如果一个值为NaN，则熊猫用NaN替换连续的所有项目

我想摆脱一些使用NaN的记录.效果很好:

I want to get rid of some records with NaNs. This works perfectly:

df.dropna(axis=0, how='any',inplace=True)

但是，它改变了我的数据框的形状，并且索引不再均匀地间隔.因此，我想用np.nan替换这些行中的所有项目.有没有简单的方法可以做到这一点?

However, it changes the shape of my dataframe, and the index is no longer uniformly spaced. Therefore, I'd like to replace all items in these rows with np.nan. Is there a simple way to do this?

我正在考虑在dropna之后重新采样数据帧，但这似乎只能以规定的间隔工作，而我宁愿使用原始索引.另一种方法是使用iterrows遍历数据帧，但这也很麻烦.

I was thinking about resampling the dataframe after dropna, but that only seems to work with a prescribed interval, whereas I would rather use the original index. Another approach would be to loop over the dataframe with iterrows, but that also feels cumbersome.