feature request: displaying lots of columns

[vgoklani]

Hi,

What's the plan for displaying head() cleanly on my dataframe when there are lots of columns?

R handle's this easily by wrapping the output across multiple lines.

Thanks,

Vishal

[lodagro]

You have two options;

• print df.to_string()
• disable the terminal size detection and the automatic switch over between full and info view e.g. pd.set_printoptions(max_columns=50).

see also:

• http://pandas.pydata.org/pandas-docs/stable/basics.html#console-output-formatting
• http://stackoverflow.com/questions/11707586/python-pandas-widen-output-display/11711637#11711637

In [41]: df = pd.DataFrame(np.random.randn(2, 20))

In [42]: df
Out[42]: 
<class 'pandas.core.frame.DataFrame'>
Int64Index: 2 entries, 0 to 1
Data columns:
0     2  non-null values
1     2  non-null values
2     2  non-null values
3     2  non-null values
4     2  non-null values
5     2  non-null values
6     2  non-null values
7     2  non-null values
8     2  non-null values
9     2  non-null values
10    2  non-null values
11    2  non-null values
12    2  non-null values
13    2  non-null values
14    2  non-null values
15    2  non-null values
16    2  non-null values
17    2  non-null values
18    2  non-null values
19    2  non-null values
dtypes: float64(20)

In [43]: print df.to_string()
         0         1         2         3         4         5         6         7         8         9         10        11        12        13        14        15        16        17        18        19
0 -1.120988 -0.014668  1.005053 -1.171275  0.819571  1.376871  0.192588  0.052807 -0.847959 -1.220135  1.367997 -1.302833  1.717463 -0.306201 -0.037465 -0.067757 -1.137955 -0.508188 -0.385194  1.289037
1 -0.348980 -0.538781  0.903626 -1.481493 -1.191302 -0.509413 -0.863830 -2.581788 -0.965519  0.152339  0.039064 -0.339086 -0.114939 -1.277665  0.537978 -1.072551 -0.772296 -0.719904  0.819310 -0.422032

In [44]: pd.set_printoptions(max_columns=50)

In [45]: df
Out[45]: 
         0         1         2         3         4         5         6         7         8         9         10        11        12        13        14        15        16        17        18        19
0 -1.120988 -0.014668  1.005053 -1.171275  0.819571  1.376871  0.192588  0.052807 -0.847959 -1.220135  1.367997 -1.302833  1.717463 -0.306201 -0.037465 -0.067757 -1.137955 -0.508188 -0.385194  1.289037
1 -0.348980 -0.538781  0.903626 -1.481493 -1.191302 -0.509413 -0.863830 -2.581788 -0.965519  0.152339  0.039064 -0.339086 -0.114939 -1.277665  0.537978 -1.072551 -0.772296 -0.719904  0.819310 -0.422032

[wesm]

I'd be willing to basically "take the R approach" of wrapping columns. Have to take a look

[vgoklani]

I've been using the approach mentioned above, but obviously R's implementation is much easier to read for high-dimensional datasets.

Thanks!
On Oct 20, 2012, at 2:32 PM, Wes McKinney notifications@github.com wrote:

I'd be willing to basically "take the R approach" of wrapping columns. Have to take a look

—
Reply to this email directly or view it on GitHub.

[ghost]

@changhiskhan , wide repr closed this?

[changhiskhan]

Yup go ahead and close
On Dec 11, 2012 12:30 AM, "y-p" notifications@github.com wrote:

@changhiskhan https://github.com/changhiskhan , wide repr closed this?

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/2076#issuecomment-11231780.

[vgoklani]

I saw the ipython notebook on twitter, the update is pretty slick! Just two quick questions:

1. what's the distinction between set_option and set_printoptions? why do you have two different setters?
2. is there a way to auto-set the line_width to match the current window size of the terminal?

Thanks!

Vishal

[ghost]

1. for back-compatibility. If you use set_printoption now, there's a deprecation warning.
2. Currently not, it's a little sticky because terminal width is not available in IPython frontends.
   maybe in the future.

[wesm]

Probably possible to add an option at some point for #2 (would be nice for interactive stuff)