mersenneforum.org April 2020
 Register FAQ Search Today's Posts Mark Forums Read

2020-04-12, 22:10   #34
0scar

Jan 2020

2710 Posts

Quote:
 Originally Posted by Kebbaj I get with the example of the Excel file attached, that after 10 days all cities are 100% contaminated. Where does the 29% come from? Something that I am not well understood?
I didn't look at your Excel file before.
According to your computations, after ten days the cities are contaminated ABOVE 100%.
You noticed it, and you hid it with well placed if statements.
But the main problem is that you are counting many infection events more than once.

As an example, on day 2 the correct values should be (using your format):
A = 10000,
B = 1900,
C = 1981,
D = 1981,
E = 199.

Last fiddled with by 0scar on 2020-04-12 at 22:19 Reason: Adding correct values on day 2

2020-04-13, 08:46   #35
Kebbaj

"Kebbaj Reda"
May 2018
Casablanca, Morocco

728 Posts

Quote:
 Originally Posted by 0scar I As an example, on day 2 the correct values should be (using your format): A = 10000, B = 1900, C = 1981, D = 1981, E = 199.
City B for example
Day 0 infection B: 0
Day 1 infection B: 1000
(10000 * 0.1)
Day 2 infection B: 2000
(1000(day 1) + 10000 * 0.1 = 2000)

Where did your 1900 come from for city B.

Last fiddled with by Kebbaj on 2020-04-13 at 08:49

2020-04-13, 09:07   #36
0scar

Jan 2020

33 Posts

Quote:
 Originally Posted by Kebbaj City B for example Day 0 infection B: 0 Day 1 infection B: 1000 (10000 * 0.1) Day 2 infection B: 2000 (1000(day 1) + 10000 * 0.1 = 2000) Where did your 1900 come from for city B. Please answer with calculations.
On day 2, city B, only the 9000 people still healthy can be infected (from A only):
1000+9000*0.1= 1900.
Otherwise, some people infected on previous days are counted over and over again, quickly exceeding 100%.

2020-04-13, 10:19   #37
Kebbaj

"Kebbaj Reda"
May 2018
Casablanca, Morocco

2×29 Posts

Quote:
 Originally Posted by 0scar On day 2, city B, only the 9000 people still healthy can be infected (from A only): 1000+9000*0.1= 1900. Otherwise, some people infected on previous days are counted over and over again, quickly exceeding 100%.
Yes your reason is better than mine. I had seen that the source city A with a high viral load of 10000 infected cases can count 1000 healthy cases in city B.

2020-04-14, 16:40   #38
uau

Jan 2017

79 Posts

Anyone have an idea what the newly added bonus thing means? It says:
Quote:
 Bonus '*' for getting the closest to 70% in 19 days; and ’**' for getting even closer.
"even closer"? What does that mean? Only one person has '**'; with the limited space for potential solutions, it doesn't seem plausible that everyone else would have failed to find the best possible solution...

2020-04-14, 17:30   #39
0scar

Jan 2020

110112 Posts

Quote:
 Originally Posted by uau Anyone have an idea what the newly added bonus thing means? It says: "even closer"? What does that mean? Only one person has '**'; with the limited space for potential solutions, it doesn't seem plausible that everyone else would have failed to find the best possible solution...
I think that someone didn't stop on day 10 or on day 19, but he always checked every t until p >=.7 was reached. And he eventually found some p very, very close.
My guess: somewhere before day 81.

 2020-04-27, 07:44 #40 Dieter   Oct 2017 2·72 Posts Are there many equivalent adjacency matrices with 70,0074873 % after 10 days or is my code wrong?
2020-04-27, 08:21   #41
SmartMersenne

Sep 2017

10101102 Posts

Quote:
 Originally Posted by Dieter Are there many equivalent adjacency matrices with 70,0074873 % after 10 days or is my code wrong?
Not sure, but there must be many equivalent matrices, at least the permutations of all the rest of the points except the originating city, minus the degenerate cases due to isomorphic connections. If the graph is dense, there will be many degenerate cases, but if the it is sparse, you will have little number of degenerate cases.

 2020-04-27, 16:30 #42 tgan   Jul 2015 17 Posts Yes i am getting this result also not sure if it is good enough
2020-04-27, 18:31   #43
yae9911

"Hugo"
Jul 2019
Germany

31 Posts

Quote:
 Originally Posted by tgan Yes i am getting this result also not sure if it is good enough
The optimum result of 70.007... for 10 days and 8 nodes corresponds to only one unlabeled connected graph. There are several ways to chose the root node in this graph all leading to the same probability of "all nodes infected". Since the updated description says
Quote:
 We were seeking the answer of Day 19 (which can give a solution with a better approximation), but we also accepted answers for Day 10.
is written in past tense, it is not clear if this means that they are still accepting the best result for 10 days, which is not accurate enough when rounded to 70.01.

2020-04-27, 20:35   #44
Dieter

Oct 2017

2·72 Posts

Quote:
 Originally Posted by yae9911 The optimum result of 70.007... for 10 days and 8 nodes corresponds to only one unlabeled connected graph. There are several ways to chose the root node in this graph all leading to the same probability of "all nodes infected". Since the updated description says is written in past tense, it is not clear if this means that they are still accepting the best result for 10 days, which is not accurate enough when rounded to 70.01.
We were seeking the answer of Day 19 (which can give a solution with a better approximation), but we also accepted answers for Day 10.

Could that be „Konjunktiv“? We also would accept... (wir würden auch...akzeptieren)? A question for a native speaker!

Last fiddled with by Dieter on 2020-04-27 at 20:38 Reason: Complement

 Similar Threads Thread Thread Starter Forum Replies Last Post ewmayer Soap Box 542 2020-11-25 21:42 gd_barnes Conjectures 'R Us 4 2020-09-23 08:59 what Puzzles 1 2020-04-24 05:46 what Puzzles 20 2020-03-04 07:55 what Puzzles 21 2020-02-02 14:11

All times are UTC. The time now is 13:44.

Thu Nov 26 13:44:03 UTC 2020 up 77 days, 10:55, 3 users, load averages: 0.90, 1.25, 1.40