create app.py

app.py

@@ -0,0 +1,68 @@
+# app.py
+from flask import Flask, request, jsonify
+import gradio as gr
+from transformers import pipeline
+import pandas as pd
+import uuid
+import os
+
+# 初始化 Flask（仅用于API）
+flask_app = Flask(__name__)
+os.makedirs("static", exist_ok=True)
+
+# 直接使用 Hugging Face 模型（无需本地存储）
+model_name = "uer/roberta-base-finetuned-jd-binary-chinese"  # 使用Hub上的模型
+classifier = pipeline("text-classification", model=model_name)
+
+@flask_app.route("/api/predict", methods=["POST"])
+def api_predict():
+    text = request.json.get("text", "")
+    result = classifier(text)
+    return jsonify(result)
+
+# Gradio 界面
+with gr.Blocks(title="情感分析系统") as demo:
+    with gr.Tab("单句分析"):
+        input_text = gr.Textbox(label="输入文本")
+        analyze_btn = gr.Button("分析")
+        output_label = gr.Label(label="结果")
+
+        def predict(text):
+            result = classifier(text)[0]
+            return {result["label"]: result["score"]}
+        
+        analyze_btn.click(
+            fn=predict,
+            inputs=input_text,
+            outputs=output_label
+        )
+
+    with gr.Tab("批量分析"):
+        file_input = gr.File(label="上传TXT文件")
+        batch_output = gr.Dataframe(headers=["文本", "标签", "置信度"])
+        process_btn = gr.Button("处理文件")
+
+        def process_file(file):
+            with open(file.name, "r", encoding="utf-8") as f:
+                texts = [line.strip() for line in f if line.strip()]
+            results = []
+            for text in texts:
+                pred = classifier(text)[0]
+                results.append([text, pred["label"], f"{pred['score']:.4f}"])
+            return pd.DataFrame(results, columns=["文本", "标签", "置信度"])
+        
+        process_btn.click(
+            fn=process_file,
+            inputs=file_input,
+            outputs=batch_output
+        )
+
+# 启动 Flask 线程（仅用于API）
+import threading
+threading.Thread(
+    target=flask_app.run,
+    kwargs={"port": 8000, "host": "0.0.0.0"}
+).start()
+
+# 启动 Gradio
+demo.launch(server_port=7860)