def test_K():
tot_correct = []
K_range = range(10, 401, 10)
for K in K_range:
teams, games = parse_data(file('scores.dat'))
do_sauceda(games, K)
correct = 0.
incorrect = 0.
even = 0
for game in games:
if game.w_idx < 6 or game.l_idx < 6: continue
w_rank = round(game.winner.history[game.w_idx], 0)
l_rank = round(game.loser.history[game.l_idx], 0)
if w_rank > l_rank: correct += 1
elif w_rank < l_rank: incorrect += 1
else: even += 1
#print "------------\nK at %d" % K
#print "Percent correct: %f%% Even: %d %d" % (correct/len(games), even, correct)
tot_correct.append(correct / len(games))
g.plot(K_range, tot_correct, '-b')
g.show()
def test_we(K):
teams, games = parse_data(file('scores.dat'))
do_sauceda(games, K)
r_strata = [(0., 0., 0) for i in range(50)] #pd, rd, n of games
for game in games:
#is this calc correct? I'm not sure
w_rank = int(round(game.winner.history[game.w_idx]))
l_rank = int(round(game.loser.history[game.l_idx]))
if w_rank < l_rank: pd = -game.pd
else: pd = game.pd
r = w_rank, l_rank
rd = max(r) - min(r)
temp_pd, temp_rd, temp_n = r_strata[rd / 50]
r_strata[rd / 50] = (temp_pd + pd, temp_rd + rd, temp_n + 1)
for i, (pd, rd, n) in enumerate(r_strata):
if n > 0:
print "%3d-%3d: %f %d %f (%d)" % \
(i * 50, (i * 50) + 50, pd / rd, pd, rd, n)
pd_tot = sum(r[0] for r in r_strata)
rd_tot = sum(r[1] for r in r_strata)
print "total: %f (%d)" % (pd_tot / rd_tot, rd_tot)
def test_K():
tot_correct = []
K_range = range(10, 401, 10)
for K in K_range:
teams, games = parse_data(file('scores.dat'))
do_sauceda(games, K)
correct = 0.
incorrect = 0.
even = 0
for game in games:
if game.w_idx < 6 or game.l_idx < 6: continue
w_rank = round(game.winner.history[game.w_idx], 0)
l_rank = round(game.loser.history[game.l_idx], 0)
if w_rank > l_rank: correct += 1
elif w_rank < l_rank: incorrect += 1
else: even += 1
#print "------------\nK at %d" % K
#print "Percent correct: %f%% Even: %d %d" % (correct/len(games), even, correct)
tot_correct.append(correct/len(games))
g.plot(K_range, tot_correct, '-b')
g.show()
def test_we(K):
teams, games = parse_data(file('scores.dat'))
do_sauceda(games, K)
r_strata = [(0., 0., 0) for i in range(50)] #pd, rd, n of games
for game in games:
#is this calc correct? I'm not sure
w_rank = int(round(game.winner.history[game.w_idx]))
l_rank = int(round(game.loser.history[game.l_idx]))
if w_rank < l_rank: pd = -game.pd
else: pd = game.pd
r = w_rank, l_rank
rd = max(r) - min(r)
temp_pd, temp_rd, temp_n = r_strata[rd / 50]
r_strata[rd / 50] = (temp_pd + pd, temp_rd + rd, temp_n + 1)
for i, (pd, rd, n) in enumerate(r_strata):
if n > 0:
print "%3d-%3d: %f %d %f (%d)" % \
(i * 50, (i * 50) + 50, pd / rd, pd, rd, n)
pd_tot = sum(r[0] for r in r_strata)
rd_tot = sum(r[1] for r in r_strata)
print "total: %f (%d)" % (pd_tot/rd_tot, rd_tot)
def test_value(K):
teams, games = parse_data(file('scores.dat'))
do_sauceda(games, K)
for game in games:
pass
def test_value(K):
teams, games = parse_data(file('scores.dat'))
do_sauceda(games, K)
for game in games:
pass
