Python Meetup Explores Generative RNN Poetry
Last night at the Ottawa Python Authors Meetup, Emily Daniels of Halogen Software presented on artificial intelligence with Python, demonstrating recurrent neural networks that generate scripts and poetry from public-domain corpora such as H.G. Wells, Jane Austen, Walt Whitman and Emily Dickinson. She explained preprocessing, a week-long training process, and fielded audience questions about dataset bias and the need to archive models and data for stewardship.
Key Points
- 1Demonstrates generative RNNs producing scripts and poetry from public-domain corpora like Austen and Whitman
- 2Highlights training complexity: models required extensive preprocessing and about a week of training on collected texts
- 3Advises attention to dataset bias and archival backup to avoid propagating harmful outputs or losing data
Scoring Rationale
Local meetup demonstrates practical RNN demos and bias awareness, but narrow scope and limited technical depth lower its industry impact.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
