def __init__(self, per_data, per_lens=None, n_states=None):
if per_lens is None:
per_lens = list(map(len, per_data))
if len(per_data.shape) > 2:
per_data = per_data.reshape(-1, per_data.shape[-1])
if n_states is None:
components = [2, 4, 6, 8, 10]
hmms = [GaussianHMM(n_components=c) for c in components]
map(lambda g: g.fit(per_data, per_lens), hmms)
scores = map(lambda g: aic(g, per_data, per_lens), hmms)
max_score, self.hmm = sorted(zip(scores, hmms))[0]
else:
self.hmm = GaussianHMM(n_components=n_states)
self.hmm.fit(per_data, per_lens)
ll = self.hmm.score(per_data, per_lens)
print "Goal HMM n_components", self.hmm.n_components, "Log likelihood", ll
upper_idxs = [per_lens[0] - 1]
start_idxs = [0]
for i in range(1, len(per_lens)):
upper_idxs.append(upper_idxs[i - 1] + per_lens[i])
start_idxs.append(start_idxs[i - 1] + per_lens[i - 1])
self.final_states = np.array(self.hmm.predict(per_data,
per_lens))[upper_idxs]
print self.final_states
self.T = int(np.mean(per_lens))
self.n_components = self.hmm.n_components
def aic(em_fit_result_dict):
'''
Compute Akaike Information Criterion.
'''
# Number of parameters:
# - mixt_target: 1
# - mixt_random: 1
# - mixt_nontarget: K
K = em_fit_result_dict['mixt_nontargets'].size + 2
return utils.aic(K, em_fit_result_dict['train_LL'])
def aic(em_fit_result_dict):
'''
Compute Akaike Information Criterion.
'''
# Number of parameters:
# - mixt_target: 1
# - mixt_random: 1
# - mixt_nontarget: K
# - kappa: 1
K = 3 + em_fit_result_dict['K']
return utils.aic(K, em_fit_result_dict['train_LL'])
def rf_func(k, combo):
regr = RandomForestClassifier(random_state=0, n_estimators=10) ##########
data = matrices[k, :, 1:][:, combo]
response = binary_labels[k]
reg_f = regr.fit(data, response)
#features = reg_f.feature_importances_
loglik = utils.loglik_logistic(reg_f, data, response)
aic = utils.aic(loglik, data.shape[0], data.shape[1])
locations = np.where(combo)[0]
model_coefs = np.array(combo).astype(float)
np.put(model_coefs, locations, reg_f.feature_importances_)
return [aic, model_coefs]
def aic(em_fit_result_dict):
'''
Compute Akaike Information Criterion.
'''
# Number of parameters:
# - mixt_target: Tnum
# - mixt_random: Tnum
# - mixt_nontargets: Tnum - 1
# - alpha: 1
# - beta: 1
K = em_fit_result_dict['mixt_target'].size + em_fit_result_dict['mixt_random'].size + em_fit_result_dict['mixt_nontargets'].size + 1
return utils.aic(K, em_fit_result_dict['train_LL'])
def aic(em_fit_result_dict):
'''
Compute Akaike Information Criterion.
'''
# Number of parameters:
# - mixt_target_tr: Tnum
# - mixt_random_tr: Tnum
# - mixt_nontarget: sum(T_space - 1)
# - alpha: 1
# - beta: 1
K = em_fit_result_dict['mixt_target_tr'].size + em_fit_result_dict['mixt_random_tr'].size + em_fit_result_dict['mixt_nontargets_tr'].size + 2
return utils.aic(K, em_fit_result_dict['train_LL'])
