define mini-batches for yourself and then push to the specified number of cores. We should just make a pool and delegate to that, or something.
This could be super helpful with some of the larger kde computations we'd like to make. Hitting it with a full pcdev2 may reduce the runtime by a factor ~60
define mini-batches for yourself and then push to the specified number of cores. We should just make a pool and delegate to that, or something.
This could be super helpful with some of the larger kde computations we'd like to make. Hitting it with a full pcdev2 may reduce the runtime by a factor ~60