Obtaining the computed task from a large h5 store can be problematic. It took either very long or caused a memory error. I suggest to fix this by managing a second dataframe of the parameters and samples of computed tasks in the h5 store file
Obtaining the computed task from a large h5 store can be problematic. It took either very long or caused a memory error.
I suggest to fix this by managing a second dataframe of the parameters and samples of computed tasks in the h5 store file