擅长:python、mysql、java
<p>你需要这样的东西</p>
<pre><code>df1 = pd.read_csv('CSV-1.csv')
df2 = pd.read_csv('CSV-2.csv')
</code></pre>
<p>合并两个数据帧</p>
<pre><code>df3 = df1.merge(df2, on="filename", how="outer",suffixes=('-1', '-2'))
</code></pre>
<p>删除更改次数相同的行</p>
<pre><code>df3 = df3[df3['number_of_changes-1'] != df3['number_of_changes-2']]
</code></pre>
<p>用0填充NAs并按文件名排序</p>
<pre><code>df3.fillna(0, inplace=True)
df3 = df3.sort_values(by ='filename').reset_index(drop=True)
</code></pre>
<p>输出:</p>
<pre><code> filename number_of_changes-1 number_of_changes-2
0 B 10.0 15.0
1 C 10.0 0.0
2 D 0.0 30.0
3 E 5.0 10.0
4 F 15.0 0.0
</code></pre>