def exp_real_with_1_comp():
# REAL
data = np.load('data/real/SCEDC-1999-2019-24hrs.npy')
data = data[:, 1:, :3]
params = np.load('results/real-24hrs-1-gcomp.npz')
da = utils.DataAdapter(init_data=data)
mu = params['mu']
beta = params['beta']
kernel = GaussianMixtureDiffusionKernel(n_comp=1,
layers=[10],
C=1.,
beta=beta,
SIGMA_SHIFT=.1,
SIGMA_SCALE=.5,
MU_SCALE=.01,
Wss=params['Wss'],
bss=params['bss'],
Wphis=params['Wphis'])
lam = HawkesLam(mu, kernel, maximum=1e+3)
print("mu", mu)
print("beta", beta)
utils.plot_spatial_kernel(
"results/learned-kernel-SCEDC-1999-2019-24hrs.pdf",
kernel.gdks[0],
S=[[-1., 1.], [-1., 1.]],
grid_size=50)
utils.spatial_intensity_on_map(
"results/map-SCEDC-1999-2019-24hrs.html",
da,
lam,
data,
seq_ind=3000,
t=8.0,
# xlim=[-23.226, -22.621],
# ylim=[-43.868, -43.050],
# ngrid=200)
xlim=da.xlim,
ylim=da.ylim,
ngrid=200)
def exp_real_with_2_comp():
# REAL
data = np.load(
'../Spatio-Temporal-Point-Process-Simulator/data/rescale.ambulance.perday.npy'
)
data = data[:, 1:, :3]
params = np.load(
'../Spatio-Temporal-Point-Process-Simulator/data/rescale_ambulance_mle_gaussian_mixture_params.npz'
)
da = utils.DataAdapter(init_data=data)
mu = .1 # params['mu']
beta = params['beta']
kernel = GaussianMixtureDiffusionKernel(n_comp=1,
layers=[5],
C=1.,
beta=beta,
SIGMA_SHIFT=.1,
SIGMA_SCALE=.5,
MU_SCALE=.01,
Wss=params['Wss'],
bss=params['bss'],
Wphis=params['Wphis'])
lam = HawkesLam(mu, kernel, maximum=1e+3)
print("mu", mu)
print("beta", beta)
print(params['Wphis'].shape)
pp = SpatialTemporalPointProcess(lam)
# generate points
points, sizes = pp.generate(T=[0., 10.],
S=[[-1., 1.], [-1., 1.]],
batch_size=100,
verbose=True)
results = da.restore(points)
print(results)
print(sizes)
np.save('results/ambulance-simulation.npy', results)
print('[%s] Train cost:\t%f' % (arrow.now(), avg_train_cost),
file=sys.stderr)
print('[%s] Test cost:\t%f' % (arrow.now(), avg_test_cost),
file=sys.stderr)
if __name__ == "__main__":
np.set_printoptions(suppress=True)
# np.random.seed(1)
# tf.set_random_seed(1)
with tf.Session() as sess:
# data preparation
data = np.load("data/northcal.earthquake.perseason.npy")
da = utils.DataAdapter(init_data=data,
S=[[-1., 1.], [-1., 1.]],
T=[0., 1.])
data = da.normalize(data)[:, 1:51, :]
mask = data == 0.
mask = mask.astype(float)
data = data + mask
print(data)
# print(data.shape)
# model configurations
lstm_hidden_size = 10
# training configurations
step_size = np.shape(data)[1]
batch_size = 5
test_ratio = 0.3
epoches = 30
# save all training cost into numpy file.
np.savetxt("results/robbery_mle_train_cost.txt",
all_train_cost,
delimiter=",")
if __name__ == "__main__":
# Unittest example
S = [[-1., 1.], [-1., 1.]]
T = [0., 10.]
data = np.load(
'../Spatio-Temporal-Point-Process-Simulator/data/rescale.ambulance.perday.npy'
)
data = data[:320,
1:51, :] # remove the first element in each seqs, since t = 0
da = utils.DataAdapter(init_data=data, S=S, T=T)
# data = np.load('../Spatio-Temporal-Point-Process-Simulator/data/northcal.earthquake.perseason.npy')
# da = utils.DataAdapter(init_data=data)
seqs = da.normalize(data)
print(da)
print(seqs.shape)
# training model
with tf.Session() as sess:
batch_size = 32
epoches = 10
layers = [5]
n_comp = 5
ppg = MLE_Hawkes_Generator(T=T,
S=S,
np.savetxt("results/robbery_rl_train_cost.txt",
all_train_cost,
delimiter=",")
if __name__ == "__main__":
# Unittest example
# np.random.seed(0)
# tf.set_random_seed(1)
data = np.load(
'../Spatio-Temporal-Point-Process-Simulator/data/apd.robbery.permonth.npy'
)
# data = np.load('../Spatio-Temporal-Point-Process-Simulator/data/northcal.earthquake.perseason.npy')
da = utils.DataAdapter(init_data=data)
seqs = da.normalize(data)
seqs = seqs[:, 1:, :] # remove the first element in each seqs, since t = 0
print(da)
print(seqs.shape)
# training model
with tf.Session() as sess:
# model configuration
batch_size = 10
epoches = 30
lr = 1e-3
T = [0., 10.]
S = [[-1., 1.], [-1., 1.]]
layers = [5]
n_comp = 5