def main():
# grid = g.read_grid('grid.txt')
grid = g.make_grid(10, 10)
# Print grid with a space between columns and a newline between rows
print('\n'.join(' '.join([str(col) for col in row]) for row in grid))
print()
start = get_user_coords(grid, 'start')
end = get_user_coords(grid, 'goal')
# Prompt user to choose breadth-first or depth-first search
while True:
print('Choose breadth-first (0) or depth-first (1) search:', end=' ')
try:
# Default to breadth-first unless 1 entered
breadth = int(input()) != 1
break
except ValueError:
print('Invalid selection. Please enter 0 or 1')
path = uninformed_search(grid, start, end, breadth)
fname = 'path.txt'
if path is None:
print('No path found.')
else:
g.output_grid(fname, grid, start, end, path)
def call(self, inputs):
if type(inputs) is not list or len(inputs) != 2:
raise Exception(
"Homography must be called on a list of 2 tensors, image and homography. Got: "
+ str(inputs)
)
images = inputs[0]
grids = make_grid(*images.shape)
homos = inputs[1]
transformed_grids = apply_homography(homos, grids)
return bilinear_sampler(images, transformed_grids)
def error_S(m_2, N_, delta_t_, theta_, l_m1_, true_price_):
l_err_DO = []
l_err_CS = []
l_err_MCS = []
l_err_HV = []
for m_1 in l_m1_:
m_0 = (m_1 + 1) * (m_2 + 1)
l_s, d_s, l_v, d_v, _, _ = grid.make_grid(m_1, S, S_0, K, c, m_2, V,
V_0, d)
a_0, a_1, a_2, a = fac.make_matrices(m_1, m_2, m_0, rho, sigma, r_d,
r_f, kappa, eta, l_s, l_v, d_s,
d_v)
b_0, b_1, b_2, b = fac.make_boundaries(m_1, m_2, m_0, r_d, r_f, N_,
l_s, delta_t_)
uu_0 = np.array([[max(l_s[i] - K, 0) for i in range(m_1 + 1)]
for _ in range(m_2 + 1)])
u_0 = uu_0.flatten()
idx_s = l_s.index(S_0)
idx_v = l_v.index(V_0)
price_DO = solver.DO_scheme(m_0, N_, u_0, delta_t_, theta_, a, a_0,
a_1, a_2, b, b_0, b_1, b_2, r_f)[0]
price_DO = np.reshape(price_DO, (m_2 + 1, m_1 + 1))
price_CS = solver.CS_scheme(m_0, N_, u_0, delta_t_, theta_, a, a_0,
a_1, a_2, b, b_0, b_1, b_2, r_f)[0]
price_CS = np.reshape(price_CS, (m_2 + 1, m_1 + 1))
price_MCS = solver.MCS_scheme(m_0, N_, u_0, delta_t_, theta_, a, a_0,
a_1, a_2, b, b_0, b_1, b_2, r_f)[0]
price_MCS = np.reshape(price_MCS, (m_2 + 1, m_1 + 1))
price_HV = solver.HV_scheme(m_0, N_, u_0, delta_t_, theta_, a, a_0,
a_1, a_2, b, b_0, b_1, b_2, r_f)[0]
price_HV = np.reshape(price_HV, (m_2 + 1, m_1 + 1))
l_err_DO.append(
abs(price_DO[idx_v, idx_s] - true_price_) / true_price_)
l_err_CS.append(
abs(price_CS[idx_v, idx_s] - true_price_) / true_price_)
l_err_MCS.append(
abs(price_MCS[idx_v, idx_s] - true_price_) / true_price_)
l_err_HV.append(
abs(price_HV[idx_v, idx_s] - true_price_) / true_price_)
return l_m1_, l_err_DO, l_err_CS, l_err_MCS, l_err_HV
def __init__(self, *args, **kwargs):
"""
GridSearchCVParallel accepts all the parameters for GridSearchCV from sklearn,
but the has some additional parameters in the constructor
callback: callable, optional
Function to be called every every iteration
Must accept 3 parameters: number of task done, total number of tasks and time elapsed (seconds)
cacher: dictionary or CacherABC subclass, optional
mapper: callable, optional: a map function (function, iterable). if not provided, itertools.imap will be used
loader: callable, optional
Parameterless function to be called on remote machines to load data.
Must return X,y as a tuple, must handle its own imports
fit_callback: callable, optional: a function to be called after _fit_and_score,
parameters: index, {estimator, X, y, parameters, fit_params, train, test, scorer}
"""
self.callback = kwargs.pop('callback', None)
self.mapper = kwargs.pop('mapper', itertools.imap)
self.cacher = kwargs.pop('cacher', {})
self.loader = kwargs.pop('loader', None)
self.fit_callback = kwargs.pop('fit_callback', None)
self.max_iter = kwargs.pop('max_iter', 100)
composite_grid = args[1]
if isinstance(composite_grid, dict):
grid = composite_grid
self.transforms = {}
else:
grid, self.transforms = make_grid(
*split_constraints_and_transforms(composite_grid),
max_iter=self.max_iter)
args = list(args)
print('grid length: %d' % len(ParameterGrid(grid)))
args[1] = grid
args = tuple(args)
super(GridSearchCVParallel, self).__init__(*args, **kwargs)
true_price = 8.8948693600540167 # from Monte Carlo simulation (ref: https://github.com/RedwanBouizi/MC-Heston)
# grid [0, S] x [0, V]
m1 = 50 # S
m2 = 25 # V
m = (m1 + 1) * (m2 + 1) # matrix A and vector U size
c = K / 5
d = V / 500
# line [0, T]
N = 20
delta_t = T / N
# model setup
Vec_s, Delta_s, Vec_v, Delta_v, X, Y = grid.make_grid(m1, S, S_0, K, c, m2, V,
V_0, d)
A_0, A_1, A_2, A = fac.make_matrices(m1, m2, m, rho, sigma, r_d, r_f, kappa,
eta, Vec_s, Vec_v, Delta_s, Delta_v)
B_0, B_1, B_2, B = fac.make_boundaries(m1, m2, m, r_d, r_f, N, Vec_s, delta_t)
# pricing
print("--True Price", true_price)
theta = 0.8
print("\n--CN Scheme")
UU_0 = np.array([[max(Vec_s[i] - K, 0) for i in range(m1 + 1)]
for j in range(m2 + 1)])
U_0 = UU_0.flatten()
price, time = solver.CN_scheme(m, N, U_0, delta_t, A, B, r_f)
done = False
clock = pygame.time.Clock()
# -------- Main Program Loop -----------
while not done:
# --- Main event loop
for event in pygame.event.get():
if event.type == pygame.QUIT:
done = True
# start working
visited = []
screen.fill(Grid.BLACK)
pygame.display.flip()
# Limit to 60 frames per second
clock.tick(60)
# Close the window and quit.
pygame.quit()
#create new Grid object for testing
input_grid = np.zeros(shape=(20, 20))
Grid = make_grid(input_grid)
Grid.print_grid()
traversal(Grid)
pg.image.load('./assets/sad_background.png'), (1650, 1000))
font = pg.font.Font('freesansbold.ttf', 50)
clockfont = pg.font.Font('./assets/clock_font.ttf', 90)
fontcols = {
1: (0, 0, 255),
2: (0, 128, 0),
3: (255, 0, 0),
4: (2, 13, 113),
5: (77, 0, 0),
6: (0, 139, 139),
7: (0, 0, 0),
8: (131, 131, 131),
'*': (255, 0, 0)
}
g = make_grid(16, 32, 80)
g2 = make_blank_grid(16, 32, 80)
flags = make_blank_grid(16, 32, 80)
flagcount = 80
flagprifix = '0'
timer = 0
timerprifix = '00'
clock = pg.time.Clock()
pg.time.set_timer(pg.USEREVENT, 1000)
play = True
start = False
lost = False
def printgrid():
global lost
