The latest articles on DEV Community by João Santiago (@jcpsantiago). https://dev.to/jcpsantiago https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.amazonaws.com%2Fuploads%2Fuser%2Fprofile_image%2F359670%2Fd6f8a5cb-6afb-45ee-bbf3-3d1a9da53356.jpeg https://dev.to/jcpsantiago en João Santiago Mon, 07 Sep 2020 11:05:56 +0000 https://dev.to/jcpsantiago/print-dvc-plots-and-metrics-directly-in-the-terminal-1c6g https://dev.to/jcpsantiago/print-dvc-plots-and-metrics-directly-in-the-terminal-1c6g <p><a href="https://dvc.org/">DVC</a> has changed the way I handle Machine Learning projects for the better. The <code>dvc plots diff</code> command is a great way to see the effect of rounds of experimentation. </p> <p>If you use the <a href="https://sw.kovidgoyal.net/kitty/index.html">kitty</a> terminal, it's possible to print DVC plots directly in the terminal without opening an external program/browser. I wrote a function for the <a href="https://fishshell.com/">fish shell</a> to simplify this process.</p> <p><a href="https://res.cloudinary.com/practicaldev/image/fetch/s--JD35hrkh--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_880/https://dev-to-uploads.s3.amazonaws.com/i/1olzyqp8nowcjmt5vdic.png" class="article-body-image-wrapper"><img src="https://res.cloudinary.com/practicaldev/image/fetch/s--JD35hrkh--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_880/https://dev-to-uploads.s3.amazonaws.com/i/1olzyqp8nowcjmt5vdic.png" alt="Alt Text"></a></p> <p>Assuming you are using <code>fish</code> as your shell, save this code as ~/.config/fish/functions/plotit.fish</p> <div class="ltag_gist-liquid-tag"> </div> <p>Then refresh your shell with <code>source ~/.config/fish/functions/plotit.fish</code>. Finally, install <code>jq</code> and <code>vega-cli</code>, if you don't have them in your system yet.</p> <p>Example usage when inside a DVC-enabled project:</p> <ul> <li> <code>plotit pr_curve.csv</code> -&gt; prints diff plot with width x height of 500px x 500px and x set to row-index and y to the second column of the csv</li> <li> <code>plotit pr_curve.csv recall precision</code> -&gt; same as above but selects the columns for x and y</li> <li> <code>plotit pr_curve.csv recall precision 800 800</code> -&gt; with custom width and height in pixels</li> </ul> <p>Enjoy!</p> datascience dvc machinelearning commandline