Update app.py

app.py

@@ -2,39 +2,27 @@ import pandas as pd
 import gradio as gr
 
 def compare_csv_files():
-    # 文件名
     file1 = "fish-speech-1.5.csv"
     file2 = "fish-speech-1.4.csv"
     
-    # 读取 CSV 文件
     df1 = pd.read_csv(file1)
     df2 = pd.read_csv(file2)
     
-    # 使用 SourceText 进行合并
-    merged_df = pd.merge(df1, df2, on="SourceText", suffixes=("_1", "_2"))
+    merged_df = pd.merge(df1, df2, on="SourceText", suffixes=("_1.5", "_1.4"))
+    merged_df["WordErrorRate_Diff"] = merged_df["WordErrorRate_1.5"] - merged_df["WordErrorRate_1.4"]
+    merged_df["CharacterErrorRate_Diff"] = merged_df["CharacterErrorRate_1.5"] - merged_df["CharacterErrorRate_1.4"]
     
-    if merged_df.empty:
-        return "两个文件中没有相同的 SourceText，请检查数据。"
-    
-    # 计算差异
-    merged_df["WordErrorRate_Diff"] = merged_df["WordErrorRate_1"] - merged_df["WordErrorRate_2"]
-    merged_df["CharacterErrorRate_Diff"] = merged_df["CharacterErrorRate_1"] - merged_df["CharacterErrorRate_2"]
-    
-    # 返回对比结果
     comparison = merged_df[[
         "SourceText", 
-        "UUID_1", "WhisperText_1", "WordErrorRate_1", "CharacterErrorRate_1",
-        "UUID_2", "WhisperText_2", "WordErrorRate_2", "CharacterErrorRate_2",
+        "WhisperText_1.5", "WordErrorRate_1.5", "CharacterErrorRate_1.5",
+        "WhisperText_1.4", "WordErrorRate_1.4", "CharacterErrorRate_1.4",
         "WordErrorRate_Diff", "CharacterErrorRate_Diff"
     ]]
     
     return comparison.to_html(index=False)
 
-# Gradio 界面
 gr.Interface(
     fn=compare_csv_files,
     inputs=None,
-    outputs="html",
-    title="CSV 文件对比工具",
-    description="自动加载目录下的 fish-speech-1.5.csv 和 fish-speech-1.4.csv，对比它们的 WordErrorRate 和 CharacterErrorRate 差异。",
+    outputs="html"
 ).launch()