itamarst

31 ms·

301.
Reducing Pandas memory usage with lossy compression(pythonspeed.com)
1 points by itamarst 7y ago|0 comments
302.
by itamarst 7y ago
For health insurance claims in the US, apparently you want to say "I want to file a grievance", at least in Massachusetts (magically fixed $5000 bill where health insurance company was obviously lying). I'm told "I want
303.
Job negotiation for programmers: the basic principles(codewithoutrules.com)
4 points by itamarst 7y ago|0 comments
304.
by itamarst 7y ago
Eliot's sociological novel, an overarching view of how society works, is part of the inspiration for my logging library, named Eliot ( https://eliot.readthedocs.io ). The intro docs are inspired by Middlemarch: https:/&
305.
by itamarst 7y ago
You would think Uber Autonomous Vehicles group would do everything they can not to hit pedestrians. Yet according to the NTSB, previous to killing a pedestrian in Arizona Uber had a weak safety culture ( https://www.ntsb.gov/
306.
Reduce Pandas memory usage by loading less data(pythonspeed.com)
2 points by itamarst 7y ago|0 comments
307.
by itamarst 7y ago
See above about streaming (chunking is basically the same thing). mmaping is a form of uncontrolled chunking, where chunk size and location is determined by operating system's filesytem caching policy, readahead heuristics, and the lik
308.
by itamarst 7y ago
Chunking is the term I used because that's more relevant to the data science domain I'm focusing on here (e.g. Pandas has "chunksize", Zarr has "chunks"). Streaming has some implication of an ongoing stream of
309.
When your data doesn’t fit in memory: the basic techniques(pythonspeed.com)
638 points by itamarst 7y ago|259 comments
310.
When to Switch to Python 3.8(pythonspeed.com)
22 points by itamarst 7y ago|26 comments
311.
by itamarst 7y ago
And of course Route 53 still has a green checkmark.
312.
by itamarst 7y ago
And, as usual, the actual service has a green checkmark even though the top of the page acknowledges problems.
313.
AWS Route 53 experiencing occasional DNS resolution errors(status.aws.amazon.com)
13 points by itamarst 7y ago|5 comments
314.
Beyond cProfile: Sampling profilers and logging for performance optimization(pythonspeed.com)
1 points by itamarst 7y ago|0 comments
315.
by itamarst 7y ago
If you want to watch a video version: https://pyvideo.org/pycon-au-2017/what-nobody-tells-you-abou...
316.
Getting through the tedium and frustration of working as a programmer(codewithoutrules.com)
2 points by itamarst 7y ago|0 comments
317.
What’s running in production? Making your Docker images identifiable(pythonspeed.com)
4 points by itamarst 7y ago|0 comments
318.
Learning Negotiation from Jane Austen(codewithoutrules.com)
1 points by itamarst 7y ago|0 comments
319.
Faster or slower: the basics of Docker build caching(pythonspeed.com)
1 points by itamarst 7y ago|0 comments
320.
What can a software developer do about climate change?(codewithoutrules.com)
4 points by itamarst 7y ago|0 comments
321.
by itamarst 7y ago
End-to-end tests work, it's just that as I talk about it slows down the feedback loop a lot. Smoke test: "Your Docker image is broken", immediately after `docker build`. End-to-end tests: 30 minutes later "all your tests
322.
Docker builds need a smoke test(pythonspeed.com)
2 points by itamarst 7y ago|2 comments
323.
Not just CPU: writing custom profilers for Python(pythonspeed.com)
1 points by itamarst 7y ago|0 comments
324.
by itamarst 7y ago
If we all say "we're doomed, there's nothing to do but think of excuses" then we're _definitely_ doomed. (We might be doomed regardless, but who knows?). So instead of worrying about whether it's a lost cause,
325.
A Python prompt into a running process: debugging with Manhole(pythonspeed.com)
2 points by itamarst 7y ago|0 comments
326.
Where’s your bottleneck? CPU time vs. wallclock time(pythonspeed.com)
2 points by itamarst 7y ago|0 comments
327.
by itamarst 7y ago
You seem to be approaching this from the wrong direction. If you're really suffering from age discrimination, instead of thinking of women and minorities as the reason you're not being hired (do you really think that's the ca
328.
Decoupling database migrations from server startup: why and how(pythonspeed.com)
3 points by itamarst 7y ago|0 comments
329.
Speed is situational: two websites, two orders of magnitude(pythonspeed.com)
1 points by itamarst 7y ago|0 comments
330.
Choosing a Docker base image for your Python application in July 2019(pythonspeed.com)
2 points by itamarst 7y ago|0 comments

More ›