Update app.py

app.py

@@ -1,10 +1,6 @@
 from transformers import pipeline, AutoTokenizer, AutoModelForCausalLM
 import torch
 import streamlit as st
-from PIL import Image
-import pytesseract
-import pandas as pd
-import plotly.express as px
 
 # ✅ Step 1: Emoji 翻译模型（你自己训练的模型）
 emoji_model_id = "jenniferhk008/roberta-hfl-emoji-aug3epoch"
@@ -26,19 +22,26 @@ model_options = {
 # ✅ 页面配置
 st.set_page_config(page_title="Emoji Offensive Text Detector", page_icon="🚨", layout="wide")
 
-# ✅ 侧边栏：模型选择
-with st.sidebar:
-    st.header("🧠 Configuration")
-    selected_model = st.selectbox("Choose classification model", list(model_options.keys()))
-    selected_model_id = model_options[selected_model]
-    classifier = pipeline("text-classification", model=selected_model_id, device=0 if torch.cuda.is_available() else -1)
+# ✅ 页面标题
+st.title("🧠 Emoji-based Offensive Language Classifier")
 
-# 初始化历史记录
-if "history" not in st.session_state:
-    st.session_state.history = []
+st.markdown("""
+This application translates emojis in a sentence and classifies whether the final sentence is offensive or not using two AI models.
+- The **first model** translates emoji or symbolic phrases into standard Chinese text.
+- The **second model** performs offensive language detection.
+""")
 
-# 分类函数
+# Streamlit 侧边栏模型选择
+selected_model = st.sidebar.selectbox("Choose classification model", list(model_options.keys()))
+selected_model_id = model_options[selected_model]
+classifier = pipeline("text-classification", model=selected_model_id, device=0 if torch.cuda.is_available() else -1)
 
+# ✅ 输入区域
+st.markdown("### ✍️ Input your sentence:")
+default_text = "你是🐷"
+text = st.text_area("Enter sentence with emojis:", value=default_text, height=150)
+
+# ✅ 主逻辑封装函数
 def classify_emoji_text(text: str):
     prompt = f"输入：{text}\n输出："
     input_ids = emoji_tokenizer(prompt, return_tensors="pt").to(emoji_model.device)
@@ -50,74 +53,21 @@ def classify_emoji_text(text: str):
     result = classifier(translated_text)[0]
     label = result["label"]
     score = result["score"]
-    reasoning = f"The sentence was flagged as '{label}' due to potentially offensive phrases. Consider replacing emotionally charged, ambiguous, or abusive terms."
 
-    st.session_state.history.append({"text": text, "translated": translated_text, "label": label, "score": score, "reason": reasoning})
-    return translated_text, label, score, reasoning
+    return translated_text, label, score
 
-# 主页面：输入与分析共存
-st.title("🚨 Emoji Offensive Text Detector & Analysis Dashboard")
-
-# 文本输入
-st.subheader("1. 输入与分类")
-default_text = "你是🐷"
-text = st.text_area("Enter sentence with emojis:", value=default_text, height=150)
-
-if st.button("🚦 Analyze Text"):
+# ✅ 触发按钮
+if st.button("🚦 Analyze"):
     with st.spinner("🔍 Processing..."):
         try:
-            translated, label, score, reason = classify_emoji_text(text)
-            st.markdown("**Translated sentence:**")
+            translated, label, score = classify_emoji_text(text)
+            st.markdown("### 🔄 Translated sentence:")
             st.code(translated, language="text")
-            st.markdown(f"**Prediction:** {label}")
-            st.markdown(f"**Confidence Score:** {score:.2%}")
-            st.markdown("**Model Explanation:**")
-            st.info(reason)
-        except Exception as e:
-            st.error(f"❌ An error occurred:\n{e}")
 
-# 图片上传与 OCR
-st.markdown("---")
-st.subheader("2. Image OCR")
-uploaded_file = st.file_uploader("Upload an image (JPG/PNG)", type=["jpg","jpeg","png"])
-if uploaded_file:
-    image = Image.open(uploaded_file)
-    st.image(image, caption="Uploaded Screenshot", use_column_width=True)
-    with st.spinner("🧠 Extracting text via OCR..."):
-        ocr_text = pytesseract.image_to_string(image, lang="chi_sim+eng").strip()
-        if ocr_text:
-            st.markdown("**Extracted Text:**")
-            st.code(ocr_text)
-            translated, label, score, reason = classify_emoji_text(ocr_text)
-            st.markdown("**Translated sentence:**")
-            st.code(translated, language="text")
-            st.markdown(f"**Prediction:** {label}")
-            st.markdown(f"**Confidence Score:** {score:.2%}")
-            st.markdown("**Model Explanation:**")
-            st.info(reason)
-        else:
-            st.info("⚠️ No text detected in the image.")
+            st.markdown(f"### 🎯 Prediction: `{label}`")
+            st.markdown(f"### 📊 Confidence Score: `{score:.2%}`")
 
-# 分析仪表盘
-st.markdown("---")
-st.subheader("3. Violation Analysis Dashboard")
-if st.session_state.history:
-    # 展示历史记录
-    df = pd.DataFrame(st.session_state.history)
-    st.markdown("### 🧾 Offensive Terms & Suggestions")
-    for item in st.session_state.history:
-        st.markdown(f"- 🔹 **Input:** {item['text']}")
-        st.markdown(f"   - ✨ **Translated:** {item['translated']}")
-        st.markdown(f"   - ❗ **Label:** {item['label']} with **{item['score']:.2%}** confidence")
-        st.markdown(f"   - 🔧 **Suggestion:** {item['reason']}")
-
-    # 雷达图
-    radar_df = pd.DataFrame({
-        "Category": ["Insult","Abuse","Discrimination","Hate Speech","Vulgarity"],
-        "Score": [0.7,0.4,0.3,0.5,0.6]
-    })
-    radar_fig = px.line_polar(radar_df, r='Score', theta='Category', line_close=True, title="⚠️ Risk Radar by Category")
-    radar_fig.update_traces(line_color='black')
-    st.plotly_chart(radar_fig)
+        except Exception as e:
+            st.error(f"❌ An error occurred during processing:\n\n{e}")
 else:
-    st.info("⚠️ No classification data available yet.")
+    st.info("👈 Please input text and click the button to classify.")