feature_data = read.read()
test_ids = list(feature_data.keys())
all_labels = list(feature_data[test_ids[0]].keys())
for test_id in test_ids:
for a_label in all_labels:
train_labels = [a for a in all_labels if a != a_label]
_train_data, _test_data = read.split(feature_data, test_id)
_train_data = read.remove_class(_train_data, [a_label])
train_data = create_train_instances(_train_data)
_support_data, _test_data = read.support_set_split(
_test_data, samples_per_class)
_support_data, _support_labels = read.flatten(_support_data)
_support_data = np.array(_support_data)
_support_data = np.expand_dims(_support_data, 3)
numsupportset = samples_per_class * classes_per_set
input1 = Input((numsupportset + 1, feature_length, 1))
modelinputs = []
base_network = conv_embedding()
for lidx in range(numsupportset):
modelinputs.append(
base_network(Lambda(lambda x: x[:, lidx, :, :])(input1)))
targetembedding = base_network(
Lambda(lambda x: x[:, -1, :, :])(input1))
modelinputs.append(targetembedding)
supportlabels = Input((numsupportset, classes_per_set))
modelinputs.append(supportlabels)
x = MaxPooling1D(pool_size=2)(x)
x = BatchNormalization()(x)
x = Flatten()(x)
x = Dense(1200, activation='relu')(x)
x = BatchNormalization()(x)
return Model(inputs=_input, outputs=x, name='embedding')
feature_data = read.read()
test_ids = list(feature_data.keys())
for test_id in test_ids:
_train_data, _test_data = read.split(feature_data, test_id)
train_data = create_train_instances(_train_data)
_train_data, _train_labels = read.flatten(_train_data)
_test_data, _test_labels = read.flatten(_test_data)
print(len(_train_data))
'''_train_data = np.array(_train_data)
_test_data = np.array(_test_data)
_train_data = np.expand_dims(_train_data, 3)
_test_data = np.expand_dims(_test_data, 3)
numsupportset = samples_per_class * classes_per_set
input1 = Input((numsupportset + 1, feature_length, 1))
modelinputs = []
base_network = conv_embedding()
for lidx in range(numsupportset):
modelinputs.append(base_network(Lambda(lambda x: x[:, lidx, :, :])(input1)))
targetembedding = base_network(Lambda(lambda x: x[:, -1, :, :])(input1))
feature_data = read.read()
test_ids = list(feature_data.keys())
all_labels = list(feature_data[test_ids[0]].keys())
for test_id in test_ids:
for a_label in all_labels:
train_labels = [a for a in all_labels if a != a_label]
_train_data, _test_data = read.split(feature_data, test_id)
_train_data = read.remove_class(_train_data, [a_label])
_support_data, _test_data = read.support_set_split(_test_data, candidates)
_train_data, _train_labels = read.flatten(_train_data)
_support_data, _support_labels = read.flatten(_support_data)
_train_data = np.array(_train_data)
_train_data = np.expand_dims(_train_data, 3)
_support_data = np.array(_support_data)
_support_data = np.expand_dims(_support_data, 3)
_embedding_model, _triplet_model = build_conv_model((feature_length,1))
_triplet_model.fit_generator(triplet_generator_minibatch(_train_data, _train_labels, mini_batch_size
, train_labels),
steps_per_epoch=steps_per_epoch, epochs=epochs, verbose=1)
_support_preds = _embedding_model.predict(_support_data)
_input = Input(shape=(180, ))
encoded = Dense(96, activation='sigmoid')(_input)
decoded = Dense(180)(encoded)
return Model(inputs=_input, outputs=decoded)
feature_data = read.read()
test_ids = list(feature_data.keys())
hc_results = []
ca_results = []
ha_results = []
for test_id in test_ids:
_train_data, _test_data = read.split(feature_data, [test_id])
h_train_data, c_train_data, a_train_data, _train_labels = read.flatten(
_train_data)
h_test_data, c_test_data, a_test_data, _test_labels = read.flatten(
_test_data)
h_train_data = np.array(h_train_data)
c_train_data = np.array(c_train_data)
a_train_data = np.array(a_train_data)
hc_train_data = np.concatenate([h_train_data, c_train_data], axis=1)
print(hc_train_data.shape)
ca_train_data = np.concatenate([c_train_data, a_train_data], axis=1)
print(ca_train_data.shape)
ha_train_data = np.concatenate([h_train_data, a_train_data], axis=1)
print(ha_train_data.shape)
h_test_data = np.array(h_test_data)
c_test_data = np.array(c_test_data)
x = Conv1D(12, kernel_size=5, activation='relu')(_input)
x = MaxPooling1D(pool_size=2)(x)
x = BatchNormalization()(x)
x = Flatten()(x)
x = Dense(1200, activation='relu')(x)
x = BatchNormalization()(x)
return Model(inputs=_input, outputs=x, name='embedding')
all_features = read.read()
test_ids = list(all_features.keys())
for test_id in test_ids:
_train_features, _test_features = read.split(all_features, test_id)
_train_features, _train_labels = read.flatten(_train_features)
_test_features, _test_labels = read.flatten(_test_features)
_train_labels_ = np_utils.to_categorical(_train_labels,
len(read.activity_list))
_test_labels_ = np_utils.to_categorical(_test_labels,
len(read.activity_list))
_train_features = np.array(_train_features)
_train_features = np.reshape(
_train_features, (_train_features.shape[0],
_train_features.shape[1] * _train_features.shape[2]))
_train_features = np.expand_dims(_train_features, 3)
print(_train_features.shape)
_test_features = np.array(_test_features)
all_features = read.read()
test_ids = list(all_features.keys())
all_labels = list(all_features[test_ids[0]].keys())
for test_id in test_ids:
for a_label in all_labels:
train_labels = [a for a in all_labels if a != a_label]
_train_features, _test_features = read.split(all_features, test_id)
_train_features = read.remove_class(_train_features, [a_label])
_support_features, _test_features = read.support_set_split(
_test_features, k_shot)
_train_features, _train_labels = read.flatten(_train_features)
_support_features, _support_labels = read.flatten(_support_features)
id_list = range(len(train_labels))
activity_id_dict = dict(zip(train_labels, id_list))
_train_labels_ = []
for item in _train_labels:
_train_labels_.append(activity_id_dict.get(item))
_train_labels_ = np_utils.to_categorical(_train_labels_,
len(train_labels))
_train_features = np.array(_train_features)
_train_features = np.reshape(
_train_features,