Our team is excited to share a new feature that you will see in DrivenData competitions: Community Code.
The Community Code section is a place where participants can share helpful code related to the competition. This could be anything from short snippets demonstrating some data processing to longer tutorials or analyses.
Community Code will be enabled on a competition-by-competition basis. The first competition that has Community Code enabled is our new Unsupervised Wisdom competition to extract insights from data about older adult fall-related injuries. You can find the Community Code section here. To encourage contributions, this competition includes a $2,500 “Most helpful shared code” bonus prize to a participant selected by judges.
Our team wants to share more than 5MB of code/ tutorial-like notebook. Is there any possibility to increase the limit. That would allow more visualizations and code snippets. Our team had to spend extra few hours to trim out data as a result of the 5MB limit.
Thank for your feedback. Are you able to speak a little more about what you are including in your notebook that you think is causing it to exceed the 5 MB size limit?
One of the purposes of the size limit is to encourage contributors to be more concise and deliberate about what is included in the notebook. For example, you should not print out a significant amount of raw data.
Our notebook was close to 4.8MB because we wanted to show the visualization outputs so that viewers could see the effects without downloading and running on their own computer.
To overcome the memory limit, these visualizations have to be trimmed out (left 2-3 example Seaborn figures in the current version).
We noticed that the interactive figures from Plotly are consuming lots of memory and so we left those commands not executed in the notebook.
We plan to update the notebook in near future and thus why wondering about the memory limit.
The current version of your notebook shows up as 242 KB for me, and the two Seaborn plots I see are 19 KB and 61 KB respectively. It seems like you have plenty of margin left to work with.
I think where you may be running into size issues is with interactive components like Plotly, which tend to embed the raw data in order to make them interactive. You may find that such components won’t work with our current implementation of Community Code anyways—we currently restrict some interactive Javascript functionality for security purposes.