WEBVTT 00:01.000 --> 00:05.640 Hello and welcome to future. In this video, I want to show you how quickly you can use futureEXPERT 00:05.640 --> 00:11.340 to get your first forecasts. We'll start on our website future-forecasting.de. 00:11.340 --> 00:15.680 Under "Start for free," we come to the registration form. Here, we only need to 00:15.680 --> 00:21.160 provide our first and last name and our email address. Today, I want to register as an 00:21.160 --> 00:25.440 individual, and confirm that I have the terms and conditions and the privacy policy, which I have 00:25.440 --> 00:30.660 taken note of. When I click on "create account," I will receive an email in which I can confirm my 00:30.660 --> 00:35.340 email address. Once my email address is confirmed, I will be automatically redirected to 00:35.340 --> 00:41.640 our frontend now.future-forecasting.de. To use futureEXPERT in my 00:41.640 --> 00:47.320 Python environment, I simply install the futureEXPERT package, which is available on GitHub. 00:47.320 --> 00:53.320 Here you can find a lot of information, such as the command to install the package. 00:54.320 --> 00:57.420 I can simply copy this and run it in my environment. 00:58.420 --> 01:01.180 The installation takes a short moment. 01:01.180 --> 01:03.200 Besides futureEXPERT, a few other helpful packages are installed, 01:03.200 --> 01:09.080 which allow, for example, creating plots or similar things. Now let's switch back to the 01:09.080 --> 01:14.960 frontend to prepare our data. In the "Data" menu field, I have the option to upload a dataset. 01:14.960 --> 01:18.280 I can simply upload this via drag-and-drop. Once it is uploaded, 01:18.280 --> 01:25.540 I can use the CHECK-IN feature to create timeseries, which will be the basis for my forecasts. I delete all the columns that I 01:25.540 --> 01:34.780 don't need and could also delete individual rows now. In the next step, I define my raw data. 01:34.780 --> 01:44.220 I specify which column contains the date and how it's formatted. I then select "demand" as value column 01:44.720 --> 01:51.480 and that the 'Material' column contains grouping information. Then I click 'next'. 01:51.480 --> 02:00.980 If there were missing values or other problems in our raw data, we would be notified here. 02:00.980 --> 02:03.840 That is not the case for us, so I can simply click 'next'. 02:03.840 --> 02:12.540 I want to create monthly time series and generate a forecast for each material. So I select this as the data level. 02:12.540 --> 02:22.300 The "Demand" column contains our values, and missing values in the target time series should be filled with zero. 02:24.140 --> 02:31.040 Once the time series are prepared, I can simply copy the version ID and continue in my Python environment 02:31.040 --> 02:34.140 to create forecasts. As a template, I use the “getting started” notebook, 02:34.140 --> 02:39.400 which I can find on GitHub. First, I need to configure my access credentials. I can either do this 02:39.400 --> 02:43.980 during the client initialization or store them in the .env file in my main directory. 02:43.980 --> 02:51.760 The individual classes allow me to make settings to tailor the forecasts precisely to my needs. 02:51.760 --> 02:58.680 For the forecast horizon, I choose twelve months, so a full year. Additionally, I want to use ensemble methods. 02:58.680 --> 03:06.020 I know that my materials are often sold in packages with fixed sizes. Therefore, I use quantization detection, 03:06.020 --> 03:12.160 which also ensures that the forecasts are predicted in these package sizes. I don't want to make any further settings. 03:12.160 --> 03:15.760 The version ID is the ID that we copied from our frontend. 03:15.760 --> 03:21.620 As soon as I run start_forecast, the forecasts will be initiated. This takes a few minutes. 03:23.620 --> 03:27.780 The forecasts run in the background. At regular intervals, we check if the calculations are complete. 03:27.780 --> 03:32.180 Once this is the case, we can easily download the results. A quick look at the plots 03:35.180 --> 03:40.520 shows meaningful and plausible results. And so, in just a few minutes, we've gone from registration 03:40.520 --> 03:44.540 to fully generated forecasts. Of course, futureEXPERT offers much more 03:44.540 --> 03:48.580 functionality than what I've used here. For a detailed description of the results, 03:48.580 --> 03:54.560 the use of covariates, or complex workflows, you can find corresponding example notebooks on our GitHub. 03:54.560 --> 03:58.560 If you have any further questions or problems, just send us an email 03:58.560 --> 04:03.640 to support[at]future-forecasting.de. Stay up to date by following us on 04:03.640 --> 04:07.040 LinkedIn or subscribing to our newsletter. If you have ideas or 04:07.040 --> 04:11.340 requests for additional features, you can create an issue on GitHub at any time. 04:11.340 --> 04:12.080 We look forward to hearing from you.