def test_setup(): """ Load a model from example dir and set it up with new interface""" import os test_dir = os.path.dirname(os.path.abspath(__file__)) dm = dismod_mr.load(test_dir + '/example_data') dm.setup_model(rate_type='p', rate_model='neg_binom')
models = {} # iter=101; burn=0; thin=1 # use these settings to run faster iter = 10000 burn = 5000 thin = 5 # use these settings to make sure MCMC converges # <markdowncell> # # Consistent fit with all data # # Let's start with a consistent fit of the simulated PD data. This includes data on prevalence, incidence, and SMR, and the assumption that remission rate is zero. All together this counts as four different data types in the DisMod-II accounting. # <codecell> model = dismod_mr.load('/homes/abie/notebook/pd_sim_data/') model.keep(areas=['GBR'], sexes=['female', 'total']) # <codecell> model.setup_model() %time model.fit(iter=iter, burn=burn, thin=thin) # <codecell> models['p, i, r, smr'] = model model.plot() # <markdowncell> # # Consistent fit without incidence
# <codecell> models = {} #iter=101; burn=0; thin=1 # use these settings to run faster iter=10000; burn=5000; thin=5 # use these settings to make sure MCMC converges # <markdowncell> # # Consistent fit with all data # # Let's start with a consistent fit of the simulated PD data. This includes data on prevalence, incidence, and SMR, and the assumption that remission rate is zero. All together this counts as four different data types in the DisMod-II accounting. # <codecell> model = dismod_mr.load('/homes/abie/notebook/pd_sim_data/') model.keep(areas=['GBR'], sexes=['female', 'total']) # <codecell> model.setup_model() %time model.fit(iter=iter, burn=burn, thin=thin) # <codecell> models['p, i, r, smr'] = model model.plot() # <markdowncell> # # Consistent fit without incidence
def test_setup(): """ Load a model from example dir and set it up with new interface""" dm = dismod_mr.load(dismod_mr.testing.get_test_data_dir()) dm.setup_model(rate_type='p', rate_model='neg_binom')