def surv_data_2_ref(surv_data):
# instantiate an empty reference
ref = Reference()
# add meta data to reference
# at present just a timestamp
from time import gmtime, strftime
ref.set_meta({'timestamp':strftime("%Y-%m-%d %H:%M:%S", gmtime())})
for channel_code in objectives_channel_codes:
# add reference objective corresponding to each model objective
if channel_code == 'prevalence':
#debug_p('surv_data ' + str(surv_data['prevalence']))
d_points = prevalence_surv_data_2_d_points(surv_data['prevalence'])
else:
msg = "Channel " + channel_code + " not implemeneted yet!\nSetting reference data to None."
warn_p(msg)
d_points = None
#debug_p('adding objective ' + channel_code)
#debug_p('num d_points ' + str(len(d_points)))
ref.add_objective(channel_code, d_points)
return ref
def get_ref(self, cluster_id):
surv_data = {}
all_ref_objs_found = True
for channel_code in objectives_channel_codes:
if channel_code == 'prevalence':
prev_data = c2p(cluster_id)
if prev_data:
surv_data[channel_code] = prev_data
else:
msg = 'Prevalence objective reference data was not found!\n Skipping plotting cluster ' + cluster_id + ' fit!'
print msg
all_ref_objs_found = False
else:
msg = "Channel objective" + channel_code + " not implemented yet!\nSetting objective reference data to None for plotting."
warn_p(msg)
surv_data[channel_code] = None
# one of the reference objective channels was not found; skipping cluster fit!
if not all_ref_objs_found:
ref = None
else:
ref = d2f(surv_data)
return ref
def sim_report_channels_model_format(reports_channels, sim_data):
report_channels_data = {}
for report_channel in reports_channels:
if report_channel == 'reinfections':
report_channels_data['reinfections'] = get_sim_report_reinfections(sim_data)
else:
msg = "Channel " + report_channel + " not implemeneted yet!\nSetting report channels data to None."
warn_p(msg)
report_channels_data[report_channel] = None
return report_channels_data
def fit(self):
best_fit = {}
if self.type == 'mmse_distance':
best_fit = self.best_fit_mmse_distance()
elif self.type == 'mmse_distance_cached':
best_fit = self.best_fit_mmse_distance(cached = True)
else:
msg = "unrecognized fit function type " + self.type + "!\nReturning None."
warn_p(msg)
return None
return best_fit
def prevalence_surv_data_2_d_points(prev_surv_data):
if prev_surv_data:
prev_ts = len(prev_surv_data)*[0]
for rnd,prev in prev_surv_data.iteritems():
if not prev == -1000:
prev_ts[int(rnd)] = prev;
else:
# if no data in reference impute to 'nan'
prev_ts[int(rnd) ] = 'nan'
return prev_ts
else:
msg = 'no reference data for prevalence available!\n Returning none.'
warn_p(msg)
return None
def generate_gazetteer_map(self):
gazetteer_base_map_file_path = os.path.join(ref_data_dir, gazetteer_base_map_file_name)
with open(gazetteer_base_map_file_path ,'r') as map_f:
self.base_map = json.load(map_f)
self.sweep_map = []
for cluster_id, cluster_record in self.best_fits.iteritems():
cluster_map_record = self.get_cluster_map_record(cluster_id)
if cluster_map_record:
#cluster_map_record['sim_avg_reinfection_rate'] = cluster_record['sim_avg_reinfection_rate']
#cluster_map_record['ref_avg_reinfection_rate'] = cluster_record['ref_avg_reinfection_rate']
cluster_map_record['temp_h'] = cluster_record['habs']['temp_h']
cluster_map_record['const_h'] = cluster_record['habs']['const_h']
mn_sim_map_file_path = os.path.join(kariba_viz_dir, gazetteer_sim_mn_base_map_file_name)
if os.path.exists(mn_sim_map_file_path):
with open(mn_sim_map_file_path ,'r') as mn_map_f:
mn_map = json.load(mn_map_f)
cluster_map_record['RDT_mn_sim'] = mn_map[cluster_id]['RDT_sim']
cluster_map_record['itn_level'] = cluster_record['ITN_cov']
cluster_map_record['drug_coverage'] = cluster_record['MSAT_cov']
cluster_map_record['fit_value'] = cluster_record['fit_value']
cluster_map_record['RDT_sn_sim'] = cluster_record['prevalence']
self.sweep_map.append(cluster_map_record)
else:
warn_p('No cluster map record found in base map. Skipping generating sim entries in cluster map record.')
gazetteer_sweep_map_file_path = os.path.join(kariba_viz_dir, self.sweep_name + '_' + gazetteer_base_map_file_name)
with open(gazetteer_sweep_map_file_path ,'w') as map_f:
json.dump(self.sweep_map, map_f, indent = 4)
def mse(self, m_points, d_points, points_weights):
if d_points:
num_obs = len(d_points)
if not len(m_points) == num_obs:
msg = "number of points in model does not match num of points in reference data!\nReturning None."
warn_p(msg)
return None
if num_obs == 0:
msg = "no observations provided!\nReturning None."
warn_p(msg)
return None
mse = 0.0
non_nan_obs = 0
for idx, m_p in enumerate(m_points):
d_p = d_points[idx]
if not (d_p == 'nan' or m_p == 'nan'):
mse = mse + math.pow(m_p - d_p, 2)*points_weights[idx]
non_nan_obs = non_nan_obs + 1
if non_nan_obs > 0:
mse = mse / non_nan_obs
else:
msg = "only nan observations provided!\nReturning None"
warn_p(msg)
return None
return mse
msg = "no reference data provided for fit!\nReturning None"
warn_p(msg)
return None
def fit(self):
models_list_prime = calib_data_2_models_list(self.calib_data)
best_fits = {}
all_fits = {}
#all_fits = {'fit':{'min_residual':float('inf')}, }
all_fits['min_residual'] = float('inf')
all_fits['max_residual'] = 0.0
all_fits['models'] = {}
debug_p('category ' + self.category)
for idx,cluster_id in enumerate(c2c(self.category)):
models_list = copy.deepcopy(models_list_prime)
print "Processing cluster " + cluster_id + "."
debug_p('Processing cluster ' + cluster_id + " in " + self.category + ".")
itn_traj = cluster_2_itn_traj(cluster_id)
drug_cov = cluster_2_drug_cov(cluster_id)
# prune models to the ones matching prior data
cluster_models = []
for model in models_list:
model_meta = model.get_meta()
if model_meta['group_key'] == get_sim_group_key(itn_traj, drug_cov):
#debug_p('model id before kariba conversion ' + str(model.get_model_id()))
group_key = model_meta['group_key']
sim_key = model_meta['sim_key']
model = KaribaModel(model, self.calib_data[group_key][sim_key], cluster_id, all_fits = self.fit_terms)
#model = kariba_model
#debug_p('model id after kariba conversion ' + str(model.get_model_id()))
cluster_models.append(model)
surv_data = {}
all_ref_objs_found = True
for channel_code in objectives_channel_codes:
if channel_code == 'prevalence':
prev_data = c2p(cluster_id)
if prev_data:
surv_data[channel_code] = prev_data
else:
msg = 'Prevalence objective reference data was not found!\n Skipping cluster ' + cluster_id + ' fit!'
print msg
all_ref_objs_found = False
else:
msg = "Channel objective" + channel_code + " not implemented yet!\nSetting objective reference data to None."
warn_p(msg)
surv_data[channel_code] = None
# one of the reference objective channels was not found; skipping cluster fit!
if not all_ref_objs_found:
continue
ref = d2f(surv_data)
# adjust highest possible fit to account for RDT+ model in dtk not reflecting reality at the upper end
obj_prev = ref.get_obj_by_name('prevalence')
d_points = obj_prev.get_points()
obj_prev.set_points([min(point, rdt_max) for point in d_points])
fitting_set = FittingSet(cluster_id, cluster_models, ref)
if load_prevalence_mse:
fit = Fit(fitting_set, type = 'mmse_distance_cached')
else:
fit = Fit(fitting_set)
best_fit_model = fit.best_fit_mmse_distance()
min_residual = fit.get_min_residual()
max_residual = fit.get_max_residual()
if min_residual < all_fits['min_residual']:
all_fits['min_residual'] = min_residual
if max_residual > all_fits['max_residual']:
all_fits['max_residual'] = max_residual
if best_fit_model:
temp_h, const_h, itn_level, drug_coverage_level = get_model_params(best_fit_model)
best_fit_meta = best_fit_model.get_meta()
best_fits[cluster_id] = {}
best_fits[cluster_id]['habs'] = {}
best_fits[cluster_id]['habs']['const_h'] = const_h
best_fits[cluster_id]['habs']['temp_h'] = temp_h
best_fits[cluster_id]['ITN_cov'] = itn_level
best_fits[cluster_id]['category'] = self.category
best_fits[cluster_id]['MSAT_cov'] = drug_coverage_level
best_fits[cluster_id]['sim_id'] = best_fit_meta['sim_id']
best_fits[cluster_id]['sim_key'] = best_fit_meta['sim_key']
best_fits[cluster_id]['group_key'] = best_fit_meta['group_key']
best_fits[cluster_id]['fit_value'] = best_fit_model.get_fit_val()
best_fits[cluster_id]['sim_avg_reinfection_rate'] = best_fit_model.get_sim_avg_reinfection_rate()
best_fits[cluster_id]['ref_avg_reinfection_rate'] = best_fit_model.get_ref_avg_reinfection_rate()
best_fits[cluster_id]['prevalence'] = best_fit_model.get_objective_by_name('prevalence').get_points()
# redundancy; to be refactored via FitEntry class
best_fits[cluster_id]['fit'] = {}
best_fits[cluster_id]['fit']['value'] = best_fit_model.get_fit_val()
best_fits[cluster_id]['fit']['temp_h'] = temp_h
best_fits[cluster_id]['fit']['const_h'] = const_h
best_fits[cluster_id]['fit']['ITN_cov'] = itn_level
best_fits[cluster_id]['fit']['MSAT_cov'] = drug_coverage_level
best_fits[cluster_id]['fit']['sim_id'] = best_fit_meta['sim_id']
best_fits[cluster_id]['fit']['sim_key'] = best_fit_meta['sim_key']
best_fits[cluster_id]['mse'] = {}
best_fits[cluster_id]['mse']['value'] = fit.get_min_mses()['prevalence']['value'] # get mmse for objective prevalence
best_fit_mse_model = fit.get_min_mses()['prevalence']['model']
temp_h, const_h, itn_level, drug_coverage_level = get_model_params(best_fit_mse_model)
model_meta_data = best_fit_mse_model.get_meta()
best_fits[cluster_id]['mse']['temp_h'] = temp_h
best_fits[cluster_id]['mse']['const_h'] = const_h
best_fits[cluster_id]['mse']['ITN_cov'] = itn_level
best_fits[cluster_id]['mse']['MSAT_cov'] = drug_coverage_level
best_fits[cluster_id]['mse']['sim_id'] = model_meta_data['sim_id']
best_fits[cluster_id]['mse']['sim_key'] = model_meta_data['sim_key']
best_fits[cluster_id]['cc_penalty'] = {}
best_fits[cluster_id]['cc_penalty']['value'] = fit.get_min_penalties()['prevalence']['value'] # get clinical penalty for objective prevalence; at present this is just the clinical cases penalty; if reinfection is considered the code needs to be adjusted
best_fit_cc_penalty_model = fit.get_min_penalties()['prevalence']['model']
temp_h, const_h, itn_level, drug_coverage_level = get_model_params(best_fit_cc_penalty_model)
model_meta_data = best_fit_cc_penalty_model.get_meta()
best_fits[cluster_id]['cc_penalty']['temp_h'] = temp_h
best_fits[cluster_id]['cc_penalty']['const_h'] = const_h
best_fits[cluster_id]['cc_penalty']['ITN_cov'] = itn_level
best_fits[cluster_id]['cc_penalty']['MSAT_cov'] = drug_coverage_level
best_fits[cluster_id]['cc_penalty']['sim_id'] = model_meta_data['sim_id']
best_fits[cluster_id]['cc_penalty']['sim_key'] = model_meta_data['sim_key']
rho = best_fit_model.get_rho()
p_val = best_fit_model.get_p_val()
if rho and p_val :
best_fits[cluster_id]['rho'] = rho
best_fits[cluster_id]['p_val'] = p_val
debug_p('rho' + str(rho))
debug_p('p_val' + str(p_val))
else:
msg = "something went wrong and the best fit for " + cluster_id + " could not be found."
warn_p(msg)
all_fits['models'][cluster_id] = cluster_models
#all_fits['models'][cluster_id] = fit.get_fitting_set_models()
print str(idx+1) + " clusters have been processed."
debug_p( str(idx+1) + " clusters have been processed in category " + self.category)
'''
if idx > 0:
break
'''
return best_fits, all_fits
