docs: move magic command forward since it restarts interpreter

notebooks/community/aisamples/AIsample - Time Series Forecasting.ipynb

@@ -25,6 +25,26 @@
                 "We will build up a model to forcast monthly volume of property trade based on history data. In order to forecast, we will use [Facebook Prophet](https://facebook.github.io/prophet/), which provides fast and automated forecast procedure and handles seasonality well."
             ]
         },
+        {
+            "cell_type": "markdown",
+            "metadata": {},
+            "source": [
+                "## Install Prophet\n",
+                "\n",
+                "Let's first install [Facebook Prophet](https://facebook.github.io/prophet/). It uses a decomposable time series model which consist of three main components: trend, seasonality and holidays. \n",
+                "For the trend part, Prophet assumes piece-wise constant rate of growth with automatic chagne point selection.\n",
+                "For seasonality part, Prophet models weekly and yearly seasonality using Fourier Series. Since we are using monthly data, so we won't have weekly seasonality and won't considering holidays."
+            ]
+        },
+        {
+            "cell_type": "code",
+            "execution_count": null,
+            "metadata": {},
+            "outputs": [],
+            "source": [
+                "!pip install prophet"
+            ]
+        },
         {
             "cell_type": "markdown",
             "metadata": {},
@@ -248,26 +268,6 @@
                 "## Step 3: Model Training and Evaluation"
             ]
         },
-        {
-            "cell_type": "markdown",
-            "metadata": {},
-            "source": [
-                "### Install Prophet\n",
-                "\n",
-                "Now let's install [Facebook Prophet](https://facebook.github.io/prophet/). It uses a decomposable time series model which consist of three main components: trend, seasonality and holidays. \n",
-                "For the trend part, Prophet assumes piece-wise constant rate of growth with automatic chagne point selection.\n",
-                "For seasonality part, Prophet models weekly and yearly seasonality using Fourier Series. Since we are using monthly data, so we won't have weekly seasonality and won't considering holidays."
-            ]
-        },
-        {
-            "cell_type": "code",
-            "execution_count": null,
-            "metadata": {},
-            "outputs": [],
-            "source": [
-                "%pip install prophet"
-            ]
-        },
         {
             "cell_type": "markdown",
             "metadata": {},