3 ms·
Why don't you run on a very small subset of data to ensure that the code works e2e in the workflow pipeline?
by verdverm 4y ago
Why don't you run on a very small subset of data to ensure that the code works e2e in the workflow pipeline?
- killingtime74 4y agoYes that’s what I thought everyone did…
- davidatbu 4y agoYes, the example @beisner gave here (the val loop crashing after the training is done) is kind of a bad example, and some frameworks (like Pytorch lightning) do do a "full workflow check" before going into training, but his overall point that variadic generics could massively impact dev/researcher productivity stands, I think.
- beisner 4y agoSure, there are of course runtime checks you can do (and I now do after having been bitten by this before), but that’s not necessarily better than having a static analyzer which can guarantee correctness without executing a single line of code.