def alphabeta(depth,
tt,
net_score=0,
alpha=-maxint,
beta=maxint,
max_node=True):
if depth == 0:
# TODO do we need static evaluator? how should we scale its value
return net_score, []
best_move = []
undo_stack = []
if max_node:
v = -maxint - 1 # negative infinity
for move in gen_moves.gen_moves(tt, debug=True):
tt.cur_player = 1
#print("max move %s: " % move)
meeple_count = make_move(move, tt.board, undo_stack)
last_step = move[-1]
score = net_score + scoring.calc_score(last_step, meeple_count,
undo_stack, tt)
#print("scored: %s" % score)
value = alphabeta(depth - 1, tt, score, alpha, beta, False)[0]
#print("after alphabeta: %s" % value)
unmake(undo_stack)
if value > v:
v = value
best_move = move
#print("new max: %d, %s" % (v, best_move))
alpha = max(alpha, v)
if beta <= alpha:
break
return v, best_move
else:
v = maxint # positive infinity
for move in gen_moves.gen_moves(tt):
tt.cur_player = 2
#print("min move %s: " % move)
meeple_count = make_move(move, tt.board, undo_stack)
last_step = move[-1]
score = net_score - scoring.calc_score(last_step, meeple_count,
undo_stack, tt)
#print("min scored: %s" % score)
value = alphabeta(depth - 1, tt, score, alpha, beta, True)[0]
#print("after alphabeta: %s" % value)
unmake(undo_stack)
if value < v:
v = value
best_move = move
#print("new min: %d, %s" % (v, best_move))
beta = min(beta, v)
if alpha >= beta:
break
return v, best_move
def greedy_solution(images,
callback=None,
callback_score_increment=1000,
thresh=0):
slides = [Slide((im, )) for im in images]
solution = [slides.pop(0)]
aaa = 0
last_callback_score = aaa
while len(slides) > 0:
max_score = 0
best = None
side = None
ind_best = 0
pos_right = len(solution[-1].tags) / 2
pos_left = len(solution[0].tags) / 2
for idx, s in enumerate(slides):
sr = calc_score([solution[-1], s])
if sr > max_score:
best = s
ind_best = idx
side = 'r'
max_score = sr
if max_score > min(thresh, pos_right):
break
sl = calc_score([s, solution[0]])
if sl > max_score:
best = s
side = 'l'
ind_best = idx
max_score = sl
if max_score > min(thresh, pos_left):
break
if best is None:
solution += [slides.pop(0)]
else:
if side == 'l':
solution = ([best] + solution)
else:
solution = (solution + [best])
slides.pop(ind_best)
aaa += max_score
print(len(slides), ' ', aaa)
if callback is not None and aaa > last_callback_score + callback_score_increment:
last_callback_score = aaa
callback(solution)
return solution
def post(self, game_id):
# TODO: is the submitted city part of the game?
# TODO: did the player already submit a guess for this city?
request_data = request.get_json()
player_id = request_data['player_id']
player_key = ndb.Key(urlsafe=player_id)
city_id = request_data['city_id']
game_key = ndb.Key(urlsafe=game_id)
game = game_key.get()
city_key = ndb.Key(urlsafe=city_id)
city = city_key.get()
if game.state == GameState.waitingForPlayers:
game.state = GameState.running
game.put()
guess = Guess(parent=game_key)
guess.game = game_key
guess.player = player_key
guess.city = city_key
guess.long = request_data['long']
guess.lat = request_data['lat']
remaining_time = request_data['remaining_time']
guess.score = calc_score("easy", city.lat, city.long, remaining_time, guess.lat, guess.long)
guess.put()
for player in game.players:
# token = Util.get_token(player.urlsafe())
# print("sending message to" + token)
msg = Messages()
msg.player = player
msg.game = game_key
msg.msg = str(Util.to_json(guess, False, False))
msg.put()
# channel.send_message(token, str(Util.to_json(guess, False, False)))
# TODO this is not stable yet - we have to make sure only one guess per city/player pair can be submitted
guess_query = Guess.query(Guess.player == player_key, ancestor=game_key)
if guess_query.count() == len(game.cities):
game.state = GameState.done
game.put()
save_score(game_key, player_id, guess_query.fetch())
return Util.to_json(guess, False), 201
def get_most_similar_restricted(limit=120):
df, df_i = import_data()
counts = df_i['ingredient'].value_counts()
ings = counts.index.values
found_ings, embeddings = retrieve_embeddings(model, ings)
#found_ings, embeddings = np.load('word2vec_embeddings.npy')
ranks = get_nearest_neighbors(embeddings)
print_nearest_neighbors(ings[:limit], found_ings, ranks)
highest_ranks, avg_rankings, random_avg_rankings = calc_score(ranks, limit,
print_scores=False, score_path='../model/scores.csv')
indices = found_ings[found_ings<highest_ranks.shape[0]]
highest_ranks = highest_ranks[indices]
avg_rankings = avg_rankings[indices]
random_avg_rankings = random_avg_rankings[indices]
print (highest_ranks<=3).sum(dtype=float) / np.isfinite(highest_ranks).sum()
print highest_ranks[np.isfinite(highest_ranks)].mean()
print avg_rankings[np.isfinite(avg_rankings)].mean()
print random_avg_rankings[np.isfinite(random_avg_rankings)].mean()
def minimax(depth, tt):
if depth == 0:
# TODO do we need static evaluator? how should we scale its value
return 0, []
best_move = []
max_score = -maxint - 1 # negative infinity
for move in gen_moves.gen_moves(tt):
undo_stack = []
meeple_count = make_move(move, tt.board, undo_stack)
last_step = move[-1]
score = scoring.calc_score(last_step, meeple_count, undo_stack, tt)\
- minimax(depth - 1, tt)[0]
unmake(undo_stack)
if (score > max_score):
print("better score: %s" % score)
max_score = score
best_move = move
return max_score, best_move
def main():
num_ingredients = 1000
ings_per_prod = None
use_embeddings = False
output_cat = 'shelf'
df, df_i = import_data()
counts = df_i['ingredient'].value_counts()
inputs_, outputs, idx_to_cat = gen_input_outputs_cat(
df, counts, num_ingredients, output_cat, ings_per_prod)
if use_embeddings:
#embeddings = np.load('embeddings/embeddings_{}.npy'.format(num_ingredients))
#embeddings = np.load('../word2vec/word2vec_embeddings.npy')[1][:num_ingredients]
embeddings = 2*np.random.random((num_ingredients, 300))-1 # Try random embeddings
embeddings = embeddings.astype('float32')
inputs = input_from_embeddings(inputs_, embeddings,
normalize=False)
#inputs = (2*np.random.random(inputs_.shape)-1).astype('float32') # Completely random inputs.
else:
inputs = inputs_
num_outputs = outputs.max()+1
print "# of data points:", len(inputs)
# Scramble inputs/outputs
np.random.seed(3)
random_idx = np.random.permutation(len(inputs))
inputs = inputs[random_idx]
outputs = outputs[random_idx]
test_split_idx = int(len(inputs))*0.8
inputs, X_test = inputs[:test_split_idx], inputs[test_split_idx:]
outputs, y_test = outputs[:test_split_idx], outputs[test_split_idx:]
X_train, X_valid, y_train, y_valid = train_test_split(
inputs, outputs, test_size=0.2, random_state=42)
print "Running models..."
# Max entropy model
# Normalize
#inputs_n = inputs / np.sum(inputs, axis=1)[:,None]
#regr = max_entropy(X_train, y_train, X_valid, y_valid)
#predict_cat(counts, regr, idx_to_cat, num_ingredients, ings)
# Neural network model
classifier, predict_model = run_nn(X_train, y_train, X_valid, y_valid,
X_train.shape[1], num_outputs,
m=100, n_epochs=10, batch_size=10,
learning_rate=0.1, L2_reg=0.0001)
pred_valid = predict_model(X_valid)
pred_test = predict_model(X_test)
if output_cat != 'aisle':
lower_to_upper_cat = get_upper_cat(df, output_cat, 'aisle')
print "Validation set:\n", calc_accuracy(pred_valid, y_valid)
print calc_accuracy(pred_valid, y_valid, lower_to_upper_cat)
print "Test set:\n", calc_accuracy(pred_test, y_test)
print calc_accuracy(pred_test, y_test, lower_to_upper_cat)
#print_predictions(X_valid, y_valid,
# np.argmax(pred_valid, axis=1), idx_to_cat, counts, limit=100)
if not use_embeddings:
embeddings_out = classifier.hiddenLayer.W.get_value()
ranks, neigh = get_nearest_neighbors(embeddings_out)
#print_nearest_neighbors(counts.index.values[:num_ingredients], ranks)
highest_rank, score, avg_rank_of_ing_cat, random_score = calc_score(
ranks, num_ingredients)
def main():
num_ingredients = 1000
use_embeddings = False
ings_per_prod = 5
frac_weighted = 0.95
invalid_multiplier = 1
df, df_i = import_data()
counts = df_i['ingredient'].value_counts()
inputs_v_, outputs_v = gen_input_outputs_valid(
df, df_i, num_ingredients, ings_per_prod)
inputs_i_w_, outputs_i_w = gen_input_outputs_invalid(inputs_v_,
invalid_multiplier*frac_weighted,
num_ingredients, ings_per_prod, weighted=True)
inputs_i_, outputs_i = gen_input_outputs_invalid(inputs_v_,
invalid_multiplier*(1-frac_weighted),
num_ingredients, ings_per_prod, weighted=False)
if use_embeddings:
embeddings = np.load('embeddings/embeddings_{}.npy'.format(num_ingredients))
#embeddings = np.load('../word2vec/word2vec_embeddings.npy')[1][:num_ingredients]
#embeddings = 2*np.random.random((num_ingredients, 20))-1 # Try random embeddings
embeddings = embeddings.astype('float32')
normalize = False
inputs_v = input_from_embeddings(inputs_v_, embeddings,
normalize=normalize)
inputs_i_w = input_from_embeddings(inputs_i_w_, embeddings,
normalize=normalize)
inputs_i = input_from_embeddings(inputs_i_, embeddings,
normalize=normalize)
else:
inputs_v = inputs_v_
inputs_i_w = inputs_i_w_
inputs_i = inputs_i_
X_train_v, X_test_v, y_train_v, y_test_v = train_test_split(
inputs_v, outputs_v, test_size=1/3., random_state=42)
X_train_i_w, X_test_i_w, y_train_i_w, y_test_i_w = train_test_split(
inputs_i_w, outputs_i_w, test_size=1/3., random_state=42)
X_train_i, X_test_i, y_train_i, y_test_i = train_test_split(
inputs_i, outputs_i, test_size=1/3., random_state=42)
X_train = np.vstack((X_train_v, X_train_i_w, X_train_i))
y_train = np.hstack((y_train_v, y_train_i_w, y_train_i))
X_test = np.vstack((X_test_v, X_test_i_w, X_test_i))
y_test = np.hstack((y_test_v, y_test_i_w, y_test_i))
# Scramble inputs/outputs
np.random.seed(3)
random_idx_tr = np.random.permutation(len(X_train))
random_idx_te = np.random.permutation(len(X_test))
X_train = X_train[random_idx_tr]
y_train = y_train[random_idx_tr]
X_test = X_test[random_idx_te]
y_test = y_test[random_idx_te]
print "Running models..."
# Max entropy model
#regr = max_entropy(X_train, y_train, X_test, y_test)
#predict_cat(counts, regr, idx_to_cat, num_ingredients, ings)
# Neural network model
classifier, predict_model = run_nn(X_train, y_train, X_test, y_test,
X_train.shape[1], num_outputs=2,
m=20, n_epochs=10, batch_size=10,
learning_rate=0.05, L2_reg=0.0001)
#pred = predict_model(X_test)
#pred_cats = np.argmax(pred, axis=1)
#print calc_accuracy(pred, y_test)
#print_predictions(X_test, y_test, pred_cats, counts, limit=100)
print "Max Entropy (Valid, Invalid, Invalid weighted):"
print calc_accuracy(regr.predict_proba(X_test_v), y_test_v)
print calc_accuracy(regr.predict_proba(X_test_i), y_test_i)
print calc_accuracy(regr.predict_proba(X_test_i_w), y_test_i_w)
print "Neural Network (Valid, Invalid, Invalid weighted):"
print calc_accuracy(predict_model(X_test_v), y_test_v)
print calc_accuracy(predict_model(X_test_i), y_test_i)
print calc_accuracy(predict_model(X_test_i_w), y_test_i_w)
if not use_embeddings:
embeddings_out = classifier.hiddenLayer.W.get_value()
ranks, neigh = get_nearest_neighbors(embeddings_out)
#print_nearest_neighbors(counts.index.values[:num_ingredients], ranks)
highest_rank, score, avg_rank_of_ing_cat, random_score = calc_score(
ranks, num_ingredients)
from scoring import calc_score
from simple_solutions import *
from hash_io import Input_parser
if __name__ == '__main__':
input_paths = ['input/a_example.txt',
'input/b_lovely_landscapes.txt',
'input/c_memorable_moments.txt',
'input/d_pet_pictures.txt',
'input/e_shiny_selfies.txt']
total_score = 0
for input_path in input_paths:
images = Input_parser(input_path).images
slides = dumb_solution(images)
total_score = total_score + calc_score(slides)
print(total_score)
