training_dict = classify_library.toDict(training) testing_dict = classify_library.toDict(testing) #################################################################### #################################################################### ################################## Script starts X_train_vids = classify_library.limited_input1(training_dict, 1000) X_test_vids = classify_library.limited_input1(testing_dict, 1000) #GET THE TRAINING AND TESTING DATA. X_train, Y_train = classify_library.make_FV_matrix(X_train_vids,training_output, class_index) X_test, Y_test = classify_library.make_FV_matrix(X_test_vids,testing_output, class_index) #PCA reduction training_PCA = classify_library.limited_input1(training_dict,40) X_PCA, _ = classify_library.make_FV_matrix(training_PCA,training_output, class_index) n_components = 1000 pca = PCA(n_components=n_components) pca.fit(X_PCA) X_train_PCA = pca.transform(X_train) X_test_PCA = pca.transform(X_test) #Exhaustive Grid Search C = [1, 10, 50, 100, 1000]
if filename.endswith('.fisher.npz')
]
training_dict = classify_library.toDict(training)
testing_dict = classify_library.toDict(testing)
####################################################################
####################################################################
################################## Script starts
X_train_vids = classify_library.limited_input1(training_dict, 1000)
X_test_vids = classify_library.limited_input1(testing_dict, 1000)
#GET THE TRAINING AND TESTING DATA.
X_train, Y_train = classify_library.make_FV_matrix(X_train_vids,
training_output,
class_index)
X_test, Y_test = classify_library.make_FV_matrix(X_test_vids,
testing_output,
class_index)
X_total = np.concatenate((X_train, X_test), 0)
Y_total = np.concatenate((Y_train, Y_test), 0)
if not args.no_pca:
#PCA reduction
training_PCA = classify_library.limited_input1(training_dict, 40)
X_PCA, _ = classify_library.make_FV_matrix(training_PCA,
training_output,
class_index)
print(train_vid_class.keys()[:5])
print('len testing:', len(testing))
training_n_dict = classify_library.toDict(training_n, train_vid_class)
training_s_dict = classify_library.toDict(training_s, train_vid_class)
testing_dict = classify_library.toDict(testing, test_vid_class)
# input('...')
#GET THE TRAINING AND TESTING DATA.
X_train_n_vids = classify_library.limited_input1(training_n_dict, args.per_class_num)
X_train_s_vids = classify_library.limited_input1(training_s_dict, args.per_class_num)
X_test_vids = classify_library.limited_input1(testing_dict, args.per_class_num)
# X_train_vids, X_test_vids = classify_library.limited_input(training_dict, testing_dict, 101, 24)
X_n_train, Y_n_train = classify_library.make_FV_matrix(X_train_n_vids,
training_n_output, class_index, train_vid_class)
X_s_train, Y_s_train = classify_library.make_FV_matrix(X_train_s_vids,
training_s_output, class_index, train_vid_class)
X_test, Y_test = classify_library.make_FV_matrix(X_test_vids,
testing_output, class_index, test_vid_class)
# pdb.set_trace()
training_n_PCA = classify_library.limited_input1(training_n_dict,1)
training_s_PCA = classify_library.limited_input1(training_s_dict,1)
if not args.PCA_dim:
X_n_train_PCA = X_n_train.tolist()
X_s_train_PCA = X_s_train.tolist()
X_n_test_PCA = X_test.tolist()
