Ask HN: What do you use in research to reproduce your results?

Reproducibility of results is important as other researchers can compare their results and also make improvements. So, either if you are a researcher or not, how do you make your set up a good reproducibility enviorment? or how would you want others to do it? What tools do you use?

Now a days that Machine Learning is everywhere and with a very open research I am interested in how to do it in this topic. How do you organize the dataset so other can use it? how do you set up your experiments? do you use jupiter notebooks?