iter=8000,
warmup=2000,
control=control,
thin=6
)
print(mcmc_result)
mcmc_result.plot()
plt.show()
# saving compiled model
if not os.path.exists('%s.pkl' % filename):
with open('%s.pkl' % filename, 'wb') as f:
pickle.dump(sm, f)
mcmc_sample = mcmc_result.extract(permuted=True)
# plot ssm of mu
tools.plot_ssm(mcmc_sample,
sales_df_3['date'],
'local liner trend',
'sales',
'mu',
sales_df_3['sales'])
# plot ssm of delta
tools.plot_ssm(mcmc_sample,
sales_df_3['date'],
'drift',
'delta',
'delta')
sm = pickle.load(open('%s.pkl' % filename, 'rb'))
# sm = pystan.StanModel(file='5-4-1-simple-reg.stan')
else:
# a model using prior for mu and sigma.
sm = pystan.StanModel(file='%s.stan' % filename)
control = {'adapt_delta': 0.8, 'max_treedepth': 10}
mcmc_result = sm.sampling(data=stan_data,
seed=1,
chains=4,
iter=8000,
warmup=2000,
control=control,
thin=6)
print(mcmc_result)
mcmc_result.plot()
plt.show()
# saving compiled model
if not os.path.exists('%s.pkl' % filename):
with open('%s.pkl' % filename, 'wb') as f:
pickle.dump(sm, f)
mcmc_sample = mcmc_result.extract(permuted=True)
# plot ssm of probs
tools.plot_ssm(mcmc_sample, np.arange(0, len(boat_df['x'])),
'autoregressive local trend', 'prob.', 'probs', boat_df['x'])
sm = pickle.load(open('%s.pkl' % filename, 'rb'))
# sm = pystan.StanModel(file='5-4-1-simple-reg.stan')
else:
# a model using prior for mu and sigma.
sm = pystan.StanModel(file='%s.stan' % filename)
control = {'adapt_delta': 0.8, 'max_treedepth': 16}
mcmc_result = sm.sampling(data=stan_data,
seed=1,
chains=4,
iter=8000,
warmup=2000,
control=control,
thin=6)
print(mcmc_result)
mcmc_result.plot()
plt.show()
# saving compiled model
if not os.path.exists('%s.pkl' % filename):
with open('%s.pkl' % filename, 'wb') as f:
pickle.dump(sm, f)
mcmc_sample = mcmc_result.extract(permuted=True)
# plot ssm of mu
tools.plot_ssm(mcmc_sample, sales_df_5['date'], 'autoregressive local trend',
'sales', 'alpha', sales_df_5['sales'])
control=control,
thin=1
)
print(mcmc_result)
mcmc_result.plot()
plt.show()
# saving compiled model
if not os.path.exists('5-3-2-local-level-interpolation.pkl'):
with open('5-3-2-local-level-interpolation.pkl', 'wb') as f:
pickle.dump(sm, f)
mcmc_sample = mcmc_result.extract()
# plot ssm
tools.plot_ssm(mcmc_sample,
sales_df['date'],
'local level model',
'sales',
'mu',
sales_df['sales'])
# plot ssm about prediction
tools.plot_ssm(mcmc_sample,
sales_df['date'],
'local level model',
'sales',
'y_pred',
sales_df['sales'])
print(mcmc_result)
mcmc_result.plot()
plt.show()
# saving compiled model
if not os.path.exists('%s.pkl' % filename):
with open('%s.pkl' % filename, 'wb') as f:
pickle.dump(sm, f)
mcmc_sample = mcmc_result.extract(permuted=True)
# plot ssm of alpha (trend and cycle)
tools.plot_ssm(mcmc_sample,
sales_df_4['date'],
'all state',
'sales',
'alpha',
sales_df_4['sales'])
# plot ssm of mu (only trend)
tools.plot_ssm(mcmc_sample,
sales_df_4['date'],
'only trend',
'sales',
'mu',
sales_df_4['sales'])
# plot ssm of gamma (only cycle)
tools.plot_ssm(mcmc_sample,
sales_df_4['date'],
'only cycle',
mcmc_result = sm.sampling(data=stan_data,
seed=1,
chains=4,
iter=3000,
warmup=2000,
control=control,
thin=6)
print(mcmc_result)
mcmc_result.plot()
plt.show()
# saving compiled model
if not os.path.exists('%s.pkl' % filename):
with open('%s.pkl' % filename, 'wb') as f:
pickle.dump(sm, f)
mcmc_sample = mcmc_result.extract(permuted=True)
# plot ssm of probs
tools.plot_ssm(mcmc_sample, fish_df['date'], 'expectation of state', 'prob.',
'lambda_exp', fish_df['fish_num'])
tools.plot_ssm(mcmc_sample, fish_df['date'],
'expectation of state without random', 'prob.', 'lambda_smooth',
fish_df['fish_num'])
tools.plot_ssm(mcmc_sample, fish_df['date'],
'expectation of state without random and fixed temperature',
'prob.', 'lambda_smooth_fix', fish_df['fish_num'])
if os.path.exists('%s.pkl' % filename):
sm = pickle.load(open('%s.pkl' % filename, 'rb'))
# sm = pystan.StanModel(file='5-4-1-simple-reg.stan')
else:
# a model using prior for mu and sigma.
sm = pystan.StanModel(file='%s.stan' % filename)
control = {'adapt_delta': 0.99, 'max_treedepth': 16}
mcmc_result = sm.sampling(data=stan_data,
seed=1,
chains=4,
iter=8000,
warmup=1200,
control=control,
thin=6)
print(mcmc_result)
mcmc_result.plot()
plt.show()
# saving compiled model
if not os.path.exists('%s.pkl' % filename):
with open('%s.pkl' % filename, 'wb') as f:
pickle.dump(sm, f)
mcmc_sample = mcmc_result.extract(permuted=True)
# plot ssm of mu
tools.plot_ssm(mcmc_sample, sales_df_3['date'], 'smooth trend model', 'sales',
'mu', sales_df_3['sales'])