3 ms·
For exploratory work, perhaps you should random sample some of the dataset (say 1k) and see the effect on them? After getting good result, you then switch to de
by pca006132 4y ago
For exploratory work, perhaps you should random sample some of the dataset (say 1k) and see the effect on them? After getting good result, you then switch to dealing with the whole dataset.
- chaps 4y agoNo need in my workflows since unix tools solve that exploratory starting point. Just about everything else is SQL.