": "中偏高（可能显得更直接）",
+        "Redirect": "不稳定（可能造成理解偏差）"
     }
-    key = tira_type.split("（")[0]
-    pdi_risk = risk_map.get(key, "低风险")
+    risk_out = risk[tira.split("（")[0]]
+    radar = draw_radar(scores)
 
-    return en_trans, markers, tira_type, analysis, pdi_risk
+    return en, scores, tira, analysis, risk_out, radar
 
 
 # ---------------------------
-# 深蓝学术 UI
+# UI (Deep Blue Academic Style)
 # ---------------------------
 css = """
-h1 {text-align: center; color: #1B3B6F;}
-.gradio-container {background-color: #F7F9FC;}
-label {font-weight: bold;}
+h1 {text-align:center; color:#1B3B6F;}
+.gradio-container {background-color:#F6F8FA;}
+label {font-weight:bold;}
 """
 
-with gr.Blocks(css=css, title="TIRA Translator") as demo:
-
+with gr.Blocks(css=css, title="TIRA Radar System") as demo:
     gr.Markdown("""
-# **📘 TIRA Academic Email Translation Analysis**
-### *Real Translation + Politeness Feature Extraction + TIRA Classification*
----
-
-输入中文学术邮件 → 系统将：
-1. 使用 Transformer 翻译为英文  
-2. 自动抽取礼貌标记  
-3. 判定 TIRA 类型  
-4. 评估 PDI 风险  
-5. 给出翻译偏移解释
-
----
-    """)
-
-    cn_input = gr.Textbox(
-        label="✉ 输入中文邮件内容",
-        lines=5,
-        placeholder="例如：能否请您帮我看一下附件？"
-    )
-
-    btn = gr.Button("🔍 翻译与 TIRA 预测", variant="primary")
-
-    en_output = gr.Textbox(label="🌐 英文翻译")
-    marker_output = gr.Textbox(label="🔎 英文中的礼貌标记")
-    tira_output = gr.Textbox(label="📌 TIRA 类型")
-    analysis_output = gr.Textbox(label="📘 解释说明")
-    risk_output = gr.Textbox(label="⚠ PDI 风险")
-
-    btn.click(
-        fn=predict_tira,
-        inputs=cn_input,
-        outputs=[en_output, marker_output, tira_output, analysis_output, risk_output]
-    )
-
-demo.launch()
+# **📘 TIRA Academic Email Translation Visualizer**
+### *Real-Time Translation + Politeness Feature Radar + TIRA Classificatio*