Cplot

A very simple matplotlib wrapper to plot columns from a data file.

Getting started

First up, clone this repo:

$ git clone https://github.com/samharrison7/cplot
$ cd cplot

Cplot is a Python script which relies on NumPy, Pandas and Matplotlib. Chances are, if you're a Python developer, you'll already have these packages installed. If not, or if you'd like to keep things clean, then you can use the provided Conda environment.yaml file to create an environment to run cplot from:

$ conda env create -f environment.yaml
$ conda activate cplot

You can then run the cplot script from this directory, or copy it to somewhere on your $PATH to make it globally available:

(cplot) $ cp ./cplot ~/bin            # For example, if ~/bin is in your $PATH

Usage

usage: cplot [-h] [-s SEP] [-x X] [-y [Y [Y ...]]] [-he] [-f [FMT [FMT ...]]]
             [-gb GROUPBY] [-a {mean,median,sum,std}] [-c COMMENT]
             file

Plot some columns.

positional arguments:
  file                  path to the data file

optional arguments:
  -h, --help            show this help message and exit
  -s SEP, --sep SEP     the separator between columns in the data, default to
                        comma ","
  -x X, --x X           column representing x, defaults to None
  -y [Y [Y ...]], --y [Y [Y ...]]
                        column(s) representing y
  -he, --header         is the first row a header?
  -f [FMT [FMT ...]], --fmt [FMT [FMT ...]]
                        plot format, same convention as pyplot.plot() fmt
                        argument. Can be list where each item is format for
                        different y column
  -gb GROUPBY, --groupby GROUPBY
                        column to group data by. Pandas groupby() is applied
                        internally
  -a {mean,median,sum,std}, --agg {mean,median,sum,std}
                        numpy function to aggregate grouped table by mean,
                        median, sum, std or cumsum. Defaults to mean
  -c COMMENT, --comment COMMENT
                        ignore lines denoted by this comment character.

The simplest example

For example, for the one column data structure (example/sin.txt):

0.8414709848078965
0.9092974268256817
...
-0.9992068341863537

To plot a line graph:

(cplot) $ cplot example/sin.txt

Multiple columns, seperators and formats

For example, for the data structure (example/multicol.txt):

x;y1;y2
1;10;30
2;20;20
3;30;10

To plot x vs y1:

(cplot) $ cplot example/multicol.txt -s ";" -he -x 0 -y 1
# OR
(cplot) $ cplot example/multicol.txt -s ";" -he -x x -y y1

To plot x vs y1 and y2:

(cplot) $ cplot example/multicol.txt -s ";" -he -x x -y y1 y2
# OR
(cplot) $ cplot example/multicol.txt -s ";" -he -x 0 -y 1 2

Changing the plot formats:

(cplot) $ cplot example/multicol.txt -s ";" -he -x x -y y1 y2 -f bo r-

Groupby

Cplot deals with the simplest of "groupby" operations; namely, grouping by one column using one aggregation function. The column to group by is provided by the -gb or --groupby flag, and the aggregation function by the -a or --agg flag (which defaults to "mean"). Internally, pandas.DataFrame.groupby() is used (i.e. DataFrame.groupby([-gb])).agg(np.-a)).

For example, if you have the data structure example/groupby.txt:

t,x,y,z
1,1,1,10
1,1,2,10
2,1,1,20
2,1,2,20
3,1,1,30
3,1,2,30
4,1,1,40
4,1,2,40

To plot t against the sum of z when grouped by x and y:

(cplot) $ cplot example/groupby.txt -he -gb t -a sum -x t -y z -f gs

Permitted aggregation functions (which are NumPy functions) are mean, median, sum and std.

Comments in files

You can ignore lines which are commented out by providing the comment character as an argument. For example, if you have the data structure:

# This is a comment
x,y
1,2
2,3
3,4

To plot x against y:

(cplot) $ cplot example/comment.txt -he -c# -x x -y y