Hello everybody
I have code where I need to redo a calculation if a large dataset or model specification change. I have a large number of such datasets. In R I would just use the digest() function which provides a SHA key for any R object, in my case a list that contains the data and model, and if the digest() of the current data/model change, then re-fit, save and return, else load the solution from disk and return.
I have not found a way to do that in Julia. The hash() function is not consistent across Julia versions, and the SHA functions can not give a hash for a dataframe or a dictionary.
We can certainly cobble something together, but that is project specific and not nice. So I wanted to ask the Julia community if a solution for a persistent hash for dictionaries/dataframes existed that had escaped my limited google powers.
all the best, Jack