def test_inception_v3_imagenet(device_id):
if cntk_device(device_id).type() != DeviceKind_GPU:
pytest.skip('test only runs on GPU')
try_set_default_device(cntk_device(device_id))
current_path = os.getcwd()
base_path = prepare_ImageNet_data()
# change dir to locate data.zip correctly
os.chdir(base_path)
from _cntk_py import set_fixed_random_seed, force_deterministic_algorithms
set_fixed_random_seed(1)
force_deterministic_algorithms()
train_data = os.path.join(base_path, 'train_map.txt')
test_data = os.path.join(base_path, 'val_map.txt')
try:
error = inception_v3_train_and_eval(train_data, test_data, minibatch_size=8, epoch_size=200,
max_epochs=4, restore=False, testing_parameters=(200, 8))
finally:
os.chdir(current_path)
expected_error = 0.99
assert np.allclose(error, expected_error, atol=TOLERANCE_ABSOLUTE)
def test_inception_v3_imagenet(device_id):
if cntk_device(device_id).type() != DeviceKind_GPU:
pytest.skip('test only runs on GPU')
try_set_default_device(cntk_device(device_id))
current_path = os.getcwd()
base_path = prepare_ImageNet_data()
# change dir to locate data.zip correctly
os.chdir(base_path)
from _cntk_py import set_fixed_random_seed, force_deterministic_algorithms
set_fixed_random_seed(1)
force_deterministic_algorithms()
train_data = os.path.join(base_path, 'train_map.txt')
test_data = os.path.join(base_path, 'val_map.txt')
try:
error = inception_v3_train_and_eval(train_data,
test_data,
minibatch_size=8,
epoch_size=200,
max_epochs=4,
restore=False,
testing_parameters=(200, 8))
finally:
os.chdir(current_path)
expected_error = 0.99
assert np.allclose(error, expected_error, atol=TOLERANCE_ABSOLUTE)
def test_alexnet_error(device_id):
if cntk_device(device_id).type() != DeviceKind_GPU:
pytest.skip('test only runs on GPU')
set_default_device(cntk_device(device_id))
base_path = prepare_ImageNet_data()
# change dir to locate data.zip correctly
os.chdir(base_path)
from _cntk_py import set_computation_network_trace_level, set_fixed_random_seed, force_deterministic_algorithms
set_computation_network_trace_level(1)
set_fixed_random_seed(
1
) # BUGBUG: has no effect at present # TODO: remove debugging facilities once this all works
#force_deterministic_algorithms()
# TODO: do the above; they lead to slightly different results, so not doing it for now
# for test purpose we train and test on same data
train_data = os.path.join(base_path, 'val1024_map.txt')
test_data = os.path.join(base_path, 'val1024_map.txt')
test_error = alexnet_train_and_eval(train_data,
test_data,
num_quantization_bits=32,
minibatch_size=16,
epoch_size=64,
max_epochs=2)
def test_alexnet_imagenet_distributed_block_momentum(device_id):
params = [
"-n", "2", "-m", "8", "-e", "16", "-datadir",
prepare_ImageNet_data(), "-b", "1600", "-r", "-device",
str(device_id)
]
mpiexec_test(device_id, script_under_test, mpiexec_params, params, 0.99,
False)
def test_alexnet_imagenet_distributed_1bitsgd(device_id):
params = [
"-n", "2", "-datadir",
prepare_ImageNet_data(), "-q", "1", "-m", "8", "-e", "16", "-r",
"-device",
str(device_id)
]
mpiexec_test(device_id, script_under_test, mpiexec_params, params, 0.99,
True)
def test_alexnet_imagenet_distributed(device_id):
params = [ "-n", "2",
"-m", "8",
"-e", "16",
"-datadir", prepare_ImageNet_data(),
"-q", "32",
"-r",
"-device", "0" ]
mpiexec_test(device_id, script_under_test, params, 0.99, True)
def disabled_test_alexnet_imagenet_distributed_block_momentum(device_id):
params = [ "-n", "2",
"-m", "8",
"-e", "16",
"-datadir", prepare_ImageNet_data(),
"-b", "1600",
"-r",
"-device", str(device_id) ]
mpiexec_test(device_id, script_under_test, mpiexec_params, params, 0.99, False)
def test_bn_inception_imagenet_distributed(device_id):
params = [
"-n", "4", "-datadir",
prepare_ImageNet_data(), "-q", "32", "-e", "300", "-m", "2", "-r",
"-device",
str(device_id)
]
mpiexec_test(device_id, script_under_test, mpiexec_params, params, 0.99,
True)
def test_alexnet_imagenet_distributed(device_id):
params = [ "-n", "2",
"-m", "4",
"-e", "16",
"-datadir", prepare_ImageNet_data(),
"-q", "32",
"-r",
"-device", "0" ]
mpiexec_test(device_id, script_under_test, params, 0.99, True)
def test_bn_inception_imagenet_distributed(device_id):
params = [ "-n", "4",
"-datadir", prepare_ImageNet_data(),
"-q", "32",
"-e", "300",
"-m", "2",
"-r",
"-device", str(device_id) ]
mpiexec_test(device_id, script_under_test, mpiexec_params, params, 0.99, True, timeout_seconds=400)
def test_VGG16_imagenet_distributed(device_id):
params = [ "-n", "2",
"-m", "2",
"-e", "2",
"-datadir", prepare_ImageNet_data(),
"-q", "32",
"-device", str(device_id),
"-r",
"-testing"]
# Currently we only test for CPU since the memory usage is very high for GPU (~6 GB)
mpiexec_test(device_id, script_under_test, mpiexec_params, params, 0.99, True, timeout_seconds=500, use_only_cpu=True)
def test_inception_v3_imagenet_distributed(device_id):
# Inception-V3 distributed test on ImageNet need plenty of memory,
# for now, the test server might feel hard to handle
pytest.skip('Mute Inception-V3 distributed test temporarily')
params = ["-n", "2",
"-datadir", prepare_ImageNet_data(),
"-q", "32",
"-e", "200",
"-m", "2",
"-r",
"-device", str(device_id)]
mpiexec_test(device_id, script_under_test, mpiexec_params, params, 0.99, True, timeout_seconds=400)
def test_VGG19_imagenet_distributed(device_id):
params = [
"-n", "2", "-m", "2", "-e", "2", "-datadir",
prepare_ImageNet_data(), "-q", "32", "-device",
str(device_id), "-r", "-testing"
]
# Currently we only test for CPU since the memory usage is very high for GPU (~6 GB)
mpiexec_test(device_id,
script_under_test,
mpiexec_params,
params,
0.99,
True,
timeout_seconds=500,
use_only_cpu=True)
def test_inception_v3_imagenet_distributed(device_id):
# Inception-V3 distributed test on ImageNet need plenty of memory,
# for now, the test server might feel hard to handle
pytest.skip('Mute Inception-V3 distributed test temporarily')
params = [
"-n", "2", "-datadir",
prepare_ImageNet_data(), "-q", "32", "-e", "200", "-m", "2", "-r",
"-device",
str(device_id)
]
mpiexec_test(device_id,
script_under_test,
mpiexec_params,
params,
0.99,
True,
timeout_seconds=400)
