将Python中具有混合类型的值列表附加到数据框

I have a data frame reading from csv using pandas.read_csv, each row of data frame looks like this:

我使用pandas.read_csv从csv读取数据帧,每行数据框如下所示:

[1, '10/18/2016 06:00', 1, 14, 0, 5.5]

Basically, it consists of integers, string, and floats. Now, I want to generate more data (newrow) based on existing data and append to the original data frame. When I try to call function append with the following code:

基本上,它由整数,字符串和浮点数组成。现在,我想基于现有数据生成更多数据(newrow)并附加到原始数据框。当我尝试使用以下代码调用函数append时:

df.append(list(newrow))

I got the error:

我收到了错误:

RuntimeWarning: unorderable types: str() < int(), sort order is undefined for incomparable objects. 
result = result.union(other)

I think the string type is playing naughty here, but I did not figure out a way to achieve this.

我认为字符串类型在这里玩顽皮,但我没有找到实现这一目标的方法。

In addition, I also tried to convert the df to df.values first, and then use numpy.vstack(df.values, numpy.array(newrow)). However, the result of this code becomes

另外,我还尝试首先将df转换为df.values,然后使用numpy.vstack(df.values,numpy.array(newrow))。但是,此代码的结果变为

['1', '10/18/2016 06:00', '1', '14', '0', '5.5']

in which all fields become strings. Any help is appreciated.

其中所有字段都成为字符串。任何帮助表示赞赏。

2 个解决方案

#1

Will this work?

这会有用吗?

#build a DataFrame first using the list and column names from original DF.
df.append(pd.DataFrame(columns=df.columns,data=[newrow]))

#2

This seems to work to append to an empty Dataframes. The columns must match.

这似乎可以附加到空的Dataframes。列必须匹配。

column_titles = ['col1','col2']
data = ['1', '10/18/2016 06:00', '1', '14', '0', '5.5']

# creates empty df w/headers
df = pd.DataFrame(columns = column_titles)

df2 = pd.DataFrame([data])

df = df.append(df2)

#1