def find_weights(datasets): results=validation_votes(datasets) loss_fun=LossFunction(ens.Votes(results)) bound_w = [(0.0, 1.0) for _ in datasets] result = differential_evolution(loss_fun, bound_w, maxiter=1000, tol=1e-7) weights=result['x'] return weights
def diff_voting(common,deep,clf="LR"): datasets=ens.read_dataset(common,deep) weights=find_weights(datasets) results=learn.train_ens(datasets,clf="LR") votes=ens.Votes(results) result=votes.weighted(weights) return result
def cv_exp(common, binary, out_path=None, clf="LR"):
datasets = ens.read_dataset(common, binary)
validation = [CrossVal(0.1 * (i + 1)) for i in range(2, 10)]
result_dict = {}
for valid_i in validation:
votes_i = ens.Votes(valid_i(datasets, clf)[1])
result_dict[str(valid_i)] = votes_i.voting(False)
exp.result_exp("no weights", result_dict, out_path)
def rename_frames(paths, json_path):
datasets = ens.read_dataset(paths["common"], paths["binary"])
helper = get_renam_fun(json_path)
if (get_fun):
return helper
new_datasets = datasets #[helper(data_i) for data_i in datasets]
votes = ens.Votes(learn.train_ens(new_datasets, clf="LR"))
result_i = votes.voting(False)
result_i.report()
def __call__(self, datasets, clf="LR"):
acc, pairs = [], []
for i in range(self.k):
self.base_val.reset()
pair_i = self.base_val(datasets, clf)
result_i = ens.Votes(pair_i[1]).voting(False)
acc.append(result_i.get_acc())
pairs.append(pair_i)
mediana = np.argsort(acc)[len(acc) // 2]
return pairs[mediana]
def full_train(common, binary, clf):
datasets = ens.read_dataset(common, binary)
results = []
for data_i in datasets:
train, test = data_i.split()
model = learn.make_model(train, clf)
X_train, y_true = train.get_X(), train.get_labels()
y_pred = model.predict_proba(X_train)
result_i = learn.Result(y_true, y_pred, test.names())
results.append(result_i)
return ens.Votes(results)
def ensemble(paths, system=None, clf="LR", s_clf=None, transform=None):
datasets = ens.read_dataset(paths["common"], paths["binary"])
if (transform):
datasets = [transform(data_i) for data_i in datasets]
results = [
learn.train_model(data_i, clf_type=clf, binary=False)
for data_i in datasets
]
if (s_clf):
results = [results[clf_i] for clf_i in s_clf]
votes = ens.Votes(results)
return voting(votes, system)
def helper(common_path, binary_path, clf="LR"):
datasets = ens.read_dataset(common_path, binary_path)
results = [person_acc(data_i) for data_i in datasets]
return ens.Votes(results)
def single_optim(self,datasets,results,clf):
weights=self.find_weights(results)
results=learn.train_ens(datasets,clf)
votes=ens.Votes(results)
result=votes.weighted(weights)
return result,weights
def __init__(self,all_votes):
self.all_votes=ens.Votes(all_votes)
self.iter=0
def test_weights(datasets, clf, weights):
votes = ens.Votes(learn.train_ens(datasets, clf))
return votes.weighted(weights)
def get_votes(train, clf, names, selector_gen):
all_votes = []
for selector_i in selector_gen(names):
results = learn.train_ens(train, clf=clf, selector=selector_i)
all_votes.append(ens.Votes(results))
return all_votes
def __init__(self, all_votes):
self.all_votes = ens.Votes(all_votes)
self.d = [result_i.true_one_hot() for result_i in all_votes]
def helper(size): ind_i= random.sample(indexes,size) subset_i=[votes.results[k] for k in ind_i] votes_i=ens.Votes(subset_i) result_i=votes_i.voting(False) return result_i.get_acc(),ind_i
def base_train(common, binary, clf):
datasets = ens.read_dataset(common, binary)
results = learn.train_ens(datasets, clf=clf)
return ens.Votes(results)
def score_dataset(results, score_weights):
pref = [as_pref(result_i, score_weights) for result_i in results]
y_true, names = results[0].y_true, results[0].names
results = [learn.Result(y_true, pref_i, names) for pref_i in pref]
return ens.Votes(results)