# Derek Jones

shape-of-code.coding-guidelines.com

 1d awarded Popular Question Aug 11 comment coxph ran out of iterations and did not converge"Ran out of iterations and did not converge" does not sound like convergence to me. No idea why this is a warning and not an error. Aug 10 comment coxph ran out of iterations and did not convergeThe data is certainly not fake. So the question of how to get get `coxph` to converge still stands. Aug 10 comment coxph ran out of iterations and did not convergeI think you have missed the point @DWin, `total_usage` is time dependent and slicing up the total time into intervals, all censored until the last one, is the solution. Of course `fault_id` is an ascending integer sequence, it was created to identify the clusters created by the slicing process. Aug 7 comment coxph ran out of iterations and did not convergeThanks for the suggestions @gung. Randomising the fault report times over a time span (instead of reguar intervals) does not make the problem go away. A histogram of the censored fault `total_usage` looks roughly Poisson with a peak around 3,000, while for actual faults the Poisson look is a bit rougher and peaks around 4,000. Aug 5 asked coxph ran out of iterations and did not converge Mar 28 awarded Tumbleweed Mar 17 revised Distribution of Levenshtein distances for partially sorted listsadded 565 characters in body Mar 17 comment Distribution of Levenshtein distances for partially sorted listsUnless there is an analytic solution the problem is computationally intensive. Limiting the search space to the set of lists that are close to \$L_1\$, in the sense of being a partially sorted version of it significantly reduces the computational effort. You are right that the original question is not clear, I will update t. Mar 17 comment Distribution of Levenshtein distances for partially sorted listsTwo lists, \$L_1\$ and \$L_2\$ where the elements in \$L_1\$ are sorted and \$L_2\$ contains the same elements as \$L_1\$ in some order. Compute the Levenshtein distance between \$L_1\$ and \$L_2\$ where \$L_2\$ takes on all possible permutations of the elements in \$L_1\$. Given this set of distances I can plot Levenshtein distance against the number of permutations having a given distance, this is the distribution I am after. Mar 17 comment Distribution of Levenshtein distances for partially sorted listsI expect the lists I am measuring to have many elements sorted, say 90% in sort order. I'm happy to ignore any lists that don't have a reasonable degree of sortedness (sorry for delay in replying). Mar 12 asked Distribution of Levenshtein distances for partially sorted lists Feb 19 awarded Supporter Feb 19 comment Combination of three items with no adjacent items the sameYes! Exactly what I was after. Thanks for this @Jair Taylor Feb 18 revised Combination of three items with no adjacent items the sameadded 56 characters in body Feb 18 revised Combination of three items with no adjacent items the sameadded 56 characters in body Feb 18 revised Combination of three items with no adjacent items the sameadded 56 characters in body Feb 18 revised Combination of three items with no adjacent items the sameadded 126 characters in body Feb 18 comment Combination of three items with no adjacent items the same@Byron Schmuland, you're right, I should have read more carefully. Thanks gnometorule (no @ because you are probably swamped). Feb 18 comment Combination of three items with no adjacent items the sameThe parameter to Taylor's equation is \$k\$, there is no \$k_1\$, \$k_2\$ and \$k_3\$ to support different numbers of each item.