diff --git a/src/pytorch/data/MNIST/raw/t10k-images-idx3-ubyte b/src/pytorch/data/MNIST/raw/t10k-images-idx3-ubyte
new file mode 100644
index 0000000..1170b2c
Binary files /dev/null and b/src/pytorch/data/MNIST/raw/t10k-images-idx3-ubyte differ
diff --git a/src/pytorch/data/MNIST/raw/t10k-images-idx3-ubyte.gz b/src/pytorch/data/MNIST/raw/t10k-images-idx3-ubyte.gz
new file mode 100644
index 0000000..5ace8ea
Binary files /dev/null and b/src/pytorch/data/MNIST/raw/t10k-images-idx3-ubyte.gz differ
diff --git a/src/pytorch/data/MNIST/raw/t10k-labels-idx1-ubyte b/src/pytorch/data/MNIST/raw/t10k-labels-idx1-ubyte
new file mode 100644
index 0000000..d1c3a97
Binary files /dev/null and b/src/pytorch/data/MNIST/raw/t10k-labels-idx1-ubyte differ
diff --git a/src/pytorch/data/MNIST/raw/t10k-labels-idx1-ubyte.gz b/src/pytorch/data/MNIST/raw/t10k-labels-idx1-ubyte.gz
new file mode 100644
index 0000000..a7e1415
Binary files /dev/null and b/src/pytorch/data/MNIST/raw/t10k-labels-idx1-ubyte.gz differ
diff --git a/src/pytorch/data/MNIST/raw/train-images-idx3-ubyte b/src/pytorch/data/MNIST/raw/train-images-idx3-ubyte
new file mode 100644
index 0000000..bbce276
Binary files /dev/null and b/src/pytorch/data/MNIST/raw/train-images-idx3-ubyte differ
diff --git a/src/pytorch/data/MNIST/raw/train-images-idx3-ubyte.gz b/src/pytorch/data/MNIST/raw/train-images-idx3-ubyte.gz
new file mode 100644
index 0000000..b50e4b6
Binary files /dev/null and b/src/pytorch/data/MNIST/raw/train-images-idx3-ubyte.gz differ
diff --git a/src/pytorch/data/MNIST/raw/train-labels-idx1-ubyte b/src/pytorch/data/MNIST/raw/train-labels-idx1-ubyte
new file mode 100644
index 0000000..d6b4c5d
Binary files /dev/null and b/src/pytorch/data/MNIST/raw/train-labels-idx1-ubyte differ
diff --git a/src/pytorch/data/MNIST/raw/train-labels-idx1-ubyte.gz b/src/pytorch/data/MNIST/raw/train-labels-idx1-ubyte.gz
new file mode 100644
index 0000000..707a576
Binary files /dev/null and b/src/pytorch/data/MNIST/raw/train-labels-idx1-ubyte.gz differ
diff --git a/src/pytorch/functions.py b/src/pytorch/functions.py
index e005810..a89c989 100644
--- a/src/pytorch/functions.py
+++ b/src/pytorch/functions.py
@@ -18,7 +18,7 @@ def __init__(self, k):
         '''
         super(linearUnified, self).__init__()
         self.k = k
-
+    @staticmethod
     def forward(self, x, w, b):
         '''
         forward propagation
@@ -36,7 +36,7 @@ def forward(self, x, w, b):
         self.add_buffer = x.new(x.size(0)).fill_(1)
         y.addr_(self.add_buffer, b)
         return y
-
+    @staticmethod
     def backward(self, dy):
         '''
         backprop with meprop
diff --git a/src/pytorch/modules.py b/src/pytorch/modules.py
index 06a6ac7..45cf883 100644
--- a/src/pytorch/modules.py
+++ b/src/pytorch/modules.py
@@ -36,9 +36,9 @@ def reset_parameters(self):
 
     def forward(self, x):
         if self.unified:
-            return linearUnified(self.k)(x, self.w, self.b)
+            return linearUnified.apply(x, self.w, self.b)#add apply
         else:
-            return linear(self.k)(x, self.w, self.b)
+            return linear.apply(x, self.w, self.b)#add apply
 
     def __repr__(self):
         return '{} ({} -> {} <- {}{})'.format(self.__class__.__name__,
diff --git a/src/pytorch/util.py b/src/pytorch/util.py
index 632ca41..90f2554 100644
--- a/src/pytorch/util.py
+++ b/src/pytorch/util.py
@@ -10,7 +10,7 @@
 import torch.nn.functional as F
 import torch.optim as optim
 import torch.utils.data
-from torch.autograd import Variable
+from torch.autograd import Variable#deprecated- no longer supported
 
 from model import NetLayer
 
@@ -102,8 +102,7 @@ def _train(self, model, opt):
         utime = 0
         tloss = 0
         for bid, (data, target) in enumerate(self.trainloader):
-            data, target = Variable(data).cuda(), Variable(
-                target.view(-1)).cuda()
+            data, target = Variable(data).cuda(), Variable(target.view(-1)).cuda()#Deprecated- need to replace
             start = torch.cuda.Event(True)
             end = torch.cuda.Event(True)
 
@@ -145,8 +144,7 @@ def _evaluate(self, model, loader, name='test'):
         test_loss = 0
         correct = 0
         for data, target in loader:
-            data, target = Variable(
-                data, volatile=True).cuda(), Variable(target).cuda()
+            data, target = Variable(data, volatile=True).cuda(), Variable(target).cuda()# Deprecated- need to update
             output = model(data)
             test_loss += F.nll_loss(output, target).data[0]
             pred = output.data.max(1)[
@@ -204,7 +202,7 @@ def run(self, k=None, epoch=None):
             start = torch.cuda.Event(True)
             end = torch.cuda.Event(True)
             start.record()
-            loss, ft, bt, ut = self._train(model, opt)
+            loss, ft, bt, ut = self._train(model, opt)#model is trained here
             end.record()
             end.synchronize()
             ttime = start.elapsed_time(end)
@@ -215,7 +213,7 @@ def run(self, k=None, epoch=None):
             btime.append(bt)
             utime.append(ut)
             # predict
-            curacc = self._evaluate(model, self.devloader, 'dev')
+            curacc = self._evaluate(model, self.devloader, 'dev')#evaluation here
             if curacc > acc:
                 e = t
                 acc = curacc