Migrations
/
pytorch-image-models
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

only run test_forward_features

Browse Source
pull/554/head
Aman Arora 4 years ago
parent a17ce022e6
commit 84fd045e4d
1 changed files with 106 additions and 106 deletions
Show Stats Download Patch File Download Diff File

212
tests/test_models.py
Unescape View File

@ -31,98 +31,98 @@ MAX_BWD_SIZE = 128
MAX_FWD_FEAT_SIZE = 448 MAX_FWD_FEAT_SIZE = 448
@pytest.mark.timeout(120) # @pytest.mark.timeout(120)
@pytest.mark.parametrize('model_name', list_models(exclude_filters=EXCLUDE_FILTERS[:-NUM_NON_STD])) # @pytest.mark.parametrize('model_name', list_models(exclude_filters=EXCLUDE_FILTERS[:-NUM_NON_STD]))
@pytest.mark.parametrize('batch_size', [1]) # @pytest.mark.parametrize('batch_size', [1])
def test_model_forward(model_name, batch_size): # def test_model_forward(model_name, batch_size):
"""Run a single forward pass with each model""" # """Run a single forward pass with each model"""
model = create_model(model_name, pretrained=False) # model = create_model(model_name, pretrained=False)
model.eval() # model.eval()
input_size = model.default_cfg['input_size'] # input_size = model.default_cfg['input_size']
if any([x > MAX_FWD_SIZE for x in input_size]): # if any([x > MAX_FWD_SIZE for x in input_size]):
# cap forward test at max res 448 * 448 to keep resource down # # cap forward test at max res 448 * 448 to keep resource down
input_size = tuple([min(x, MAX_FWD_SIZE) for x in input_size]) # input_size = tuple([min(x, MAX_FWD_SIZE) for x in input_size])
inputs = torch.randn((batch_size, *input_size)) # inputs = torch.randn((batch_size, *input_size))
outputs = model(inputs) # outputs = model(inputs)
assert outputs.shape[0] == batch_size # assert outputs.shape[0] == batch_size
assert not torch.isnan(outputs).any(), 'Output included NaNs' # assert not torch.isnan(outputs).any(), 'Output included NaNs'
@pytest.mark.timeout(120) # @pytest.mark.timeout(120)
@pytest.mark.parametrize('model_name', list_models(exclude_filters=EXCLUDE_FILTERS)) # @pytest.mark.parametrize('model_name', list_models(exclude_filters=EXCLUDE_FILTERS))
@pytest.mark.parametrize('batch_size', [2]) # @pytest.mark.parametrize('batch_size', [2])
def test_model_backward(model_name, batch_size): # def test_model_backward(model_name, batch_size):
"""Run a single forward pass with each model""" # """Run a single forward pass with each model"""
model = create_model(model_name, pretrained=False, num_classes=42) # model = create_model(model_name, pretrained=False, num_classes=42)
num_params = sum([x.numel() for x in model.parameters()]) # num_params = sum([x.numel() for x in model.parameters()])
model.eval() # model.eval()
input_size = model.default_cfg['input_size'] # input_size = model.default_cfg['input_size']
if any([x > MAX_BWD_SIZE for x in input_size]): # if any([x > MAX_BWD_SIZE for x in input_size]):
# cap backward test at 128 * 128 to keep resource usage down # # cap backward test at 128 * 128 to keep resource usage down
input_size = tuple([min(x, MAX_BWD_SIZE) for x in input_size]) # input_size = tuple([min(x, MAX_BWD_SIZE) for x in input_size])
inputs = torch.randn((batch_size, *input_size)) # inputs = torch.randn((batch_size, *input_size))
outputs = model(inputs) # outputs = model(inputs)
outputs.mean().backward() # outputs.mean().backward()
for n, x in model.named_parameters(): # for n, x in model.named_parameters():
assert x.grad is not None, f'No gradient for {n}' # assert x.grad is not None, f'No gradient for {n}'
num_grad = sum([x.grad.numel() for x in model.parameters() if x.grad is not None]) # num_grad = sum([x.grad.numel() for x in model.parameters() if x.grad is not None])
assert outputs.shape[-1] == 42 # assert outputs.shape[-1] == 42
assert num_params == num_grad, 'Some parameters are missing gradients' # assert num_params == num_grad, 'Some parameters are missing gradients'
assert not torch.isnan(outputs).any(), 'Output included NaNs' # assert not torch.isnan(outputs).any(), 'Output included NaNs'
@pytest.mark.timeout(120) # @pytest.mark.timeout(120)
@pytest.mark.parametrize('model_name', list_models(exclude_filters=NON_STD_FILTERS)) # @pytest.mark.parametrize('model_name', list_models(exclude_filters=NON_STD_FILTERS))
@pytest.mark.parametrize('batch_size', [1]) # @pytest.mark.parametrize('batch_size', [1])
def test_model_default_cfgs(model_name, batch_size): # def test_model_default_cfgs(model_name, batch_size):
"""Run a single forward pass with each model""" # """Run a single forward pass with each model"""
model = create_model(model_name, pretrained=False) # model = create_model(model_name, pretrained=False)
model.eval() # model.eval()
state_dict = model.state_dict() # state_dict = model.state_dict()
cfg = model.default_cfg # cfg = model.default_cfg
classifier = cfg['classifier'] # classifier = cfg['classifier']
pool_size = cfg['pool_size'] # pool_size = cfg['pool_size']
input_size = model.default_cfg['input_size'] # input_size = model.default_cfg['input_size']
if all([x <= MAX_FWD_FEAT_SIZE for x in input_size]) and \ # if all([x <= MAX_FWD_FEAT_SIZE for x in input_size]) and \
not any([fnmatch.fnmatch(model_name, x) for x in EXCLUDE_FILTERS]): # not any([fnmatch.fnmatch(model_name, x) for x in EXCLUDE_FILTERS]):
# output sizes only checked if default res <= 448 * 448 to keep resource down # # output sizes only checked if default res <= 448 * 448 to keep resource down
input_size = tuple([min(x, MAX_FWD_FEAT_SIZE) for x in input_size]) # input_size = tuple([min(x, MAX_FWD_FEAT_SIZE) for x in input_size])
input_tensor = torch.randn((batch_size, *input_size)) # input_tensor = torch.randn((batch_size, *input_size))
# test forward_features (always unpooled) # # test forward_features (always unpooled)
outputs = model.forward_features(input_tensor) # outputs = model.forward_features(input_tensor)
assert outputs.shape[-1] == pool_size[-1] and outputs.shape[-2] == pool_size[-2] # assert outputs.shape[-1] == pool_size[-1] and outputs.shape[-2] == pool_size[-2]
# test forward after deleting the classifier, output should be poooled, size(-1) == model.num_features # # test forward after deleting the classifier, output should be poooled, size(-1) == model.num_features
model.reset_classifier(0) # model.reset_classifier(0)
outputs = model.forward(input_tensor) # outputs = model.forward(input_tensor)
assert len(outputs.shape) == 2 # assert len(outputs.shape) == 2
assert outputs.shape[-1] == model.num_features # assert outputs.shape[-1] == model.num_features
# test model forward without pooling and classifier # # test model forward without pooling and classifier
model.reset_classifier(0, '') # reset classifier and set global pooling to pass-through # model.reset_classifier(0, '') # reset classifier and set global pooling to pass-through
outputs = model.forward(input_tensor) # outputs = model.forward(input_tensor)
assert len(outputs.shape) == 4 # assert len(outputs.shape) == 4
if not isinstance(model, timm.models.MobileNetV3): # if not isinstance(model, timm.models.MobileNetV3):
# FIXME mobilenetv3 forward_features vs removed pooling differ # # FIXME mobilenetv3 forward_features vs removed pooling differ
assert outputs.shape[-1] == pool_size[-1] and outputs.shape[-2] == pool_size[-2] # assert outputs.shape[-1] == pool_size[-1] and outputs.shape[-2] == pool_size[-2]
# check classifier name matches default_cfg # # check classifier name matches default_cfg
assert classifier + ".weight" in state_dict.keys(), f'{classifier} not in model params' # assert classifier + ".weight" in state_dict.keys(), f'{classifier} not in model params'
# check first conv(s) names match default_cfg # # check first conv(s) names match default_cfg
first_conv = cfg['first_conv'] # first_conv = cfg['first_conv']
if isinstance(first_conv, str): # if isinstance(first_conv, str):
first_conv = (first_conv,) # first_conv = (first_conv,)
assert isinstance(first_conv, (tuple, list)) # assert isinstance(first_conv, (tuple, list))
for fc in first_conv: # for fc in first_conv:
assert fc + ".weight" in state_dict.keys(), f'{fc} not in model params' # assert fc + ".weight" in state_dict.keys(), f'{fc} not in model params'
if 'GITHUB_ACTIONS' not in os.environ: if 'GITHUB_ACTIONS' not in os.environ:
@ -147,20 +147,20 @@ EXCLUDE_JIT_FILTERS = [
] ]
@pytest.mark.timeout(120) # @pytest.mark.timeout(120)
@pytest.mark.parametrize('model_name', list_models(exclude_filters=EXCLUDE_FILTERS + EXCLUDE_JIT_FILTERS)) # @pytest.mark.parametrize('model_name', list_models(exclude_filters=EXCLUDE_FILTERS + EXCLUDE_JIT_FILTERS))
@pytest.mark.parametrize('batch_size', [1]) # @pytest.mark.parametrize('batch_size', [1])
def test_model_forward_torchscript(model_name, batch_size): # def test_model_forward_torchscript(model_name, batch_size):
"""Run a single forward pass with each model""" # """Run a single forward pass with each model"""
with set_scriptable(True): # with set_scriptable(True):
model = create_model(model_name, pretrained=False) # model = create_model(model_name, pretrained=False)
model.eval() # model.eval()
input_size = (3, 128, 128) # jit compile is already a bit slow and we've tested normal res already... # input_size = (3, 128, 128) # jit compile is already a bit slow and we've tested normal res already...
model = torch.jit.script(model) # model = torch.jit.script(model)
outputs = model(torch.randn((batch_size, *input_size))) # outputs = model(torch.randn((batch_size, *input_size)))
assert outputs.shape[0] == batch_size # assert outputs.shape[0] == batch_size
assert not torch.isnan(outputs).any(), 'Output included NaNs' # assert not torch.isnan(outputs).any(), 'Output included NaNs'
EXCLUDE_FEAT_FILTERS = [ EXCLUDE_FEAT_FILTERS = [

Write Preview
Loading…
Cancel
Save