# The SVM baseline for SEMAINE import shared_defs_DISFA import data_preparation import numpy import mlp pca_loc = "../pca_generation/generic_face_rigid" (all_aus, train_recs, disfa_dir, hog_data_dir) = shared_defs_DISFA.shared_defs() devel_recs = train_recs[14:-1] train_recs = train_recs[0:14] [train_samples, train_labels, valid_samples, valid_labels, _, PC, means, scaling] = \ data_preparation.Prepare_HOG_AU_data_generic_DISFA_dynamic(train_recs, devel_recs, all_aus, disfa_dir, hog_data_dir, pca_loc, geometry=True) # binarise the labels train_labels[train_labels > 1] = 1 valid_labels[valid_labels > 1] = 1 print train_samples.shape, valid_samples.shape, numpy.mean(train_labels, axis=0), numpy.mean( valid_labels, axis=0) import validation_helpers train_fn = mlp.train_mlp_probe test_fn = mlp.test_mlp_class
# The SVM baseline for BO4D import shared_defs_DISFA import data_preparation (all_aus, users, DISFA_dir, hog_data_dir) = shared_defs_DISFA.shared_defs() train_recs = users[0:int(len(users)/2)] devel_recs = users[(int(len(users)/2)):] pca_loc = "../pca_generation/generic_face_rigid" f = open("./trained/DISFA_train_dynamic_lin_svm.txt", 'w') for au in all_aus: hyperparams = {"C": [0.000001, 0.00001, 0.0001, 0.001, 0.01, 0.1, 1, 10, 100, 1000], "validate_params": ["C"]} # load the training and testing data for the current fold [train_samples, train_labels, valid_samples, valid_labels, raw_valid, PC, means, scaling] = \ data_preparation.Prepare_HOG_AU_data_generic_DISFA_dynamic(train_recs, devel_recs, [au], DISFA_dir, hog_data_dir, pca_loc, geometry=True) train_labels[train_labels > 1] = 1 valid_labels[valid_labels > 1] = 1 import linear_SVM import validation_helpers train_fn = linear_SVM.train_SVM test_fn = linear_SVM.test_SVM # Cross-validate here
# The SVM baseline for SEMAINE import shared_defs_DISFA import data_preparation import numpy import mlp pca_loc = "../pca_generation/generic_face_rigid" (all_aus, train_recs, disfa_dir, hog_data_dir) = shared_defs_DISFA.shared_defs() devel_recs = train_recs[14:-1] train_recs = train_recs[0:14] [train_samples, train_labels, valid_samples, valid_labels, _, PC, means, scaling] = \ data_preparation.Prepare_HOG_AU_data_generic_DISFA(train_recs, devel_recs, all_aus, disfa_dir, hog_data_dir, pca_loc, geometry=True) # binarise the labels train_labels[train_labels > 1] = 1 valid_labels[valid_labels > 1] = 1 print train_samples.shape, valid_samples.shape, numpy.mean(train_labels, axis=0), numpy.mean(valid_labels, axis=0) import validation_helpers train_fn = mlp.train_mlp_probe test_fn = mlp.test_mlp_class hyperparams = { 'batch_size': [100],
# The SVM baseline for BO4D import shared_defs_DISFA import data_preparation (all_aus, users, DISFA_dir, hog_data_dir) = shared_defs_DISFA.shared_defs() train_recs = users[0:int(len(users) / 2)] devel_recs = users[(int(len(users) / 2)):] pca_loc = "../pca_generation/generic_face_rigid" f = open("./trained/DISFA_train_dynamic_lin_svm.txt", 'w') for au in all_aus: hyperparams = { "C": [0.000001, 0.00001, 0.0001, 0.001, 0.01, 0.1, 1, 10, 100, 1000], "validate_params": ["C"] } # load the training and testing data for the current fold [train_samples, train_labels, valid_samples, valid_labels, raw_valid, PC, means, scaling] = \ data_preparation.Prepare_HOG_AU_data_generic_DISFA_dynamic(train_recs, devel_recs, [au], DISFA_dir, hog_data_dir, pca_loc, geometry=True) train_labels[train_labels > 1] = 1 valid_labels[valid_labels > 1] = 1 import linear_SVM import validation_helpers train_fn = linear_SVM.train_SVM
import shared_defs_DISFA import data_preparation import numpy import mlp (all_aus_bp4d, train_recs, devel_recs, BP4D_dir, hog_data_dir) = shared_defs_BP4D.shared_defs_intensity() pca_loc = "../pca_generation/generic_face_rigid" # load the training and testing data for the current fold [train_samples_bp4d, train_labels_bp4d, valid_samples_bp4d, valid_labels_bp4d, raw_valid, PC, means, scaling] = \ data_preparation.Prepare_HOG_AU_data_generic_BP4D_intensity(train_recs, devel_recs, all_aus_bp4d, BP4D_dir, hog_data_dir, pca_loc, geometry=True) (all_aus_disfa, train_recs, disfa_dir, hog_data_dir) = shared_defs_DISFA.shared_defs() devel_recs = train_recs[0:1] [train_samples_disfa, train_labels_disfa, _, _, _, PC, means, scaling] = \ data_preparation.Prepare_HOG_AU_data_generic_DISFA(train_recs, devel_recs, all_aus_disfa, disfa_dir, hog_data_dir, pca_loc, geometry=True) import validation_helpers train_labels_bp4d = train_labels_bp4d / 5.0 valid_labels_bp4d = valid_labels_bp4d / 5.0 train_labels_disfa = train_labels_disfa / 5.0 # Train on all three # Do the fully joint models first (2, 12, 17) aus_exp = [6, 12, 17]