def test_profiling(self):
with cuda.profiling():
a = cuda.device_array(10)
del a
with cuda.profiling():
a = cuda.device_array(100)
del a
def test_profiling(self):
with cuda.profiling():
a = cuda.device_array(10)
del a
with cuda.profiling():
a = cuda.device_array(100)
del a
if flat_slice[j] < flat_slice[j + 1]:
flat_slice[j], flat_slice[j + 1] = flat_slice[j + 1], flat_slice[j]
if slice.size % 2 == 0:
upper = int(slice.size / 2)
lower = upper - 1
out[x, y, z] = (flat_slice[upper] + flat_slice[lower]) / 2
else:
out[x, y, z] = flat_slice[slice.size // 2]
im_array = np.array(Image.open("lab02img.bmp")).astype(np.uint8)
with allocated_gpu() as gpu:
threadsperblock = (
math.floor(math.sqrt(gpu.MAX_THREADS_PER_BLOCK / (2 * im_array.shape[2]))),
math.floor(math.sqrt(gpu.MAX_THREADS_PER_BLOCK / (2 * im_array.shape[2]))),
im_array.shape[2]
)
blockspergrid = (
math.ceil(im_array.shape[0] / threadsperblock[0]),
math.ceil(im_array.shape[1] / threadsperblock[1])
)
result = np.zeros(shape=(im_array.shape[0], im_array.shape[1], im_array.shape[2]))
with cuda.profiling(), timed("on GPU"):
calculate_gpu[blockspergrid, threadsperblock](im_array, result)
Image.fromarray(result.astype(np.uint8), 'RGB').save('lab02result.bmp')
N_runs = int(T // t_run)
#Adjust t_run:
t_run = T / N_runs
else:
N_runs = 1
t_run = T
#print('N_runs =', N_runs)
#Mean number of encounters in time t_run
N_mean = t_run * encounterRate(n_p, v_rel, b_min, b_max_max, v_min, v_max)
print('Evolving population, start time:', datetime.datetime.now())
for i in range(N_runs):
#Number of encounters
N_enc = np.random.poisson(N_mean, size=N_bin)
rng_states = create_xoroshiro128p_states(threadsperblock * blockspergrid,
seed=1)
with cuda.profiling():
run_encounters[blockspergrid,
threadsperblock](a, e, m1, m2, M_p, v_rel, a_T,
b_max_max, v_min, v_max, N_enc, delta,
G_new, N_broken, rng_states)
e = e[np.where(a > 0.0)]
a = a[np.where(a > 0.0)]
#N_broken = N_bin - np.size(a)
#print('a_final/au =', a)
#print('e_final =', e)
print('Finish time:', datetime.datetime.now())
print('N_broken =', N_broken)
print('Saving')
np.savez('simulation_GPU_{}Msol_{}e{}.npz'.format(
int(M_p * mass_unit / (2.0 * 10.0**30.0)),