pid = os.getpid()
print('PID: {}'.format(pid))
f = open('pid_' + str(train_split), 'wb')
f.write(str(pid) + '\n')
f.close()
# Load model
model = SupervisedModel('experiment', './', learning_rate=1e-2)
monitor = util.Monitor(model,
checkpoint_directory='checkpoints_' + str(train_split),
save_steps=1000)
# Loading CK+ dataset
print('Loading Data')
supervised_data_loader = SupervisedDataLoaderCrossVal(
data_paths.ck_plus_data_path)
train_data_container = supervised_data_loader.load('train', train_split)
test_data_container = supervised_data_loader.load('test', train_split)
X_train = train_data_container.X
X_train = numpy.float32(X_train)
X_train /= 255.0
X_train *= 2.0
y_train = train_data_container.y
X_test = test_data_container.X
X_test = numpy.float32(X_test)
X_test /= 255.0
X_test *= 2.0
y_test = test_data_container.y
pid = os.getpid()
print('PID: {}'.format(pid))
f = open('pid_'+str(train_split), 'wb')
f.write(str(pid)+'\n')
f.close()
# Load model
model = SupervisedModel('experiment', './', learning_rate=1e-2)
monitor = util.Monitor(model,
checkpoint_directory='checkpoints_'+str(train_split),
save_steps=1000)
# Loading CK+ dataset
print('Loading Data')
supervised_data_loader = SupervisedDataLoaderCrossVal(
data_paths.ck_plus_data_path)
train_data_container = supervised_data_loader.load('train', train_split)
test_data_container = supervised_data_loader.load('test', train_split)
X_train = train_data_container.X
X_train = numpy.float32(X_train)
X_train /= 255.0
X_train *= 2.0
y_train = train_data_container.y
X_test = test_data_container.X
X_test = numpy.float32(X_test)
X_test /= 255.0
X_test *= 2.0
y_test = test_data_container.y
f.write(str(pid) + '\n')
f.close()
# Load model
model = SupervisedModel('experiment', './', learning_rate=1e-2)
monitor = util.Monitor(model,
checkpoint_directory='checkpoints_' + str(train_split),
save_steps=1000)
# Add dropout to fully-connected layer
model.fc4.dropout = 0.5
model._compile()
# Loading CK+ dataset
print('Loading Data')
supervised_data_loader = SupervisedDataLoaderCrossVal(
'/data/Expr_Recog/CK+_condensed/npy_files/')
train_data_container = supervised_data_loader.load('train', train_split)
test_data_container = supervised_data_loader.load('test', train_split)
train_mask = numpy.logical_and(train_data_container.y != 0,
train_data_container.y != 2)
train_data_container.X = train_data_container.X[train_mask, :, :, :]
train_data_container.y = train_data_container.y[train_mask]
train_data_container.y = reindex_labels(train_data_container.y)
test_mask = numpy.logical_and(test_data_container.y != 0,
test_data_container.y != 2)
test_data_container.X = test_data_container.X[test_mask, :, :, :]
test_data_container.y = test_data_container.y[test_mask]
test_data_container.y = reindex_labels(test_data_container.y)
if train_split == 9:
parser.add_argument("-s", "--split", default="0", help="Training split of CK+ to use. (0-9)")
parser.add_argument("checkpoint_file", help="Path to a single model checkpoint (.pkl file).")
args = parser.parse_args()
checkpoint_file = args.checkpoint_file
fold = int(args.split)
dataset_path = data_paths.ck_plus_data_path
print "Checkpoint: %s" % checkpoint_file
print "Testing on split %d\n" % fold
# Load model
model = SupervisedModel("evaluation", "./")
# Load dataset
supervised_data_loader = SupervisedDataLoaderCrossVal(dataset_path)
test_data_container = supervised_data_loader.load(mode="test", fold=fold)
test_data_container.X = numpy.float32(test_data_container.X)
test_data_container.X /= 255.0
test_data_container.X *= 2.0
# Remove samples with neutral and contempt labels
mask = numpy.logical_and(test_data_container.y != 0, test_data_container.y != 2)
test_data_container.X = test_data_container.X[mask, :, :, :]
test_data_container.y = test_data_container.y[mask]
test_data_container.y = reindex_labels(test_data_container.y)
num_test_samples = len(test_data_container.y)
if fold == 9:
test_data_container.X, test_data_container.y = add_padding(test_data_container.X, test_data_container.y)
parser.add_argument("checkpoint_file",
help='Path to a single model checkpoint (.pkl file).')
args = parser.parse_args()
checkpoint_file = args.checkpoint_file
fold = int(args.split)
dataset_path = data_paths.ck_plus_data_path
print 'Checkpoint: %s' % checkpoint_file
print 'Testing on split %d\n' % fold
# Load model
model = SupervisedModel('evaluation', './')
# Load dataset
supervised_data_loader = SupervisedDataLoaderCrossVal(dataset_path)
test_data_container = supervised_data_loader.load(mode='test', fold=fold)
test_data_container.X = numpy.float32(test_data_container.X)
test_data_container.X /= 255.0
test_data_container.X *= 2.0
# Construct evaluator
preprocessor = [util.Normer3(filter_size=5, num_channels=1)]
evaluator = util.Evaluator(model, test_data_container, checkpoint_file,
preprocessor)
# For the inputted checkpoint, compute the overall test accuracy
accuracies = []
print 'Checkpoint: %s' % os.path.split(checkpoint_file)[1]
evaluator.set_checkpoint(checkpoint_file)
