3 ms·
A vectorized Search Visualizer written with NumPy
Hey Y Combinator community,
Last semester, I delved into a Data Structures and Algorithms course with a hunger for visualizing complex algorithms and data structures.
While I found some success using [grid-based search visualizers](https://i.imgur.com/4jQwRHu.png), my desire for a versatile graph search visualizer persisted.
So, I decided to create one. Now, it not only supports generic graph search visualization but also tackles challenges like the traveling salesman problem and various other graph-based algorithms.
The API is straightforward. The `graphViz` module provides granular access at the node and edge levels within a given `Graph` object, allowing you to dynamically alter colors while running your graph-based algorithm.
For a more in-depth understanding, I've put together:
- [A detailed walkthrough of recent updates](https://youtu.be/BKF-PEgd1PA)
- [A concise overview of the initial version](https://youtu.be/kNGsOoWh9fM)
I'd love for you to check out the [GitHub Repo](https://github.com/gyanantaran/searchViz) and share your thoughts with me. Your feedback is highly valued!
- anigbrowl 3y agoFYI, it's best to submit a URL (probably your github) and then add further information as a comment. As for the tool itself, it looks pretty nice! The default layout is good although it will be worthwhile to experiment with others. Consider also using colormaps for metrics like path length, degreeness, or network centrality. Gephi is a highly developed open source graph visualization project with a lot of plugins, which will give you a lot of ideas and allow you to experiment freely.
- gyanantaran 3y agoThank you for replying. I was confused about the URL as it said it made the additional info optional. I now understand. I am not able to edit the post now though. Thanks for the feedback about the tool. Yes, for the layout(if you mean the distribution), I've written a module to write or supply any distribution/layout to the Graph. I've implemented some already: gaussian, uniform, and exponential distributions for the node-locations. There is also a module for the edge-connections models. Oh, yes, path-length, degree-ness and network centrality have not been implemented but are controlled indirectly by the edge connections model. I was unaware of Gephi, seems a full-fledged application. Thanks for sharing!