Updated the srnn ipython notebook (gpu support)

pytorch/examples/SRNN/SRNN_Example.ipynb

@@ -29,6 +29,7 @@
     "from pytorch_edgeml.graph.rnn import SRNN2\n",
     "from pytorch_edgeml.trainer.srnnTrainer import SRNNTrainer\n",
     "import pytorch_edgeml.utils as utils"
+    "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")"
    ]
   },
   {
@@ -127,8 +128,8 @@
     }
    ],
    "source": [
-    "srnn2 = SRNN2(numInput, numClasses, hiddenDim0, hiddenDim1, cellType)\n",
-    "trainer = SRNNTrainer(srnn2, learningRate, lossType='xentropy')"
+    "srnn2 = SRNN2(numInput, numClasses, hiddenDim0, hiddenDim1, cellType).to(device)\n",
+    "trainer = SRNNTrainer(srnn2, learningRate, lossType='xentropy', device=device)"
    ]
   },
   {