qd

quick-draw: a cli plotting tool

qd is a command line tool to quickly make plots from csv and json files or streams. It is built on top of the pandas and plotly libraries.

Installation and usage

qd can be installed via pip install qd-plot. It requires Python 3.7+.

Basic usage

Make a quick plot using the first columns available and display the output in a gui (default web browser).

cat data/trig.json | qd --gui

By default qd reads data from stdin and writes image bytes to stdout, however it also accepts input and output files as arguments, as well as the --gui option shown above.

qd -i data/trig.json -o trig.png

All the cli functionality available can be seen via the --help option.

qd --help

MacOS + iTerm2

Since qd writes to stdout by default, the images can be displayed right in the terminal window if using a compatible terminal, such as iTerm2 with imgcat:

cat data/trig.json | qd | imgcat

More examples

Mean in bins

Plot the mean values in some bins specifying the x and y columns.

cat data/trig.json | qd -x x -y sin,cos --mean --gui

Percentile in bins

Plot the 95th percentile values in 20 bins

cat data/trig.json | qd -x x -y sin --quant -q 95 --nbins 20 --gui

Histogram of values

Make a histogram from two sets of data using 20 bins.

cat data/dists.csv | qd -x gauss,expo --hist -n20 --gui

Local Development

Clone this repo from github and in a virtual environment do the following:

pip install .  # installs qd based on local code changes
pip install -r requirements-dev.txt  # installs extra packages for dev and testing

Tests can be run via:

pytest -n4