Overview documentation: mention versioning and link to dedicated document.

PiperOrigin-RevId: 257510847

Author: pierrot0

docs/overview.ipynb

@@ -213,31 +213,32 @@
         "mnist_train"
       ],
       "execution_count": 5,
-      "outputs": [
-        {
-          "output_type": "stream",
-          "text": [
-            "INFO:tensorflow:Reusing dataset mnist (/root/tensorflow_datasets/mnist/1.0.0)\n",
-            "WARNING:tensorflow:From /usr/local/lib/python2.7/dist-packages/tensorflow/python/ops/control_flow_ops.py:423: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
-            "Instructions for updating:\n",
-            "Colocations handled automatically by placer.\n"
-          ],
-          "name": "stdout"
-        },
-        {
-          "output_type": "execute_result",
-          "data": {
-            "text/plain": [
-              "<DatasetV1Adapter shapes: {image: (28, 28, 1), label: ()}, types: {image: tf.uint8, label: tf.int64}>"
-            ]
-          },
-          "metadata": {
-            "tags": []
-          },
-          "execution_count": 5
-        }
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "D4A6yp_Ii3G-",
+        "colab_type": "text"
+      },
+      "source": [
+        "When loading a dataset, the canonical default version is used. It is however recommended to specify the major version of the dataset to use, and to advertise which version of the dataset was used in your results. See the\n",
+        "[documentation on datasets versioning](https://github.com/tensorflow/datasets/blob/master/docs/) for more details."
       ]
     },
+    {
+      "cell_type": "code",
+      "metadata": {
+        "id": "OCXCz-vhj0kE",
+        "colab_type": "code",
+        "colab": {}
+      },
+      "source": [
+        "mnist = tfds.load(\"mnist:1.*.*\")"
+      ],
+      "execution_count": 0,
+      "outputs": []
+    },
     {
       "metadata": {
         "colab_type": "text",