Which combination of five players has the largest combined scores subject to the condition that their combined ELO's does not exceed 11,000?

I have a dataset which is something like the following:

Name, ELO, Score
Derek Aufderhar, 2134, 1
Hadley Kuhn, 2044, 0
Myrtie Lueilwitz, 2207, 2
Mitchell Schiller, 2036, 2
Javier Walter MD, 2485, 4
Waino Leuschke, 2486, 2
Ariel Jacobson, 2015, 3
Melvin Bailey, 2485, 0
Dovie Emmerich, 2383, 4
Adrian Stroman Jr., 2180, 1
Helen Douglas, 2352, 4
Yessenia O'Reilly, 2247, 2

I'd like to find the 5 players where they scored the most (their total score) and their ELO sum doesn't pass a specific number (which is 11.000). The dataset is dynamic every time. Any tip on where to look for doing it?


Name, ELO, Score
Derek Aufderhar, 2134, 1
Hadley Kuhn, 2044, 0
Myrtie Lueilwitz, 2207, 2
Mitchell Schiller, 2036, 2
Javier Walter MD, 2485, 4
Waino Leuschke, 2486, 2
Ariel Jacobson, 2015, 3
Melvin Bailey, 2485, 0
Dovie Emmerich, 2383, 4
Adrian Stroman Jr., 2180, 1
Helen Douglas, 2352, 4
Yessenia O'Reilly, 2247, 2


2 Answers
2

Code

def best_five(players, max_elo)
players.combination(5).with_object({ names:, tot_scores: -1 }) do |arr, best|
names, elos, scores = arr.map(&:values).transpose
best.replace({ names: names, tot_scores: scores.sum }) unless
elos.sum > max_elo || scores.sum <= best[:tot_scores]
end
end


Here players is an array of hashes, each with keys :name, :elo and score, where the value of :name is a string and values of the other two keys are integers.

players

:name

:elo

score

:name

Example

players =<<_
Derek Aufderhar, 2134, 1
Hadley Kuhn, 2044, 0
Myrtie Lueilwitz, 2207, 2
Mitchell Schiller, 2036, 2
Javier Walter MD, 2485, 4
Waino Leuschke, 2486, 2
Ariel Jacobson, 2015, 3
Melvin Bailey, 2485, 0
Dovie Emmerich, 2383, 4
Adrian Stroman Jr., 2180, 1
Helen Douglas, 2352, 4
Yessenia O’Reilly, 2247, 2
_


It is convenient to convert this string to a hash, both to address the current problem and to perform other operations with the data.

players_by_name = players.each_line.with_object({}) do |line, h|
name, elo, score = line.split(',')
h[name] = { name: name, elo: elo.to_i, score: score.to_i }
end
#=> {"Derek Aufderhar" =>{:name=>"Derek Aufderhar", :elo=>2134, :score=>1},
# "Hadley Kuhn" =>{:name=>"Hadley Kuhn", :elo=>2044, :score=>0},
# ...
# "Yessenia O’Reilly"=>{:name=>"Yessenia O’Reilly", :elo=>2247, :score=>2}}


We may now compute the best five for max_elo = 11000:

max_elo = 11000

best = best_five(players_by_name.values, 11000)
#=> {:names=>["Myrtie Lueilwitz", "Mitchell Schiller", "Ariel Jacobson",
# "Dovie Emmerich", "Helen Douglas"],
# :tot_scores=>15}


To retrieve information for these five players we compute the following:

a = players_by_name.values_at(*best[:names])
#=> [{:name=>"Myrtie Lueilwitz" , :elo=>2207, :score=>2},
# {:name=>"Mitchell Schiller", :elo=>2036, :score=>2},
# {:name=>"Ariel Jacobson" , :elo=>2015, :score=>3},
# {:name=>"Dovie Emmerich" , :elo=>2383, :score=>4},
# {:name=>"Helen Douglas" , :elo=>2352, :score=>4}]


We already know the scores sum to 15. As

a.map { |h| h[:elo] }.sum
#=> 10993


we see that the combined ELO limit is not exceeded.

Array#sum made its debut in Ruby v2.4.

If the dataset is not huge, here you go (possibly quite inefficient):

data =
%|Derek Aufderhar, 2134, 1
Hadley Kuhn, 2044, 0
Myrtie Lueilwitz, 2207, 2
Mitchell Schiller, 2036, 2
Javier Walter MD, 2485, 4
Waino Leuschke, 2486, 2
Ariel Jacobson, 2015, 3
Melvin Bailey, 2485, 0
Dovie Emmerich, 2383, 4
Adrian Stroman Jr., 2180, 1
Helen Douglas, 2352, 4
Yessenia O’Reilly, 2247, 2|

# unnecessary: transform to hash for clarity
values =
data.
split($/).
map { |e| e.split(',') }.
map { |name, elo, score| {name: name, elo: elo.to_i, score: score.to_i } }

# find the top
values.
permutation(5).
reject { |data| data.map { |e| e[:elo] }.inject(:+) > 11_000 }.
max { |data| data.map { |e| e[:score] }.inject(:+) }
#⇒ [{:name=>"Yessenia O'Reilly", :elo=>2247, :score=>2},
# {:name=>"Helen Douglas", :elo=>2352, :score=>4},
# {:name=>"Adrian Stroman Jr.", :elo=>2180, :score=>1},
# {:name=>"Ariel Jacobson", :elo=>2015, :score=>3},
# {:name=>"Mitchell Schiller", :elo=>2036, :score=>2}]


By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.