def geom_wrap(expt, qnum, sn, subject_start=0, subject_end=33):
beliefs, metas, actual, _, qnames = vars_from_data(dataset=expt)
beliefs = beliefs[qnum]
metas = metas[qnum]
#beliefs = beliefs[:, subject_start:subject_end]
#metas = metas[:, subject_start:subject_end]
num_qs = 1
num_ppl = beliefs.shape[1]
params = setup_params.init_params_demo1()
params['num_qs'] = num_qs
params['num_responses'] = num_ppl
params['binary_beliefs'] = False
params['meta_noise_hierarchy'] = "default"
params['meta_noise_default'] = sn
params['own_noise_hierarchy'] = "default"
params['own_noise_default'] = sn
params['use_expertise'] = False
params['expertise_hierarchy'] = "individual"
params['prior_wp_sm'] = "signal_assumption"
params['do_parallel'] = False
wp_grid = 10 #33,
sm_grid = 50 #200
results = geom_grid(beliefs, metas, params, wp_grid, sm_grid)
probs = [geom_probs_ll(res, wp_grid, sm_grid) for res in results]
for pp in probs:
print(np.nansum(pp[0,:,:,:]))
reorder = [geom_reorder(pp, wp_grid, sm_grid) for pp in probs]
1/0
def demo8(num_mh=None):
params = setup_params.init_params_demo1()
params['prior_wp_sm'] = "gen_bij_infer_all"
params['binary_beliefs'] = False
#params['meta_noise_hierarchy'] = "individual"
#params['own_noise_hierarchy'] = "individual"
params['meta_noise_hierarchy'] = "default"
params['own_noise_hierarchy'] = "default"
params['meta_noise_type'] = "truncnorm"
params['own_noise_type'] = "truncnorm"
params['meta_noise_default'] = 1
params['own_noise_default'] = 1
params['num_responses'] = 15
params['use_expertise'] = False
params['expertise_hierarchy'] = "individual"
params['num_qs'] = 1 #48
if num_mh is None:
params['num_mh_samples'] = 5000 #5000
else:
params['num_mh_samples'] = num_mh
params['thin'] = 1
params['num_mh_expertise'] = 0 #50
params['num_mh_outer'] = 1 #30
params['use_meta'] = True
params['do_parallel'] = False
params['fxd_actual'] = False
params['all_same'] = False
params['transition'] = "separate"
return generic_demo(params)
def lesions_expt(qnum):
beliefs, meta, act, nom = vars_from_data(dataset="lesions")
params = setup_params.init_params_demo1()
params['num_qs'] = 1
params['num_responses'] = beliefs[qnum].shape[1]
params['num_mh_samples'] = 200
states, indiv_states, accept, internals = do_inference(params, beliefs[qnum:qnum+1], meta[qnum:qnum+1])
mh = diagnostics.summarise_qs(states, indiv_states)
return mh, beliefs, meta, act
def simple_mh_all(expt, num_mh, qnums=None):
beliefs, meta, act, nom, qnames = vars_from_data(dataset=expt)
if qnums is not None:
beliefs = [beliefs[qn] for qn in qnums]
meta = [meta[qn] for qn in qnums]
act = [act[qn] for qn in qnums]
nom = [nom[qn] for qn in qnums]
params = setup_params.init_params_demo1()
params['num_qs'] = len(beliefs)
params['num_mh_samples'] = num_mh
params['do_parallel'] = False
states, indiv_states, accept, internals = do_inference(params, beliefs, meta)
mh = diagnostics.summarise_qs(states, indiv_states, params)
save_mh_results(qnames, mh, act, params)
1/0
return mh, beliefs, meta, act, qnames
def wp_expt(sm=np.array([[.7, .4], [.3, .6]]), wp=np.array([.1, .9]),
change_wp=True):
params = setup_params.init_params_demo1()
params['binary_beliefs'] = False
indiv = fwd.generate_individual(params)
grid = 100
diff = 0
res0 = np.zeros((grid-grid*diff, grid-grid*diff))
res1 = np.zeros((grid-grid*diff, grid-grid*diff))
bm0 = np.zeros((grid-grid*diff, grid-grid*diff))
bm1 = np.zeros((grid-grid*diff, grid-grid*diff))
mm0 = np.zeros((grid-grid*diff, grid-grid*diff))
mm1 = np.zeros((grid-grid*diff, grid-grid*diff))
mm0avg = np.zeros((grid-grid*diff, grid-grid*diff))
#hacky since grid needn't be nice num.:
for k, gen in enumerate(np.arange(diff * grid, grid)):
actual = {}
gen_s0 = gen / grid + .01
if change_wp:
gen_wp = np.array([gen_s0, 1 - gen_s0])
actual['wp'] = gen_wp
actual['sm'] = sm
else:
actual['wp'] = wp
gen_sm = np.array([[gen_s0, gen_s0 - diff],
[1 - gen_s0, 1 - gen_s0 + diff]])
actual['sm'] = gen_sm
actual['world'] = 0
actual['signals'] = fwd.sample_signals(actual['world'], actual['sm'],
params['num_responses'])
beliefs, meta = fwd.generate_data(actual, indiv, params)
for j, ans in enumerate(np.arange(grid*diff, grid)):
ans_s0 = ans / grid + .01
if change_wp:
ans_wp = np.array([ans_s0, 1 - ans_s0])
ans_sm = actual['sm']
else:
ans_wp = actual['wp']
ans_sm = np.array([[ans_s0, ans_s0 - diff],
[1 - ans_s0, 1 - ans_s0 + diff]])
ll = np.array([aggregation.reports_ll_marg(beliefs, meta, w,
ans_sm, ans_wp,
indiv['own_noise'],
indiv['meta_noise'],
params)
for w in [0, 1]])
ll += aggregation.log_prob_world(actual['world'], ans_wp)
bel = fwd.calc_belief_matrix(ans_sm, ans_wp)
bm0[k, j] = bel[0,0]
bm1[k, j] = bel[0,1]
met = fwd.calc_meta_matrix(ans_sm, ans_wp)
mm0[k, j] = met[0,0]
mm1[k, j] = met[0,1]
if bel[0, 0] < .5:
mm0avg[k,j] = 0
elif bel[0, 1] > .5:
mm0avg[k,j] = 1
else:
mm0avg[k,j] = met[0,0]
if bel[0, 1] > .5:
non_bij = 1
elif bel[0,0] < .5:
non_bij = -1
else:
non_bij = 0
res0[k, j] = ll[0]
res1[k, j] = ll[1]
plt.imshow(res0, cmap=cm.Greys_r)
plt.imshow(res1, cmap=cm.Greys_r)
1/0
return res
