А все не так сложно: ты хочешь реплицируемое или воспроизводимое? What’s

Question

А все не так сложно: ты хочешь реплицируемое или воспроизводимое? What’s

the difference between replication and reproduction?
There is no consensus yet on what exactly these two terms mean, so here is how we understand and use them.

Reproduction of a computational study means running the same computation on the same input data, and then checking if the results are the same, or at least “close enough” when it comes to numerical approximations. Reproduction can be considered as software testing at the level of a complete study.

Replication of a scientific study (computational or other) means repeating a published protocol, respecting its spirit and intentions but varying the technical details. For computational work, this would mean using different software, running a simulation from different initial conditions, etc. The idea is to change something that everyone believes shouldn’t matter, and see if the scientific conclusions are affected or not.

Reproduction verifies that a computation was recorded with enough detail that it can be analyzed later or by someone else. Replication explores which details matter for reaching a specific scientific conclusion. A replication attempt is most useful if reproducibility has already been verified. Otherwise, if replication fails, leads to different conclusions, you cannot trace back the differences in the results to the underlying code and data.

In the ideal world, the reproducibility of computational studies would be checked automatically before publication. In the real world, most published studies do not even claim to be reproducible, because the code or the input data are not made public. Given this situation, a ReScience C publication provides two important services: by replicating the original work, it provides an independent implementation of the original computational protocol, and by making this new implementation public, it is reproducible and thus a safer basis for future research to build on.

#programming #r #russian

0

03.08.2020

1 ответов

18 просмотров

Philipp Upravitelev · Accepted Answer

Philipp Upravitelev

именно воспроизводимое то есть, где пайплайн и код анализа один и тот же, только данные разные

0

03.08.2020

169 похожих чатов

А все не так сложно: ты хочешь реплицируемое или воспроизводимое? What’s

1 ответов

Похожие вопросы