docs: update excel add in docs to reflect changes in the plugin (#406)

Nixtla · Jul 4, 2024 · c7eac7d · c7eac7d
1 parent 1491cc5
commit c7eac7d
Showing 1 changed file with 33 additions and 29 deletions.
diff --git a/nbs/docs/reference/03_excel_addin.ipynb b/nbs/docs/reference/03_excel_addin.ipynb
@@ -6,6 +6,11 @@
    "source": [
     "# TimeGPT Excel Add-in\n",
     "\n",
+    "## Installation\n",
+    "\n",
+    "Head to the [TimeGTP excel add-in page in Microsoft Appsource](https://appsource.microsoft.com/en-us/product/office/WA200006429?tab=Overview) and click on \"Get it now\"\n",
+    "\n",
+    "## Usage\n",
     "> 📘 Access token required\n",
     "> \n",
     "> The TimeGPT Excel Add-in requires an access token. Get your token on the [Nixtla Dashboard](http://dashboard.nixtla.io).\n",
@@ -20,37 +25,31 @@
     "\n",
     "If it is your first time using the add-in, a welcome screen will guide you to a settings screen to configure your access token.\n",
     "\n",
-    "To access the setting later, click the three dots in the top right and click `Settings`.\n",
+    "To access the setting later, click the gear icon in the top left.\n",
     "\n",
     "### Data Requirements\n",
     "\n",
+    "* Put your dates in one column and your values in another.\n",
+    "* Ensure your date format is recognized as a valid date by excel.\n",
+    "* Ensure your values are recognized as valid number by excel.\n",
     "* All data inputs must exist in the same worksheet. The add-in does not support forecasting using multiple worksheets.\n",
-    "\n",
-    "* Time series and exogenous variables (optional) must exist in their own columns. Values that you want predicted can be left blank.\n",
-    "\n",
-    "* Headers are optional.\n",
+    "* Do not include headers\n",
     "\n",
     "Example:\n",
     "\n",
-    "| timestamp      | value  | exogenous1 | exogenous2 |\n",
-    "| :------------- | :----- | :--------- | :--------- |\n",
-    "| 12/1/2016 0:00 | 72     | 61507      | 71066      |\n",
-    "| 12/1/16 1:00   | 65.8   | 59528      | 67311      |\n",
-    "| 12/1/16 2:00   | 59.99  | 58812      | 67470      |\n",
-    "| 12/1/16 3:00   | 50.69  | 57676      | 64529      |\n",
-    "| 12/1/16 4:00   | 52.58  | 56804      | 62773      |\n",
-    "| 12/1/16 5:00   | 65.05  | 58491      | 63780      |\n",
-    "| 12/1/16 6:00   | 80.4   | 61622      | 68346      |\n",
-    "| 12/1/16 7:00   | 200    | 68070      | 76799      |\n",
-    "| 12/1/16 8:00   | 200.63 | 70490      | 81312      |\n",
-    "| 12/1/16 9:00   | 155.47 | 70187      | 80506      |\n",
-    "| 12/1/16 10:00  | 150.91 | 68304      | 80066      |\n",
-    "| 12/1/16 11:00  |        | 66789      | 78745      |\n",
-    "| 12/1/16 12:00  |        | 65456      | 77975      |\n",
-    "| 12/1/16 13:00  |        | 64818      | 77031      |\n",
-    "| 12/1/16 14:00  |        | 63390      | 73875      |\n",
-    "| 12/1/16 15:00  |        | 63135      | 71576      |\n",
-    "| 12/1/16 16:00  |        | 63316      | 70560      |\n",
+    "| dates      | values  | \n",
+    "| :------------- | :----- | \n",
+    "| 12/1/16 0:00 | 72     | \n",
+    "| 12/1/16 1:00   | 65.8   | \n",
+    "| 12/1/16 2:00   | 59.99  | \n",
+    "| 12/1/16 3:00   | 50.69  | \n",
+    "| 12/1/16 4:00   | 52.58  | \n",
+    "| 12/1/16 5:00   | 65.05  | \n",
+    "| 12/1/16 6:00   | 80.4   | \n",
+    "| 12/1/16 7:00   | 200    | \n",
+    "| 12/1/16 8:00   | 200.63 | \n",
+    "| 12/1/16 9:00   | 155.47 | \n",
+    "| 12/1/16 10:00  | 150.91 | \n",
     "\n",
     "#### Forecasting\n",
     "\n",
@@ -62,15 +61,20 @@
     "\n",
     "* **Values** - The column of the timeseries values for each point in time\n",
     "\n",
-    "* **Exogenous Variable(s)** - The exogenous variables for each point in time. This should also include forecasting horizon (fh) additional timestamps to calculate the future values.\n",
+    "* **Frequency** - The frequency of the data (hourly / daily / weekly / monthly)\n",
+    "\n",
+    "* **Horizon** - The forecasting horizon. This represents the number of time steps into the future that the forecast should predict.\n",
+    "\n",
+    "When you're ready, click Run Forecast to generate the predicted values. The add-in will generate a new worksheet with the prediced values and a chart depicting the forecast.\n",
+    "\n",
+    "\n",
+    "#### Anomaly Detection\n",
     "\n",
-    "* **Frequency** - The frequency of the data\n",
+    "The requirements are the same as for the forecasting functionality, so if you already tried it you are ready to run the anomaly detection one. Go to the main page in the add-in and select \"Anomaly Detection\", then choose your dates and values cell ranges and click on submit. We'll run the model and mark the anomalies cells in yellow while adding a third column for expected values with a green background.\n",
     "\n",
-    "* **Frequency Horizon** - The forecasting horizon. This represents the number of time steps into the future that the forecast should predict. This is usually the number of empty cells in the value column.\n",
     "\n",
-    "* **Finetune Steps (advanced)** - The number of tuning steps used to train the large time model on the data. Set this value to 0 for zero-shot inference, i.e., to make predictions without any further model tuning.\n",
     "\n",
-    "When you're ready, click Run Forecast to generate the predicted values. The add-in will generate a new worksheet with the prediced values and a chart depicting the forecast."
+    "\n"
    ]
   }
  ],