Ttykuu

I am writing a script to help me analyze and get statistics for poker hands. After a lot of research I was able to work with bit manipulation and get my statistics very efficiently for one hand. I want, thought, to be able to run it for several times (millions) and I don't know how to to this efficiently. I tried using multiprocess map (probably wrong) and it slowed the script down significantly.

This is what I had originally:

def get_best_combination(hand, board):

 candidate_hand = []
 best_score = 9999 
 best_rank = -1
 for h in itertools.combinations(hand, 2):
 for b in itertools.combinations(board,3):
 candidate_hand = h + b
 score = eval_hand(candidate_hand)
 if (score < best_score):
 best_score = score
 best_rank = hand_rank(best_score)
 return best_score, best_rank

What the function is doing:
- It combines 2 cards from hand = [Card1, Card2, Card3, Card4, Card5] and 3 cards from board = [Card1, Card2, Card3, Card4, Card5]
- Evaluates the combination and assign a score for it
- Stores and returns the best score and best rank for all possible combinations between hand and board (total of 100 combinations).

This runs really fast for (1 hand, 1 board) (0.02s) but I want to call this function for a lot of different hands and a lot of different boards for every hand. Tried to implement multiprocessing for this but for my surprise it slowed down to almost 40s.

This is what I tried:

def get_best_combination(hand, board):

 candidate_hand = []
 best_score = 9999 
 best_rank = -1
 p = Pool()
 for h in itertools.combinations(hand, 2):
 for b in itertools.combinations(board,3):
 candidate_hand.append(h+b)
 results = p.map(eval_hand, candidate_hand)
 for score in results:
 if (score < best_score):
 best_score = score
 best_rank = hand_rank(best_score)
 return best_score, best_rank

What is wrong with it? Whys is it slowing the script down instead of speeding it up?

Thanks!