diff --git a/notebooks/end2end_example/cybersecurity/1-train-mlp-with-brevitas.ipynb b/notebooks/end2end_example/cybersecurity/1-train-mlp-with-brevitas.ipynb
index ff4c5704002219ca18bb07eeb8c768f860f3ffbf..e0ce00c1beefe8172ac5fd2aeaaa076b9bb574c1 100644
--- a/notebooks/end2end_example/cybersecurity/1-train-mlp-with-brevitas.ipynb
+++ b/notebooks/end2end_example/cybersecurity/1-train-mlp-with-brevitas.ipynb
@@ -103,23 +103,23 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 2,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "--2021-02-24 16:57:33--  https://zenodo.org/record/4519767/files/unsw_nb15_binarized.npz?download=1\n",
+      "--2021-05-10 18:14:00--  https://zenodo.org/record/4519767/files/unsw_nb15_binarized.npz?download=1\n",
       "Resolving zenodo.org (zenodo.org)... 137.138.76.77\n",
       "Connecting to zenodo.org (zenodo.org)|137.138.76.77|:443... connected.\n",
       "HTTP request sent, awaiting response... 200 OK\n",
       "Length: 13391907 (13M) [application/octet-stream]\n",
-      "Saving to: 'unsw_nb15_binarized.npz'\n",
+      "Saving to: ‘unsw_nb15_binarized.npz’\n",
       "\n",
-      "unsw_nb15_binarized 100%[===================>]  12.77M  2.17MB/s    in 8.9s    \n",
+      "unsw_nb15_binarized 100%[===================>]  12.77M  3.96MB/s    in 3.4s    \n",
       "\n",
-      "2021-02-24 16:57:44 (1.44 MB/s) - 'unsw_nb15_binarized.npz' saved [13391907/13391907]\n",
+      "2021-05-10 18:14:04 (3.77 MB/s) - ‘unsw_nb15_binarized.npz’ saved [13391907/13391907]\n",
       "\n"
      ]
     }
@@ -422,7 +422,9 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Training loss = 0.132480 test accuracy = 0.797989: 100%|██████████| 10/10 [00:58<00:00,  5.70s/it]\n"
+      "Training loss:   0%|          | 0/10 [00:00<?, ?it/s]/opt/conda/lib/python3.8/site-packages/torch/autograd/__init__.py:130: UserWarning: CUDA initialization: Found no NVIDIA driver on your system. Please check that you have an NVIDIA GPU and installed a driver from http://www.nvidia.com/Download/index.aspx (Triggered internally at  /opt/conda/conda-bld/pytorch_1607370172916/work/c10/cuda/CUDAFunctions.cpp:100.)\n",
+      "  Variable._execution_engine.run_backward(\n",
+      "Training loss = 0.131708 test accuracy = 0.805398: 100%|██████████| 10/10 [01:04<00:00,  6.42s/it]\n"
      ]
     }
    ],
@@ -457,7 +459,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -483,7 +485,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 432x288 with 1 Axes>"
       ]
@@ -507,7 +509,7 @@
     {
      "data": {
       "text/plain": [
-       "0.7979886313948404"
+       "0.8053976582616722"
       ]
      },
      "execution_count": 15,
@@ -546,7 +548,7 @@
     {
      "data": {
       "text/plain": [
-       "IncompatibleKeys(missing_keys=[], unexpected_keys=[])"
+       "<All keys matched successfully>"
       ]
      },
      "execution_count": 17,
@@ -773,12 +775,14 @@
     "# Export to FINN-ONNX <a id=\"export_finn_onnx\" ></a>\n",
     "\n",
     "\n",
-    "[ONNX](https://onnx.ai/) is an open format built to represent machine learning models, and the FINN compiler expects an ONNX model as input. We'll now export our network into ONNX to be imported and used in FINN for the next notebooks. Note that the particular ONNX representation used for FINN differs from standard ONNX, you can read more about this [here](https://finn.readthedocs.io/en/latest/internals.html#intermediate-representation-finn-onnx)."
+    "[ONNX](https://onnx.ai/) is an open format built to represent machine learning models, and the FINN compiler expects an ONNX model as input. We'll now export our network into ONNX to be imported and used in FINN for the next notebooks. Note that the particular ONNX representation used for FINN differs from standard ONNX, you can read more about this [here](https://finn.readthedocs.io/en/latest/internals.html#intermediate-representation-finn-onnx).\n",
+    "\n",
+    "You can see below how we export a trained network in Brevitas into a FINN-compatible ONNX representation. Note how we create a `QuantTensor` instance with dummy data to tell Brevitas how our inputs look like, which will be used to set the input quantization annotation on the exported model."
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 27,
    "metadata": {
     "scrolled": true
    },
@@ -787,69 +791,38 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Model saved to cybsec-mlp.onnx\n"
+      "Model saved to cybsec-mlp-ready.onnx\n"
      ]
     },
     {
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "/opt/conda/lib/python3.6/site-packages/ipykernel_launcher.py:15: TracerWarning: torch.tensor results are registered as constants in the trace. You can safely ignore this warning if you use this function to create tensors out of constant variables that would be the same every time you call this function. In any other case, this might cause the trace to be incorrect.\n",
-      "  from ipykernel import kernelapp as app\n"
+      "<ipython-input-22-78c27bb59095>:15: TracerWarning: torch.tensor results are registered as constants in the trace. You can safely ignore this warning if you use this function to create tensors out of constant variables that would be the same every time you call this function. In any other case, this might cause the trace to be incorrect.\n",
+      "  x = (x + torch.tensor([1.0])) / 2.0\n"
      ]
     }
    ],
    "source": [
     "import brevitas.onnx as bo\n",
+    "from brevitas.quant_tensor import QuantTensor\n",
     "\n",
-    "export_onnx_path = \"cybsec-mlp.onnx\"\n",
+    "ready_model_filename = \"cybsec-mlp-ready.onnx\"\n",
     "input_shape = (1, 600)\n",
-    "bo.export_finn_onnx(model_for_export, input_shape, export_onnx_path)\n",
-    "\n",
-    "print(\"Model saved to %s\" % export_onnx_path)"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "## One final fix: input datatype\n",
-    "\n",
-    "There's one more thing we'll do: we will mark the input tensor datatype as `DataType.BIPOLAR`, which will be used by the compiler later on. To do this, we'll utilize the `ModelWrapper` component from FINN, which lets us examine and manipulate the ONNX graph in an easier way.\n",
-    "\n",
-    "*In the near future it will be possible to add this information to the model [while exporting](https://github.com/Xilinx/brevitas/issues/232), instead of having to add it manually.*"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 26,
-   "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "Input tensor name: 0\n",
-      "Input tensor shape: [1, 600]\n",
-      "Input tensor datatype: DataType.BIPOLAR\n"
-     ]
-    }
-   ],
-   "source": [
-    "from finn.core.modelwrapper import ModelWrapper\n",
-    "from finn.core.datatype import DataType\n",
+    "# create a QuantTensor instance to mark input as bipolar during export\n",
+    "input_a = np.random.randint(0, 1, size=input_shape).astype(np.float32)\n",
+    "input_a = 2 * input_a - 1\n",
+    "scale = 1.0\n",
+    "input_t = torch.from_numpy(input_a * scale)\n",
+    "input_qt = QuantTensor(\n",
+    "    input_t, scale=torch.tensor(scale), bit_width=torch.tensor(1.0), signed=True\n",
+    ")\n",
     "\n",
-    "finn_model = ModelWrapper(export_onnx_path)\n",
+    "bo.export_finn_onnx(\n",
+    "    model_for_export, export_path=ready_model_filename, input_t=input_qt\n",
+    ")\n",
     "\n",
-    "finnonnx_in_tensor_name = finn_model.graph.input[0].name\n",
-    "finnonnx_model_in_shape = finn_model.get_tensor_shape(finnonnx_in_tensor_name)\n",
-    "finn_model.set_tensor_datatype(finnonnx_in_tensor_name, DataType.BIPOLAR)\n",
-    "print(\"Input tensor name: %s\" % finnonnx_in_tensor_name)\n",
-    "print(\"Input tensor shape: %s\" % str(finnonnx_model_in_shape))\n",
-    "print(\"Input tensor datatype: %s\" % str(finn_model.get_tensor_datatype(finnonnx_in_tensor_name)))\n",
-    "\n",
-    "ready_model_filename = \"cybsec-mlp-ready.onnx\"\n",
-    "finn_model.save(ready_model_filename)"
+    "print(\"Model saved to %s\" % ready_model_filename)"
    ]
   },
   {
@@ -870,7 +843,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 27,
+   "execution_count": 28,
    "metadata": {},
    "outputs": [
     {
@@ -894,10 +867,10 @@
        "        "
       ],
       "text/plain": [
-       "<IPython.lib.display.IFrame at 0x7f77214fa630>"
+       "<IPython.lib.display.IFrame at 0x7f49738bffa0>"
       ]
      },
-     "execution_count": 27,
+     "execution_count": 28,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -940,7 +913,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.6.8"
+   "version": "3.8.5"
   }
  },
  "nbformat": 4,