"Easter egg" at 11:06 :) ironically, the code under test will blow up because of a NameError (see spelling of max_bound -- fixed on my speakerdeck speakerdeck.com/marcobonzanini/building-data-pipelines-in-python-pydata-london-2016)
Great talk, +marcobonzanini. Thanks! I'm curious: if you were to start a new data pipeline today, would you still pick Luigi? It seems to me that Airflow is more advanced, although I like Luigi's simplicity.
Hi, I'm more or less in line with your observation: today Airflow offers a few nice features out of the box that Luigi doesn't (scheduling, history, nice dashboard, etc), but the way you write your pipeline code with Luigi feels more straightforward. I haven't tested Airflow extensively though
Very impressive talk. Gutted I missed it, but I'm incredibly appreciative for this upload. Thank you!
+SFSylvester Glad you liked it, feel free to get in touch if you want to discuss. Cheers, Marco
"Easter egg" at 11:06 :) ironically, the code under test will blow up because of a NameError (see spelling of max_bound -- fixed on my speakerdeck speakerdeck.com/marcobonzanini/building-data-pipelines-in-python-pydata-london-2016)
Great talk, +marcobonzanini. Thanks! I'm curious: if you were to start a new data pipeline today, would you still pick Luigi? It seems to me that Airflow is more advanced, although I like Luigi's simplicity.
Hi, I'm more or less in line with your observation: today Airflow offers a few nice features out of the box that Luigi doesn't (scheduling, history, nice dashboard, etc), but the way you write your pipeline code with Luigi feels more straightforward. I haven't tested Airflow extensively though