def prepare_training_set(params):
db_values_x_training, db_values_y_training, db_names_training, db_values_x_test, db_values_y_test, db_names_test = load_db(
params)
if params[
"normalise_data"] == 3: #We normalise the target and input values.
print("Normalasing the input and target values")
db_values_x_training, db_values_x_test, men, std = ut.normalise_data(
db_values_x_training, db_values_x_test)
params["x_men"] = men
params["x_std"] = std
db_values_y_training, db_values_y_test, men, std = ut.normalise_data(
db_values_y_training, db_values_y_test)
params["y_men"] = men
params["y_std"] = std
if params["subsample"] > 0:
print("Subsampling frames %s" % params["subsample"])
db_values_x_training, db_values_y_training, db_names_training, seq_rel_train = subsample_frames(
params, db_values_x_training, db_values_y_training,
db_names_training)
db_values_x_test, db_values_y_test, db_names_test, seq_rel_test = subsample_frames(
params, db_values_x_test, db_values_y_test, db_names_test)
params["seq_rel_train"] = seq_rel_train
params["seq_rel_test"] = seq_rel_test
X_train, Y_train, F_list_train, G_list_train, S_Train_list, R_L_Train_list = prepare_sequences(
params, db_values_x_training, db_values_y_training, db_names_training)
X_test, Y_test, F_list_test, G_list_test, S_Test_list, R_L_Test_list = prepare_sequences(
params, db_values_x_test, db_values_y_test, db_names_test)
return (params, X_train, Y_train, F_list_train, G_list_train, S_Train_list,
R_L_Train_list, X_test, Y_test, F_list_test, G_list_test,
S_Test_list, R_L_Test_list)
def load_flat_data(params):
db_values_x_training, db_values_y_training, db_names_training, db_values_x_test, db_values_y_test, db_names_test = load_db(
params)
if params[
"normalise_data"] == 3: #We normalise the target and input values.
print("Normalasing the input and target values")
db_values_x_training, db_values_x_test, men, std = ut.normalise_data(
db_values_x_training, db_values_x_test)
params["x_men"] = men
params["x_std"] = std
db_values_y_training, db_values_y_test, men, std = ut.normalise_data(
db_values_y_training, db_values_y_test)
params["y_men"] = men
params["y_std"] = std
if params[
"normalise_data"] == 4: # We normalise the target and input values.
print(
"Normalasing the input and target values with same std of entire training set"
)
db_values_x_training, db_values_y_training, db_values_x_test, db_values_y_test, men, std = \
ut.complete_normalise_data(db_values_x_training,db_values_y_training,db_values_x_test,db_values_y_test)
params["x_men"] = men
params["x_std"] = std
if params["subsample"] > 0:
print("Subsampling frames %s" % params["subsample"])
db_values_x_training, db_values_y_training, db_names_training, seq_rel_train = subsample_frames(
params, db_values_x_training, db_values_y_training,
db_names_training)
db_values_x_test, db_values_y_test, db_names_test, seq_rel_test = subsample_frames(
params, db_values_x_test, db_values_y_test, db_names_test)
params["seq_rel_train"] = seq_rel_train
params["seq_rel_test"] = seq_rel_test
return params, db_values_x_training, db_values_y_training, db_names_training, db_values_x_test, db_values_y_test, db_names_test
# label_path_1 = 'demo/Emotion/data/sensor_b8r3_c5_y_s40_e80.npy'
azi = np.load(azi_data_path)
ele = np.load(ele_data_path)
label = np.zeros((len(emotion_list), 80))
for i in range(len(label)):
label[i] = i
label = label.flatten()
# expand dims
azi = np.expand_dims(azi, axis=1)
ele = np.expand_dims(ele, axis=1)
# normalize
azi = normalise_data(azi)
ele = normalise_data(ele)
# split data
azi_train, azi_test, ele_train, ele_test, label_train, label_test = train_test_split(
azi, ele, label, test_size=0.2, random_state=25, stratify=label)
train_loader = senor_heatmap_label_data_loader(azi_train,
ele_train,
label_train,
batch_size=BATCH_SIZE)
test_loader = senor_heatmap_label_data_loader(
azi_test, ele_test, label_test, batch_size=np.shape(azi_test)[0])
# log path
# path = dir_path("sensor_heatmap_3dcnn", result_dir)
'flatten_factor': 30,
}
# results dir
result_dir = "FER/results"
# load data
# df_x = np.load('C:/Users/Zber/Documents/Dev_program/OpenRadar/demo/Emotion/data/emotion_3s_diff_segment_x.npy')
# df_y = np.load('C:/Users/Zber/Documents/Dev_program/OpenRadar/demo/Emotion/data/emotion_3s_diff_segment_y.npy')
df_x = np.load('//data/emotion_3s_diff_x.npy')
df_y = np.load('//data/emotion_3s_diff_y.npy')
# normalization
df_x = normalise_data(df_x)
df_x = np.expand_dims(df_x, axis=2)
# df_y_eye = np_to_eye(df_y, num_class=7)
# split data
x_train, x_test, y_train, y_test = train_test_split(df_x, df_y, test_size=0.2, random_state=25, stratify=df_y)
train_loader = data_loader(x_train, y_train, batch_size=BATCH_SIZE)
test_loader = data_loader(x_test, y_test, batch_size=BATCH_SIZE)
# log path
path = dir_path("emotion_3s_diff_segment", result_dir)
# create model
model = LeNet(**model_para)
model = model.to(device)