20210423, 06:25  #45  
"Alexander"
Nov 2008
The Alamo City
17·41 Posts 
Quote:
Quote:
Quote:
Quote:


20210423, 08:14  #46  
"Garambois JeanLuc"
Oct 2011
France
1010000000_{2} Posts 
Quote:
I don't think the variable names matter, it won't change anything. This name will not appear in the database, just where I will explain the meaning of each number in the row. Is not it ? But if that suits you, we can call it p. Quote:
And you are talking about the averages of the terms of the maxima and minima or the averages of the number of digits of these terms ? I haven't started running the program yet, because I still changed a few lines of code and ran speed tests with different settings. But I will wait for your response to this message, to find out which averages of the maxima and minima would be useful to you. Because this will add two variables and I will still have to modify the program. If I add these two variables after the other 15, the order of the variables will also be quite weird : maybe I should change the order of the variables so that all the means are "close". Do you have an opinion on the order of the variables ? 

20210423, 23:14  #47  
"Alexander"
Nov 2008
The Alamo City
17×41 Posts 
Quote:
Quote:
Quote:
Last fiddled with by Happy5214 on 20210423 at 23:18 

20210424, 07:51  #48  
"Garambois JeanLuc"
Oct 2011
France
2^{7}·5 Posts 
First of all, thank you very much Happy for your interest in this work.
I think the database will be even richer now ! I will take your advice. 1) I will add the two arithmetic means you ask for and I will change the order of the variables, it will be much clearer this way. 2) I will add another variable : the number of minimums encountered in a sequence. This is not necessarily the same as the number of maximums, for example in the case of a sequence with a Vshaped graph. Quote:
I still do not understand what you are suggesting. Let me make up a digital example below with a fictitious sequence (deficient terms in red, abundant terms in black, c is a cofactor without the 3) : index. term 0. 2*c 1. 2^2*c 2. 2^2*c 3. 2^3*c 4. 2^3*c 5. 2^3*c 6. 2*3*c 7. 2*3*c 8. 2*c 9. 2*c 10. 2*c 11. 2*c 12. 2^2*c 13. 2^2*c 14. 2^2*c 15. 2^2*c 16. 2^2*c 17. 2^3*c 18. 2^3*c 19. 2^3*c 20. 2^3*c 21. 2^3*c 22. 2^3*c Please, what is the number q in this example, and why ? 

20210425, 06:36  #49  
"Alexander"
Nov 2008
The Alamo City
17×41 Posts 
Quote:


20210425, 17:41  #50 
"Garambois JeanLuc"
Oct 2011
France
2^{7}·5 Posts 
Thank you all very much for your advice and ideas.
The new "regina" program is finally running ... and it calculates 19 variables for each sequence. I will make final checks during this week and I will keep you posted ... 
20210502, 22:02  #51 
"Ed Hall"
Dec 2009
Adirondack Mtns
7372_{8} Posts 
JeanLuc, Could you post a sample of your new file, with the new key for the changed elements? I'm hoping to work some more on getting the latest version to work with both formats. I am also trying to add some "features" but some of them are still giving me a bit of a challenge. Ed

20210504, 16:59  #52  
"Garambois JeanLuc"
Oct 2011
France
2^{7}·5 Posts 
Quote:
To explain the meaning of all the variables in this file, I think it is much easier to take the letters of the alphabet in order and explain the meaning of each variable. Each line of the file represents a sequence and is written like this : [A,B,C,D,E,F,G,H,I,J,K,L,M,N,O,P,Q,R,S,T] A : starting integer of the aliquot sequence. B : relative integer which is 1 if the aliquot sequence from start A ends up at 1, which is 0 if the sequence is openend, which is negative if the sequence falls on a cycle of length B, with B being the absolute value of B. C : number of iterations at the end of the computation of the sequence. We stop the calculation if we end up with 1, if we find a cycle or if the terms are greater than 10^50 and we consider it as an openend sequence (or if the sequence meets a term of 25 digits of an openend sequence already calculated previously, to decrease the calculation time). D : if B is 1, D is the prime number on which the starting sequence A fell. If B is 0, D is the smallest starting integer of the openend sequence merged by the starting sequence A. If B is negative, D is the integer belonging to the cycle by which the starting sequence A entered this cycle. E : if B is 1, E tells for how many other sequences one has already fallen on the prime number D, including the starting sequence A. If B is 0, E tells for how many other sequences one has already fallen on the openend beginning with D, including the starting sequence A. If B is negative, E tells for how many other sequences one has already fallen on the same cycle, including the starting sequence A. F : number of digits of the largest term of the starting sequence A. G : number of relative minimums of the starting sequence A. H : number of relative maximums encountered in the starting sequence A. I : number of parity changes encountered in the starting sequence A. J : record number of consecutive even abundant terms in the starting sequence A. K : record number of consecutive even deficient terms in the starting sequence A. L : record number of consecutive odd deficient terms in the starting sequence A. M : record number of consecutive odd abundant terms in the starting sequence A. N : number of downdriver extracts in the sequence : terms of the form 2^1 * cofactor with 3 which does not divide cofactor. O : smallest quotient of two consecutive terms s_{u+1}(A)/s_{u}(A) found in the whole starting sequence A, except the last quotient if B=1 and A not prime (if A is prime, the only quotient is the last quotient and it has no sense, because the last but one term of the aliquot sequence is prime and so the last quotient can be very small if this prime number is large). P : largest quotient of two consecutive terms s_{u+1}(A)/s_{u}(A) found in the whole starting sequence A. Q : arithmetic mean of s_{u+1}(A)/s_{u}(A), that is to say arithmetic mean of all the quotients of two consecutive terms of the starting sequence A, except for the last quotient if B=1 (this last quotient is the unique one if A is prime and in this last case, we count it anyway). We make the sum of all s_{u+1}(A)/s_{u}(A), 0<=u<=C, and we divide it by the number C; or by C1 if B=1 and if A is not prime. R : geometric mean of s_{u+1}(A)/s_{u}(A), i.e. geometric mean of all the quotients of two consecutive terms of the starting sequence A, except for the last quotient if B=1 (this last quotient is the only one if A is prime and in this last case, it is counted anyway). We make the product of all s_{u+1}(A)/s_{u}(A) with 0<=u<=C, that comes back to consider finally only the quotient of the last term of the sequence by the first term, and we raise this exponent to the power 1/C ; or by C1 if B=1 and if A is not prime. Be careful with the entry in the cycles : here, we only count all the members of the cycle once. But if we want another geometric mean, we can find it with the other parameters of the line. For example, we can want the geometric mean of the sequence without the cycle, or by counting the number of entries in the cycle twice... S : arithmetic mean of the number of digits of all the minimums of the sequence. T : arithmetic mean of the number of digits of all the maximums of the sequence. Here are some lines from the regina file as an example : 2,1,1,2,1,1,0,0,1,0,0,0,0,1,0.5000000000,0.5000000000,0.5000000000,0.5000000000,0,0 3,1,1,3,1,1,0,0,0,0,0,1,0,0,0.3333333333,0.3333333333,0.3333333333,0.3333333333,0,0 4,1,2,3,2,1,0,0,1,0,0,1,0,0,0.7500000000,0.7500000000,0.7500000000,0.7500000000,0,0 5,1,1,5,1,1,0,0,0,0,0,1,0,0,0.2000000000,0.2000000000,0.2000000000,0.2000000000,0,0 6,1,1,6,1,1,0,0,0,0,0,0,0,0,1.000000000,1.000000000,1.000000000,1.000000000,0,0 7,1,1,7,1,1,0,0,0,0,0,1,0,0,0.1428571429,0.1428571429,0.1428571429,0.1428571429,0,0 8,1,2,7,2,1,0,0,1,0,0,1,0,0,0.8750000000,0.8750000000,0.8750000000,0.8750000000,0,0 9,1,3,3,3,1,0,0,2,0,0,1,0,0,0.4444444444,0.7500000000,0.5972222222,0.5773502692,0,0 10,1,3,7,3,2,0,0,1,0,1,1,0,1,0.8000000000,0.8750000000,0.8375000000,0.8366600265,0,0 11,1,1,11,1,2,0,0,0,0,0,1,0,0,0.09090909091,0.09090909091,0.09090909091,0.09090909091,0,0 12,1,6,3,4,2,0,1,3,1,0,1,0,0,0.4444444444,1.333333333,0.8130555555,0.7578582833,0,2.000000000 13,1,1,13,1,2,0,0,0,0,0,1,0,0,0.07692307692,0.07692307692,0.07692307692,0.07692307692,0,0 14,1,4,7,4,2,0,0,1,0,2,1,0,1,0.7142857143,0.8750000000,0.7964285714,0.7937005260,0,0 ... ... 138,1,177,59,2,12,14,15,1,31,37,2,0,9,0.2226415094,2.472557387,1.097090035,0.9951837026,5.714285714,6.533333333 ... ... 276,0,491,276,1,51,30,30,0,164,55,0,0,2,0.5000000000,2.918415621,1.365056702,1.250148383,26.43333333,27.03333333 277,1,1,277,1,3,0,0,0,0,0,1,0,0,0.003610108303,0.003610108303,0.003610108303,0.003610108303,0,0 278,1,6,43,42,3,1,1,1,1,3,1,0,2,0.5107913669,1.250000000,0.7364917336,0.6884695423,2.000000000,2.000000000 279,1,2,137,2,3,0,0,0,0,0,2,0,0,0.4910394265,0.4910394265,0.4910394265,0.4910394265,0,0 280,1,15,41,31,3,2,3,1,3,6,2,0,2,0.5066079295,1.571428571,0.9314937741,0.8717698494,2.500000000,2.666666667 281,1,1,281,1,3,0,0,0,0,0,1,0,0,0.003558718861,0.003558718861,0.003558718861,0.003558718861,0,0 282,1,16,163,2,5,0,1,1,11,0,4,0,0,0.1080225678,1.971428571,1.188064593,0.9641158730,0,5.000000000 283,1,1,283,1,3,0,0,0,0,0,1,0,0,0.003533568905,0.003533568905,0.003533568905,0.003533568905,0,0 284,2,2,284,2,3,1,0,0,1,1,0,0,0,0.7746478873,1.290909091,1.032778489,1.000000000,3.000000000,0 ... ... 306,0,144,276,2,26,1,1,0,136,2,0,0,1,0.8047493404,2.872578052,1.512782551,1.437485463,4.000000000,4.000000000 For information, after almost 10 days of calculation, I am at 2.5 M. Don't hesitate to let me know if something is not clear to you, I had a hard time explaining all this in English ! Don't hesitate either if you find an anomaly in the file ! 

20210504, 23:59  #53 
"Ed Hall"
Dec 2009
Adirondack Mtns
2·3^{3}·71 Posts 
Thanks! I'll let you know how it turns out. It may be a while, but let's see where it goes. . .

20210511, 15:15  #54 
"Ed Hall"
Dec 2009
Adirondack Mtns
2·3^{3}·71 Posts 
Version 2 of seqinfo
Here is the new version of the tool, which I'm calling "seqinfo2." Of course, it can be named anything one wishes when it is compiled, but I will refer to it as seqinfo2.
This file was written to read the entire NEW version of regina_file (up to 20M). It will not work with the older version, which has a different order and number of elements per line. The default filename is coded as regina_file, but an alternate name can be provided on the command line, e.g. "regina_file_1e5" for the 2100000 set. The biggest addition to this version is the ability to do "Advanced Searches" of the data. All of the reginafile is read into memory, so be sure your machine has a fair amount of RAM. Since many of the new fields contain precision values, (almost) all entries for the Advanced Search function are set up to accept up to two values separated by a space. In the case of the first entry, which filters the sequence values, the entry is made in the form of a and b, where a (mod b) will be evaluated. To choose 3 (mod 7), enter "3 7" (without quotes) and the values of 3, 10, 17, ... will be evaluated. A single value equals 0 (mod <value>) and leaving the entry blank equals 0 (mod 1), evaluating all sequences in the full list. The next entry is for the range of sequences and is inclusive. If a single value is entered, it is the range. If this entry is left blank, the entire set is evaluated. The third entry is the only one that expects a single letter to choose whether to limit the search to primes, openendeds or cycles. A blank entry is treated as "all." All the rest of the entries are of the "min max" format and a single entry is treated as both values. A blank entry is treated as the range being 0 to 10000, which should mean all values.* * Note: If any values other than element B can be <0, let me know, so I can change the program to include them. The source should compile under linux with "g++ seqinfo2.cpp o seqinfo2" on the command line and should run with "./seqinfo2" (or "./seqinfo2 <alternate filename>") in a terminal. If any messages are returned when compiling, let me know of them. 
20210513, 09:08  #55 
"Garambois JeanLuc"
Oct 2011
France
2^{7}×5 Posts 
Thanks Edwin for all your hard work !
I have some questions, remarks, answers, information... 1) Would it be possible to show us an example of an entry to achieve an "Advanced Search" ? 2) After less than 3 weeks of computation, the new program came to more than 4,000,000. So I think it should reach 20 M sometime this summer. But then, the RAM needed to run the program should become too large for almost all computers, unless I am wrong ! For sequences up to 20 M, the file will be over 2 GB. And in python, in my experience, it is impossible to process such a file by storing it in RAM : it takes much more memory than the file size. But maybe it's different in C, I hope so ! 3) Just for information, I am thinking of letting the program run for several months and this new program could process all the sequences up to 50 M, or even more... if my old computer does not break down. 4) There can be no other negative value than B in the table, that is for sure. 
Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Aliquot sequence reservations  schickel  Aliquot Sequences  3529  20210729 19:46 
A new tool to identify aliquot sequence margins and acquisitions  garambois  Aliquot Sequences  24  20210225 23:31 
Extending an aliquot sequence backwards  arbooker  Aliquot Sequences  5  20201107 15:58 
Another Aliquot Sequence site  schickel  Aliquot Sequences  67  20120120 17:53 
Useful aliquotsequence links  10metreh  Aliquot Sequences  2  20090731 17:43 