diff --git a/transformer/Models.py b/transformer/Models.py
index 9af9fd6..884dff7 100644
--- a/transformer/Models.py
+++ b/transformer/Models.py
@@ -29,20 +29,17 @@ def __init__(self, d_hid, n_position=200):
         self.register_buffer('pos_table', self._get_sinusoid_encoding_table(n_position, d_hid))
 
     def _get_sinusoid_encoding_table(self, n_position, d_hid):
-        ''' Sinusoid position encoding table '''
-        # TODO: make it with torch instead of numpy
-
+    ''' Sinusoid position encoding table '''
+        
         def get_position_angle_vec(position):
-            return [position / np.power(10000, 2 * (hid_j // 2) / d_hid) for hid_j in range(d_hid)]
+            return [position / torch.pow(10000, 2 * (hid_j // 2) / d_hid) for hid_j in range(d_hid)]
 
-        sinusoid_table = np.array([get_position_angle_vec(pos_i) for pos_i in range(n_position)])
-        sinusoid_table[:, 0::2] = np.sin(sinusoid_table[:, 0::2])  # dim 2i
-        sinusoid_table[:, 1::2] = np.cos(sinusoid_table[:, 1::2])  # dim 2i+1
+        sinusoid_table = torch.tensor([get_position_angle_vec(pos_i) for pos_i in range(n_position)], dtype=torch.float32)
+        sinusoid_table[:, 0::2] = torch.sin(sinusoid_table[:, 0::2])  # dim 2i
+        sinusoid_table[:, 1::2] = torch.cos(sinusoid_table[:, 1::2])  # dim 2i+1
 
-        return torch.FloatTensor(sinusoid_table).unsqueeze(0)
+        return sinusoid_table.unsqueeze(0)
 
-    def forward(self, x):
-        return x + self.pos_table[:, :x.size(1)].clone().detach()
 
 
 class Encoder(nn.Module):