Timeline for Finding permutations efficiently
Current License: CC BY-SA 4.0
15 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Apr 8, 2020 at 7:22 | vote | accept | Xbel | ||
| Apr 7, 2020 at 16:04 | answer | added | Maarten Fabré | timeline score: 1 | |
| Apr 7, 2020 at 15:00 | history | tweeted | twitter.com/StackCodeReview/status/1247539744054996996 | ||
| Apr 7, 2020 at 14:44 | comment | added | Xbel |
Added type definitions, and imports. I hope I didn't miss anything. The objective of the function is to find among P times series those that are more correlated with K. That is way only K are returned. In the ideal case, would be at least one 1 for each row, but can happen that two series in permuted are more similar to only one in true, that's prevented by tracking them using used_comps.
|
|
| Apr 7, 2020 at 14:37 | history | edited | Xbel | CC BY-SA 4.0 |
Added type definitions
|
| Apr 7, 2020 at 13:14 | comment | added | Maarten Fabré |
In your docstring you mention P >K, but in your example P == K and is every time series represented at least once? or better, is there a 1 in each row of corr_matrix
|
|
| Apr 7, 2020 at 13:08 | comment | added | Maarten Fabré | The cython code as is will not work for me. There are some imports and typedefs missing. Can you include thos? | |
| Apr 7, 2020 at 12:17 | comment | added | Xbel | Added in the python code part :) | |
| Apr 7, 2020 at 12:17 | history | edited | Xbel | CC BY-SA 4.0 |
change python code to add minimal example.
|
| Apr 7, 2020 at 10:04 | comment | added | Maarten Fabré | do you have some (dummy) sample data with which you call this? | |
| Apr 7, 2020 at 8:59 | history | edited | Xbel | CC BY-SA 4.0 |
Typo in docstrings
|
| Apr 7, 2020 at 7:34 | history | edited | Xbel | CC BY-SA 4.0 |
Bug in the Cython code
|
| Apr 7, 2020 at 7:28 | history | edited | Xbel | CC BY-SA 4.0 |
Added a Cython version of the code
|
| Apr 7, 2020 at 7:07 | history | edited | Xbel | CC BY-SA 4.0 |
edited body: typos
|
| Apr 6, 2020 at 14:37 | history | asked | Xbel | CC BY-SA 4.0 |