|
Apr
20 |
|
awarded | Teacher |
|
Apr
20 |
|
awarded | Supporter |
|
Apr
20 |
|
answered | Use regular expression to handle nested parenthesis in math equation? |
|
Apr
19 |
|
comment |
Efficiently sum a small numpy array, broadcast across a ginormous numpy array? steveha is right, they're single-byte (dtype=bool) values and it's feasible to keep them in ram. And with my performance requirements, I really can't afford to touch the disk at all, even to swap. But I agree that this is a useful addition for someone looking to do the same thing on a slower time scale, with less ram, thanks! |
|
Apr
19 |
|
awarded | Scholar |
|
Apr
19 |
|
comment |
Efficiently sum a small numpy array, broadcast across a ginormous numpy array? This is awesome, I didn't understand the dot function at all from my wandering through the docs. I did time it, and unfortunately it's not quite fast enough even on my high CPU ec2 instance, but this exactly what I asked for, and I'm glad to know about it, thanks! |
|
Apr
19 |
|
accepted | Efficiently sum a small numpy array, broadcast across a ginormous numpy array? |
|
Apr
19 |
|
comment |
Efficiently sum a small numpy array, broadcast across a ginormous numpy array? Yes, pure numpy is a win, but this is a nicely succinct solution as well, thanks! |
|
Apr
19 |
|
comment |
Efficiently sum a small numpy array, broadcast across a ginormous numpy array? stride_tricks is worth knowing to be sure! I wondered if something like that were possible, and tried to see if arrays could be constructed by reference, but gave up. I can imagine using this in the future, thanks! |
|
Apr
19 |
|
awarded | Student |
|
Apr
19 |
|
asked | Efficiently sum a small numpy array, broadcast across a ginormous numpy array? |
