diff --git a/denoising_diffusion_pytorch/denoising_diffusion_pytorch.py b/denoising_diffusion_pytorch/denoising_diffusion_pytorch.py
index 4d68919..d211f39 100644
--- a/denoising_diffusion_pytorch/denoising_diffusion_pytorch.py
+++ b/denoising_diffusion_pytorch/denoising_diffusion_pytorch.py
@@ -444,9 +444,10 @@ class Trainer(object):
 
     def load(self, milestone):
         data = torch.load(f'./model-{milestone}.pt')
+
         self.step = data['step']
-        self.model = data['model']
-        self.ema_model = data['ema']
+        self.model.load_state_dict(data['model'])
+        self.ema_model.load_state_dict(data['ema'])
 
     def train(self):
         while self.step < self.train_num_steps:
diff --git a/setup.py b/setup.py
index d59e43d..db80780 100644
--- a/setup.py
+++ b/setup.py
@@ -3,7 +3,7 @@ from setuptools import setup, find_packages
 setup(
   name = 'denoising-diffusion-pytorch',
   packages = find_packages(),
-  version = '0.2.1',
+  version = '0.2.2',
   license='MIT',
   description = 'Denoising Diffusion Probabilistic Models - Pytorch',
   author = 'Phil Wang',