The latest articles on DEV Community by Rahul Raj (@rahulrajpl). https://dev.to/rahulrajpl 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%2F343607%2F1dbfb1ae-5fbb-4b65-bf7e-09dfccb90a08.png https://dev.to/rahulrajpl en Rahul Raj Fri, 05 Jun 2020 05:32:25 +0000 https://dev.to/rahulrajpl/wexda-web-based-data-exploration-tool-le0 https://dev.to/rahulrajpl/wexda-web-based-data-exploration-tool-le0 <p><a href="https://res.cloudinary.com/practicaldev/image/fetch/s--mlpghP9T--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_880/https://dev-to-uploads.s3.amazonaws.com/i/j64bi5tb6j4ieg49q3it.png" class="article-body-image-wrapper"><img src="https://res.cloudinary.com/practicaldev/image/fetch/s--mlpghP9T--/c_limit%2Cf_auto%2Cfl_progressive%2Cq_auto%2Cw_880/https://dev-to-uploads.s3.amazonaws.com/i/j64bi5tb6j4ieg49q3it.png" alt="Alt Text"></a><br> WExDA is a web based data exploration tool primarily useful for data preperation/ data analysis stage. This automates the EDA via web ui and is built using <a href="https://discuss.streamlit.io/">streamlit</a></p> <h2> How to use </h2> <p>Go to <a href="https://wexda.herokuapp.com/">https://wexda.herokuapp.com/</a> and start using ;)</p> <h2> Features </h2> <ul> <li>upload csv files upto 200MB</li> <li>describe data set</li> <li>find missing values</li> <li>find unique values and its count</li> <li>plot data via scatter, bar and pair plots</li> <li>plot correlation matrix in heatmap style</li> </ul> datascience machinelearning python Rahul Raj Mon, 23 Mar 2020 08:11:52 +0000 https://dev.to/rahulrajpl/diffusion-map-for-manifold-learning-theory-and-implementation-2b81 https://dev.to/rahulrajpl/diffusion-map-for-manifold-learning-theory-and-implementation-2b81 <p>‘Curse of dimensionality’ is a well-known problem in Data Science, which often causes poor performance, inaccurate results, and, most importantly, a similarity measure break-down. The primary cause of this is because high dimensional datasets are typically sparse, and often a lower-dimensional structure or ‘Manifold’ would embed this data. So there is a non-linear relationship among the variables (or features or dimensions), which we need to learn to compute better similarity.</p> <p>Read more at <a href="https://randomwalk.in/python/ml/2020/03/14/Diffusion-Map.html">https://randomwalk.in/python/ml/2020/03/14/Diffusion-Map.html</a></p> machinelearning datascience Rahul Raj Wed, 18 Mar 2020 11:16:54 +0000 https://dev.to/rahulrajpl/hans-rosling-style-analytics-on-covid19-3jd7 https://dev.to/rahulrajpl/hans-rosling-style-analytics-on-covid19-3jd7 <p>Hi all,</p> <p>I made a 'Hans Rosling' styled analytics on the Covid19 pandemic in India.<br> Check out the project at<br> <a href="http://randomwalk.in/covid19India/">http://randomwalk.in/covid19India/</a></p> <p>I welcome all comments/feedback at <a href="https://twitter.com/rahulrajpl">https://twitter.com/rahulrajpl</a></p> datascience