def test_running_stats(self):
inspector = dp_inspector.DPModelInspector(should_throw=False)
self.assertTrue(inspector.validate(nn.InstanceNorm1d(16)))
self.assertTrue(inspector.validate(nn.InstanceNorm1d(16, affine=True)))
self.assertTrue(
inspector.validate(nn.InstanceNorm1d(16,
track_running_stats=True)))
self.assertFalse(
inspector.validate(
nn.InstanceNorm1d(16, affine=True, track_running_stats=True)))
def test_unsupported_layer(self):
class SampleNetWithTransformer(nn.Module):
def __init__(self):
super().__init__()
self.fc = nn.Linear(8, 16)
self.encoder = nn.Transformer()
def forward(self, x):
x = self.fc(x)
x = self.encoder(x)
return x
model = SampleNetWithTransformer()
inspector = dp_inspector.DPModelInspector(should_throw=False)
self.assertFalse(inspector.validate(model))
def test_extra_param(self):
inspector = dp_inspector.DPModelInspector(should_throw=False)
class SampleNetWithExtraParam(nn.Module):
def __init__(self):
super().__init__()
self.fc = nn.Linear(8, 16)
self.extra_param = nn.Parameter(torch.Tensor(16, 2))
def forward(self, x):
x = self.fc(x)
x = x.matmul(self.extra_param)
return x
model = SampleNetWithExtraParam()
self.assertFalse(inspector.validate(model))
model.extra_param.requires_grad = False
self.assertTrue(inspector.validate(model))
def test_conv2d(self):
inspector = dp_inspector.DPModelInspector(should_throw=False)
self.assertTrue(
inspector.validate(
nn.Conv2d(in_channels=3,
out_channels=6,
kernel_size=1,
groups=1)))
self.assertTrue(
inspector.validate(
nn.Conv2d(in_channels=3,
out_channels=6,
kernel_size=1,
groups=3)))
self.assertFalse(
inspector.validate(
nn.Conv2d(in_channels=6,
out_channels=6,
kernel_size=1,
groups=2)))
def test_returns_true(self):
inspector = dp_inspector.DPModelInspector()
model = utils.convert_batchnorm_modules(models.resnet50())
self.assertTrue(inspector.validate(model))
def test_returns_False(self):
inspector = dp_inspector.DPModelInspector()
model = models.resnet50()
inspector.should_throw = False
self.assertFalse(inspector.validate(model))
def test_raises_exception(self):
inspector = dp_inspector.DPModelInspector()
model = models.resnet50()
with self.assertRaises(dp_inspector.IncompatibleModuleException):
inspector.validate(model)
