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 calibrate_mh():
num_mh = 20000
results = demo8(num_mh)
(individual, actual, beliefs, meta, states,
indiv_states, params, accept) = results
mh = diagnostics.summarise_qs(states, indiv_states)
#params['transition'] = "separate"
#res_sep = do_inference(params, beliefs, meta)
#states_sep, indiv_sep, _, _ = res_sep
#mh_sep = diagnostics.summarise_qs(states_sep, indiv_sep)
return actual, mh
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
ulinop = np.array([np.mean(b[1,:]) for b in beliefs])
ulinop_ans = np.where(ulinop > .5, 1, 0)
ulinop_correct = ulinop_ans == actual_ans
consensus = np.array([np.mean([ulinop[n] if nominal[n][k] == 1
else 1 - ulinop[n]
for n in range(num_qs)])
for k in range(num_ppl)])
votes = np.array([nom for nom in nominal])
1/0
return mh
if __name__ == '__main__':
warnings.simplefilter('error', RuntimeWarning)
np.seterr(divide='ignore') #so that / 0.0 gives inf on ubuntu
np.seterr(invalid='ignore')
#call some demo function
#diagnostics.pymc_diagnostics(states[0], actual[0])
mh = diagnostics.summarise_qs(states, indiv_states)
nominal = np.array([(np.array(beliefs[k][1,:]) > .5).astype(int)
for k in range(len(beliefs))])
accuracy = np.array([np.sum([nominal[n, k] == actual[n]['world']
for n in range(params['num_qs'])])
for k in range(params['num_responses'])])
