我有一個 df，其中有一列 Critic_Score，該列具有 NaN 值。我試圖用同一平臺的評論家分數的平均值來替換它們。這個問題已經在堆棧溢出上被問過好幾次了，我使用了 4 個建議，但沒有給我想要的輸出。請告訴我如何解決這個問題。這是 df 的子集：x[['Platform','Critic_Score']].head()Platform    Critic_Score0   wii 76.01   nes NaN2   wii 82.03   wii 80.04   gb  NaN有關原始 df 的更多信息：x.head().to_dict('list'){'Name': ['wii sports',  'super mario bros.',  'mario kart wii',  'wii sports resort',  'pokemon red/pokemon blue'], 'Platform': ['wii', 'nes', 'wii', 'wii', 'gb'], 'Year_of_Release': [2006.0, 1985.0, 2008.0, 2009.0, 1996.0], 'Genre': ['sports', 'platform', 'racing', 'sports', 'role-playing'], 'NA_sales': [41.36, 29.08, 15.68, 15.61, 11.27], 'EU_sales': [28.96, 3.58, 12.76, 10.93, 8.89], 'JP_sales': [3.77, 6.81, 3.79, 3.28, 10.22], 'Other_sales': [8.45, 0.77, 3.29, 2.95, 1.0], 'Critic_Score': [76.0, nan, 82.0, 80.0, nan], 'User_Score': ['8', nan, '8.3', '8', nan], 'Rating': ['E', nan, 'E', 'E', nan]}這些是我嘗試過的語句及其輸出：1.x['Critic_Score'] = x['Critic_Score'].fillna(x.groupby('Platform')['Critic_Score'].transform('mean'), inplace = True)0    None1    None2    None3    None4    NoneName: Critic_Score, dtype: objectx.loc[x.Critic_Score.isnull(), 'Critic_Score'] = x.groupby('Platform').Critic_Score.transform('mean')#no change in column0    76.01     NaN2    82.03    80.04     NaNx['Critic_Score'] = x.groupby('Platform')['Critic_Score']\    .transform(lambda y: y.fillna(y.mean()))#no change in column0    76.01     NaN2    82.03    80.04     NaNName: Critic_Score, dtype: float64x['Critic_Score']=x.groupby('Platform')['Critic_Score'].apply(lambda y:y.fillna(y.mean()))x['Critic_Score'].head()Out[73]:0    76.01     NaN2    82.03    80.04     NaNName: Critic_Score, dtype: float64