Love the concept, but every time I try it out, my brain fail to find good properties for testing my real life systems. It's easy to find them on toy examples from tutorials and slides, but how do you do that for thins like complex django endpoints or tensorflow models? Any tips?
-
-
-
This is a great question. Often times if you find yourself with big chunks of code whose properties aren't obvious, it is indicative that you should break that code into functions and write tests at a smaller scale. Or that you should have started writing tests earlier.
- Još 6 drugih odgovora
Novi razgovor -
-
-
I like this area of thought; always love more resources. Link to slides?
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
Can definitely confirm that! I'd recommend both property-based and "traditional" testing. See:https://datascience.blog.wzb.eu/2019/11/08/property-based-testing-for-scientific-code-in-python/ …
Hvala. Twitter će to iskoristiti za poboljšanje vaše vremenske crte. PoništiPoništi
-
-
-
Any property-based testing packages tou can recommend?
-
He mentioned the Hypothesis - https://hypothesis.readthedocs.io/en/latest/
Kraj razgovora
Novi razgovor -
Čini se da učitavanje traje već neko vrijeme.
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.