[Script Info] Title: [Events] Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text Dialogue: 0,0:00:01.92,0:00:04.68,Default,,0000,0000,0000,,Great, so I think we can start since the Dialogue: 0,0:00:04.68,0:00:07.86,Default,,0000,0000,0000,,meeting is recorded, so if everyone, uh Dialogue: 0,0:00:07.86,0:00:11.16,Default,,0000,0000,0000,,jump-jumps in later can-can watch the Dialogue: 0,0:00:11.16,0:00:12.42,Default,,0000,0000,0000,,recording. Dialogue: 0,0:00:12.42,0:00:15.78,Default,,0000,0000,0000,,So, hi everyone and welcome to this Dialogue: 0,0:00:15.78,0:00:18.00,Default,,0000,0000,0000,,um, Cloud Skill Challenge study session Dialogue: 0,0:00:18.00,0:00:20.88,Default,,0000,0000,0000,,around a create classification models Dialogue: 0,0:00:20.88,0:00:24.00,Default,,0000,0000,0000,,with Azure Machine learning designer. Dialogue: 0,0:00:24.00,0:00:27.24,Default,,0000,0000,0000,,So today I'm thrilled to be here with Dialogue: 0,0:00:27.24,0:00:29.82,Default,,0000,0000,0000,,John. Uh, John do you mind introduce briefly Dialogue: 0,0:00:29.82,0:00:31.62,Default,,0000,0000,0000,,yourself? Dialogue: 0,0:00:31.62,0:00:34.16,Default,,0000,0000,0000,,Uh, thank you Carlotta. Hello everyone. Dialogue: 0,0:00:34.16,0:00:38.16,Default,,0000,0000,0000,,Welcome to our workshop today. I hope Dialogue: 0,0:00:38.16,0:00:40.56,Default,,0000,0000,0000,,that you are all excited for it. I am Dialogue: 0,0:00:40.56,0:00:43.14,Default,,0000,0000,0000,,John Aziz a gold Microsoft Learn student Dialogue: 0,0:00:43.14,0:00:47.46,Default,,0000,0000,0000,,ambassador and I will be here with, uh, Dialogue: 0,0:00:47.46,0:00:50.76,Default,,0000,0000,0000,,Carlotta to, like, do the practical part Dialogue: 0,0:00:50.76,0:00:53.82,Default,,0000,0000,0000,,about this module of the Cloud Skills Dialogue: 0,0:00:53.82,0:00:57.00,Default,,0000,0000,0000,,Challenge. Thank you for having me. Dialogue: 0,0:00:57.00,0:00:59.22,Default,,0000,0000,0000,,Perfect, thanks John. So for those who Dialogue: 0,0:00:59.22,0:01:03.44,Default,,0000,0000,0000,,don't know me I'm Carlotta Castelluccio, Dialogue: 0,0:01:03.44,0:01:06.48,Default,,0000,0000,0000,,based in Italy and focused on AI Dialogue: 0,0:01:06.48,0:01:08.76,Default,,0000,0000,0000,,machine learning technologies and about Dialogue: 0,0:01:08.76,0:01:11.20,Default,,0000,0000,0000,,the use in education. Dialogue: 0,0:01:11.20,0:01:12.34,Default,,0000,0000,0000,,Um, so, Dialogue: 0,0:01:12.74,0:01:14.54,Default,,0000,0000,0000,,um this Cloud Skill Challenge study Dialogue: 0,0:01:14.54,0:01:17.12,Default,,0000,0000,0000,,session is based on a learn module, a Dialogue: 0,0:01:17.12,0:01:21.08,Default,,0000,0000,0000,,dedicated learn module. I sent to you, uh Dialogue: 0,0:01:21.32,0:01:23.94,Default,,0000,0000,0000,,the link to this module, uh, in the chat Dialogue: 0,0:01:23.94,0:01:25.62,Default,,0000,0000,0000,,in a way that you can follow along the Dialogue: 0,0:01:25.62,0:01:28.68,Default,,0000,0000,0000,,module if you want, or just have a look at Dialogue: 0,0:01:28.68,0:01:32.47,Default,,0000,0000,0000,,the module later at your own pace. Dialogue: 0,0:01:32.47,0:01:33.78,Default,,0000,0000,0000,,Um... Dialogue: 0,0:01:33.78,0:01:37.02,Default,,0000,0000,0000,,So, before starting I would also like to Dialogue: 0,0:01:37.02,0:01:40.62,Default,,0000,0000,0000,,remember to remember you, uh, the code of Dialogue: 0,0:01:40.62,0:01:43.44,Default,,0000,0000,0000,,conduct and guidelines of our student Dialogue: 0,0:01:43.44,0:01:47.51,Default,,0000,0000,0000,,ambassadors community. So please during this Dialogue: 0,0:01:47.51,0:01:51.00,Default,,0000,0000,0000,,meeting be respectful and inclusive and Dialogue: 0,0:01:51.00,0:01:53.58,Default,,0000,0000,0000,,be friendly, open, and welcoming and Dialogue: 0,0:01:53.58,0:01:56.16,Default,,0000,0000,0000,,respectful of other-each other Dialogue: 0,0:01:56.16,0:01:57.72,Default,,0000,0000,0000,,differences. Dialogue: 0,0:01:57.72,0:02:01.20,Default,,0000,0000,0000,,If you want to learn more about the code Dialogue: 0,0:02:01.20,0:02:03.39,Default,,0000,0000,0000,,of conduct, you can use this link in the Dialogue: 0,0:02:03.39,0:02:08.88,Default,,0000,0000,0000,,deck: aka.ms/SACoC. Dialogue: 0,0:02:09.66,0:02:11.73,Default,,0000,0000,0000,,And now we are, Dialogue: 0,0:02:11.73,0:02:15.42,Default,,0000,0000,0000,,um, we are ready to to start our session. Dialogue: 0,0:02:15.42,0:02:18.96,Default,,0000,0000,0000,,So as we mentioned it we are going to Dialogue: 0,0:02:18.96,0:02:21.98,Default,,0000,0000,0000,,focus on classification models and Azure ML, Dialogue: 0,0:02:21.98,0:02:24.90,Default,,0000,0000,0000,,uh, today. So, first of all, we are going Dialogue: 0,0:02:24.90,0:02:28.43,Default,,0000,0000,0000,,to, um, identify, uh, the kind of Dialogue: 0,0:02:28.43,0:02:31.08,Default,,0000,0000,0000,,um, of scenarios in which you should Dialogue: 0,0:02:31.08,0:02:34.49,Default,,0000,0000,0000,,choose to use a classification model. Dialogue: 0,0:02:34.49,0:02:36.66,Default,,0000,0000,0000,,We're going to introduce Azure Machine Dialogue: 0,0:02:36.66,0:02:39.06,Default,,0000,0000,0000,,Learning and Azure Machine Designer. Dialogue: 0,0:02:39.06,0:02:41.88,Default,,0000,0000,0000,,We're going to understand, uh, which are Dialogue: 0,0:02:41.88,0:02:43.68,Default,,0000,0000,0000,,the steps to follow, to create a Dialogue: 0,0:02:43.68,0:02:46.20,Default,,0000,0000,0000,,classification model in Azure Machine Dialogue: 0,0:02:46.20,0:02:48.08,Default,,0000,0000,0000,,Learning, and then John will, Dialogue: 0,0:02:48.08,0:02:49.50,Default,,0000,0000,0000,,um, Dialogue: 0,0:02:49.50,0:02:52.22,Default,,0000,0000,0000,,lead an amazing demo about training and Dialogue: 0,0:02:52.22,0:02:54.30,Default,,0000,0000,0000,,publishing a classification model in Dialogue: 0,0:02:54.30,0:02:57.00,Default,,0000,0000,0000,,Azure ML Designer. Dialogue: 0,0:02:57.00,0:02:59.82,Default,,0000,0000,0000,,So, let's start from the beginning. Let's Dialogue: 0,0:02:59.82,0:03:02.64,Default,,0000,0000,0000,,start from identifying classification Dialogue: 0,0:03:02.64,0:03:05.22,Default,,0000,0000,0000,,machine learning scenarios. Dialogue: 0,0:03:05.22,0:03:07.64,Default,,0000,0000,0000,,So, first of all, what is classification? Dialogue: 0,0:03:07.64,0:03:09.96,Default,,0000,0000,0000,,Classification is a form of machine Dialogue: 0,0:03:09.96,0:03:12.12,Default,,0000,0000,0000,,learning that is used to predict which Dialogue: 0,0:03:12.12,0:03:15.60,Default,,0000,0000,0000,,category or class an item belongs to. For Dialogue: 0,0:03:15.60,0:03:17.34,Default,,0000,0000,0000,,example, we might want to develop a Dialogue: 0,0:03:17.34,0:03:19.80,Default,,0000,0000,0000,,classifier able to identify if an Dialogue: 0,0:03:19.80,0:03:22.20,Default,,0000,0000,0000,,incoming email should be filtered or not Dialogue: 0,0:03:22.20,0:03:25.08,Default,,0000,0000,0000,,according to the style, the sender, the Dialogue: 0,0:03:25.08,0:03:28.14,Default,,0000,0000,0000,,length of the email, etc. In this case, the Dialogue: 0,0:03:28.14,0:03:30.06,Default,,0000,0000,0000,,characteristics of the email are the Dialogue: 0,0:03:30.06,0:03:31.08,Default,,0000,0000,0000,,features. Dialogue: 0,0:03:31.08,0:03:34.20,Default,,0000,0000,0000,,And the label is a classification of Dialogue: 0,0:03:34.20,0:03:38.10,Default,,0000,0000,0000,,either a zero or one, representing a spam Dialogue: 0,0:03:38.10,0:03:40.86,Default,,0000,0000,0000,,or non-spam for the incoming email. So Dialogue: 0,0:03:40.86,0:03:42.36,Default,,0000,0000,0000,,this is an example of a binary Dialogue: 0,0:03:42.36,0:03:44.40,Default,,0000,0000,0000,,classifier. If you want to assign Dialogue: 0,0:03:44.40,0:03:46.26,Default,,0000,0000,0000,,multiple categories to the incoming Dialogue: 0,0:03:46.26,0:03:48.96,Default,,0000,0000,0000,,email like work letters, love letters, Dialogue: 0,0:03:48.96,0:03:52.08,Default,,0000,0000,0000,,complaints, or other categories, in this Dialogue: 0,0:03:52.08,0:03:54.00,Default,,0000,0000,0000,,case a binary classifier is no longer Dialogue: 0,0:03:54.00,0:03:55.74,Default,,0000,0000,0000,,enough, and we should develop a Dialogue: 0,0:03:55.74,0:03:58.32,Default,,0000,0000,0000,,multi-class classifier. So classification Dialogue: 0,0:03:58.32,0:04:00.60,Default,,0000,0000,0000,,is an example of what is called Dialogue: 0,0:04:00.60,0:04:02.52,Default,,0000,0000,0000,,supervised machine learning Dialogue: 0,0:04:02.52,0:04:05.28,Default,,0000,0000,0000,,in which you train a model using data Dialogue: 0,0:04:05.28,0:04:07.08,Default,,0000,0000,0000,,that includes both the features and Dialogue: 0,0:04:07.08,0:04:08.88,Default,,0000,0000,0000,,known values for label Dialogue: 0,0:04:08.88,0:04:11.10,Default,,0000,0000,0000,,so that the model learns to fit the Dialogue: 0,0:04:11.10,0:04:13.56,Default,,0000,0000,0000,,feature combinations to the label. Then, Dialogue: 0,0:04:13.56,0:04:15.42,Default,,0000,0000,0000,,after training has been completed, you Dialogue: 0,0:04:15.42,0:04:17.04,Default,,0000,0000,0000,,can use the train model to predict Dialogue: 0,0:04:17.04,0:04:19.50,Default,,0000,0000,0000,,labels for new items for-for which the Dialogue: 0,0:04:19.50,0:04:22.32,Default,,0000,0000,0000,,label is unknown. Dialogue: 0,0:04:22.32,0:04:25.44,Default,,0000,0000,0000,,But let's see some examples of scenarios Dialogue: 0,0:04:25.44,0:04:27.12,Default,,0000,0000,0000,,for classification machine learning Dialogue: 0,0:04:27.12,0:04:29.16,Default,,0000,0000,0000,,models. So, we already mentioned an Dialogue: 0,0:04:29.16,0:04:31.02,Default,,0000,0000,0000,,example of a solution in which we would Dialogue: 0,0:04:31.02,0:04:33.66,Default,,0000,0000,0000,,need a classifier, but let's explore Dialogue: 0,0:04:33.66,0:04:35.70,Default,,0000,0000,0000,,other scenarios for classification in Dialogue: 0,0:04:35.70,0:04:37.98,Default,,0000,0000,0000,,other industries. For example, you can use Dialogue: 0,0:04:37.98,0:04:40.38,Default,,0000,0000,0000,,a classification model for a health Dialogue: 0,0:04:40.38,0:04:43.68,Default,,0000,0000,0000,,clinic scenario, and use clinical data to Dialogue: 0,0:04:43.68,0:04:45.72,Default,,0000,0000,0000,,predict whether patient will become sick Dialogue: 0,0:04:45.72,0:04:47.06,Default,,0000,0000,0000,,or not. Dialogue: 0,0:04:47.06,0:04:49.55,Default,,0000,0000,0000,,You can use, um... Dialogue: 0,0:04:49.55,0:04:59.25,Default,,0000,0000,0000,,[NO AUDIO] Dialogue: 0,0:04:59.25,0:05:00.93,Default,,0000,0000,0000,,Carlotta, you are muted. Dialogue: 0,0:05:03.78,0:05:07.86,Default,,0000,0000,0000,,Oh, sorry. So, when I became muted, it's a Dialogue: 0,0:05:07.86,0:05:11.94,Default,,0000,0000,0000,,long time, or? You can use-you can use, uh Dialogue: 0,0:05:11.94,0:05:13.56,Default,,0000,0000,0000,,some models for classification. For Dialogue: 0,0:05:13.56,0:05:16.92,Default,,0000,0000,0000,,example, you can use...You were saying this. Dialogue: 0,0:05:16.92,0:05:21.66,Default,,0000,0000,0000,,Uh, so I was in this deck, or the previous one? Dialogue: 0,0:05:21.66,0:05:24.18,Default,,0000,0000,0000,,This one, like you have been muted Dialogue: 0,0:05:24.18,0:05:27.06,Default,,0000,0000,0000,,for, uh, one second [laughs]. Okay, okay perfect, Dialogue: 0,0:05:27.06,0:05:30.42,Default,,0000,0000,0000,,perfect. Uh, yeah I was talking...sorry for Dialogue: 0,0:05:30.42,0:05:33.28,Default,,0000,0000,0000,,that. So, I was talking about the possible Dialogue: 0,0:05:33.28,0:05:34.56,Default,,0000,0000,0000,,scenarios in which you, Dialogue: 0,0:05:34.56,0:05:37.32,Default,,0000,0000,0000,,you can use a classification model. Like Dialogue: 0,0:05:37.32,0:05:39.66,Default,,0000,0000,0000,,have clinic scenario, financial scenario, Dialogue: 0,0:05:39.66,0:05:41.70,Default,,0000,0000,0000,,or the third one is business type of Dialogue: 0,0:05:41.70,0:05:44.10,Default,,0000,0000,0000,,scenario. You can use characteristics of Dialogue: 0,0:05:44.10,0:05:45.90,Default,,0000,0000,0000,,small business to predict if a new Dialogue: 0,0:05:45.90,0:05:47.88,Default,,0000,0000,0000,,venture will succeed or not, for Dialogue: 0,0:05:47.88,0:05:49.56,Default,,0000,0000,0000,,example. And these are all types of Dialogue: 0,0:05:49.56,0:05:52.16,Default,,0000,0000,0000,,binary classification. Dialogue: 0,0:05:52.16,0:05:55.20,Default,,0000,0000,0000,,Uh, but today we are also going to talk Dialogue: 0,0:05:55.20,0:05:57.24,Default,,0000,0000,0000,,about Azure Machine Learning. So let's Dialogue: 0,0:05:57.24,0:05:58.14,Default,,0000,0000,0000,,see. Dialogue: 0,0:05:58.14,0:06:00.66,Default,,0000,0000,0000,,What is Azure Machine Learning? So Dialogue: 0,0:06:00.66,0:06:02.16,Default,,0000,0000,0000,,training and deploying an effective Dialogue: 0,0:06:02.16,0:06:04.20,Default,,0000,0000,0000,,machine learning model involves a lot of Dialogue: 0,0:06:04.20,0:06:06.54,Default,,0000,0000,0000,,work, much of it time-consuming and Dialogue: 0,0:06:06.54,0:06:08.88,Default,,0000,0000,0000,,resource intensive. So, Azure Machine Dialogue: 0,0:06:08.88,0:06:11.04,Default,,0000,0000,0000,,Learning is a cloud-based service that Dialogue: 0,0:06:11.04,0:06:12.78,Default,,0000,0000,0000,,helps simplify some of the tasks it Dialogue: 0,0:06:12.78,0:06:15.72,Default,,0000,0000,0000,,takes to prepare data, train a model, and Dialogue: 0,0:06:15.72,0:06:18.06,Default,,0000,0000,0000,,also deploy it as a predictive service. Dialogue: 0,0:06:18.06,0:06:20.22,Default,,0000,0000,0000,,So it helps that the scientists increase Dialogue: 0,0:06:20.22,0:06:22.38,Default,,0000,0000,0000,,their efficiency by automating many of Dialogue: 0,0:06:22.38,0:06:24.66,Default,,0000,0000,0000,,the time-consuming tasks associated to Dialogue: 0,0:06:24.66,0:06:27.54,Default,,0000,0000,0000,,creating and training a model. Dialogue: 0,0:06:27.54,0:06:29.52,Default,,0000,0000,0000,,And it enables them also to use Dialogue: 0,0:06:29.52,0:06:31.74,Default,,0000,0000,0000,,cloud-based compute resources that scale Dialogue: 0,0:06:31.74,0:06:33.72,Default,,0000,0000,0000,,effectively to handle large volumes of Dialogue: 0,0:06:33.72,0:06:36.30,Default,,0000,0000,0000,,data while incurring costs only when Dialogue: 0,0:06:36.30,0:06:38.70,Default,,0000,0000,0000,,actually used. Dialogue: 0,0:06:38.70,0:06:41.22,Default,,0000,0000,0000,,To use Azure Machine Learning, you, Dialogue: 0,0:06:41.22,0:06:43.20,Default,,0000,0000,0000,,first thing's first, you need to create a Dialogue: 0,0:06:43.20,0:06:44.94,Default,,0000,0000,0000,,workspace resource in your Azure Dialogue: 0,0:06:44.94,0:06:47.52,Default,,0000,0000,0000,,subscription, and you can then use these Dialogue: 0,0:06:47.52,0:06:50.22,Default,,0000,0000,0000,,workspace to manage data, compute Dialogue: 0,0:06:50.22,0:06:52.44,Default,,0000,0000,0000,,resources, code models and other Dialogue: 0,0:06:52.44,0:06:55.14,Default,,0000,0000,0000,,artifacts after you have created an Dialogue: 0,0:06:55.14,0:06:56.82,Default,,0000,0000,0000,,Azure Machine Learning workspace, you can Dialogue: 0,0:06:56.82,0:06:58.56,Default,,0000,0000,0000,,develop solutions with the Azure Machine Dialogue: 0,0:06:58.56,0:07:00.84,Default,,0000,0000,0000,,Learning service, either with developer Dialogue: 0,0:07:00.84,0:07:02.58,Default,,0000,0000,0000,,tools or the Azure Machine Learning Dialogue: 0,0:07:02.58,0:07:04.38,Default,,0000,0000,0000,,studio web portal. Dialogue: 0,0:07:04.38,0:07:06.36,Default,,0000,0000,0000,,In particular, Azure Machine Learning Dialogue: 0,0:07:06.36,0:07:07.80,Default,,0000,0000,0000,,studio is a web portal for Machine Dialogue: 0,0:07:07.80,0:07:09.72,Default,,0000,0000,0000,,Learning Solutions in Azure, and it Dialogue: 0,0:07:09.72,0:07:11.64,Default,,0000,0000,0000,,includes a wide range of features and Dialogue: 0,0:07:11.64,0:07:13.80,Default,,0000,0000,0000,,capabilities that help data scientists Dialogue: 0,0:07:13.80,0:07:16.26,Default,,0000,0000,0000,,prepare data, train models, publish Dialogue: 0,0:07:16.26,0:07:18.48,Default,,0000,0000,0000,,predictive services, and monitor also Dialogue: 0,0:07:18.48,0:07:19.68,Default,,0000,0000,0000,,their usage. Dialogue: 0,0:07:19.68,0:07:22.14,Default,,0000,0000,0000,,So to begin using the web portal, you Dialogue: 0,0:07:22.14,0:07:23.88,Default,,0000,0000,0000,,need to assign the workspace you created Dialogue: 0,0:07:23.88,0:07:26.82,Default,,0000,0000,0000,,in the Azure portal to the Azure Machine Dialogue: 0,0:07:26.82,0:07:29.42,Default,,0000,0000,0000,,Learning studio. Dialogue: 0,0:07:29.52,0:07:31.80,Default,,0000,0000,0000,,At its core, Azure Machine Learning is a Dialogue: 0,0:07:31.80,0:07:33.72,Default,,0000,0000,0000,,service for training and managing Dialogue: 0,0:07:33.72,0:07:36.00,Default,,0000,0000,0000,,machine learning models for which you Dialogue: 0,0:07:36.00,0:07:38.22,Default,,0000,0000,0000,,need compute resources on which to run Dialogue: 0,0:07:38.22,0:07:39.92,Default,,0000,0000,0000,,the training process. Dialogue: 0,0:07:39.92,0:07:44.28,Default,,0000,0000,0000,,Compute targets are, um, one of the main Dialogue: 0,0:07:44.28,0:07:46.74,Default,,0000,0000,0000,,basic concepts of Azure Machine Learning. Dialogue: 0,0:07:46.74,0:07:48.78,Default,,0000,0000,0000,,They are cloud-based resources on which Dialogue: 0,0:07:48.78,0:07:50.64,Default,,0000,0000,0000,,you can run model training and data Dialogue: 0,0:07:50.64,0:07:53.22,Default,,0000,0000,0000,,exploration processes. Dialogue: 0,0:07:53.22,0:07:54.78,Default,,0000,0000,0000,,So in Azure Machine Learning studio, you Dialogue: 0,0:07:54.78,0:07:56.76,Default,,0000,0000,0000,,can manage the compute targets for your Dialogue: 0,0:07:56.76,0:07:58.74,Default,,0000,0000,0000,,data science activities, and there are Dialogue: 0,0:07:58.74,0:08:03.24,Default,,0000,0000,0000,,four kinds of of compute targets you can Dialogue: 0,0:08:03.24,0:08:05.94,Default,,0000,0000,0000,,create. We have the compute instances, Dialogue: 0,0:08:05.94,0:08:09.54,Default,,0000,0000,0000,,which are vital machine set up for Dialogue: 0,0:08:09.54,0:08:10.98,Default,,0000,0000,0000,,running machine learning code during Dialogue: 0,0:08:10.98,0:08:13.32,Default,,0000,0000,0000,,development, so they are not designed for Dialogue: 0,0:08:13.32,0:08:14.46,Default,,0000,0000,0000,,production. Dialogue: 0,0:08:14.46,0:08:17.10,Default,,0000,0000,0000,,Then we have compute clusters, which are Dialogue: 0,0:08:17.10,0:08:19.80,Default,,0000,0000,0000,,a set of virtual machines that can scale Dialogue: 0,0:08:19.80,0:08:22.20,Default,,0000,0000,0000,,up automatically based on traffic. Dialogue: 0,0:08:22.20,0:08:24.60,Default,,0000,0000,0000,,We have inference clusters, which are Dialogue: 0,0:08:24.60,0:08:26.70,Default,,0000,0000,0000,,similar to compute clusters, but they are Dialogue: 0,0:08:26.70,0:08:29.34,Default,,0000,0000,0000,,designed for deployment, so they are Dialogue: 0,0:08:29.34,0:08:31.98,Default,,0000,0000,0000,,deployment targets for predictive Dialogue: 0,0:08:31.98,0:08:35.82,Default,,0000,0000,0000,,services that use trained models. Dialogue: 0,0:08:35.82,0:08:38.34,Default,,0000,0000,0000,,And finally, we have attached compute, Dialogue: 0,0:08:38.34,0:08:41.34,Default,,0000,0000,0000,,which are any compute target that you Dialogue: 0,0:08:41.34,0:08:44.16,Default,,0000,0000,0000,,manage yourself outside of Azure ML, like, Dialogue: 0,0:08:44.16,0:08:46.56,Default,,0000,0000,0000,,for example, virtual machines or Azure Dialogue: 0,0:08:46.56,0:08:49.70,Default,,0000,0000,0000,,data bricks clusters. Dialogue: 0,0:08:49.98,0:08:52.80,Default,,0000,0000,0000,,So we talked about Azure Machine Dialogue: 0,0:08:52.80,0:08:54.30,Default,,0000,0000,0000,,Learning, but we also mentioned- Dialogue: 0,0:08:54.30,0:08:55.50,Default,,0000,0000,0000,,mentioned Azure Machine Learning Dialogue: 0,0:08:55.50,0:08:57.54,Default,,0000,0000,0000,,designer. What is Azure Machine Learning Dialogue: 0,0:08:57.54,0:09:00.12,Default,,0000,0000,0000,,designer? So, in Azure Machine Learning Dialogue: 0,0:09:00.12,0:09:02.88,Default,,0000,0000,0000,,Studio, there are several ways to author Dialogue: 0,0:09:02.88,0:09:04.56,Default,,0000,0000,0000,,classification machine learning models. Dialogue: 0,0:09:04.56,0:09:08.10,Default,,0000,0000,0000,,One way is to use a visual interface, and Dialogue: 0,0:09:08.10,0:09:10.26,Default,,0000,0000,0000,,this visual interface is called designer, Dialogue: 0,0:09:10.26,0:09:13.14,Default,,0000,0000,0000,,and you can use it to train, test, and Dialogue: 0,0:09:13.14,0:09:15.54,Default,,0000,0000,0000,,also deploy machine learning models. And Dialogue: 0,0:09:15.54,0:09:17.94,Default,,0000,0000,0000,,the drag-and-drop interface makes use of Dialogue: 0,0:09:17.94,0:09:20.28,Default,,0000,0000,0000,,clearly defined inputs and outputs that Dialogue: 0,0:09:20.28,0:09:22.68,Default,,0000,0000,0000,,can be shared, reused, and also version Dialogue: 0,0:09:22.68,0:09:23.88,Default,,0000,0000,0000,,control. Dialogue: 0,0:09:23.88,0:09:25.92,Default,,0000,0000,0000,,And using the designer, you can identify Dialogue: 0,0:09:25.92,0:09:28.08,Default,,0000,0000,0000,,the building blocks or components needed Dialogue: 0,0:09:28.08,0:09:30.84,Default,,0000,0000,0000,,for your model, place and connect them on Dialogue: 0,0:09:30.84,0:09:33.12,Default,,0000,0000,0000,,your canvas, and run a machine learning Dialogue: 0,0:09:33.12,0:09:35.30,Default,,0000,0000,0000,,job. Dialogue: 0,0:09:35.40,0:09:36.78,Default,,0000,0000,0000,,So, Dialogue: 0,0:09:36.78,0:09:39.12,Default,,0000,0000,0000,,each designer project, so each project Dialogue: 0,0:09:39.12,0:09:42.36,Default,,0000,0000,0000,,in the designer is known as a pipeline. Dialogue: 0,0:09:42.36,0:09:45.60,Default,,0000,0000,0000,,And in the design, we have a left panel Dialogue: 0,0:09:45.60,0:09:48.36,Default,,0000,0000,0000,,for navigation and a canvas on your Dialogue: 0,0:09:48.36,0:09:50.64,Default,,0000,0000,0000,,right hand side in which you build your Dialogue: 0,0:09:50.64,0:09:53.94,Default,,0000,0000,0000,,pipeline visually. So pipelines let you Dialogue: 0,0:09:53.94,0:09:56.10,Default,,0000,0000,0000,,organize, manage, and reuse complex Dialogue: 0,0:09:56.10,0:09:58.26,Default,,0000,0000,0000,,machine learning workflows across Dialogue: 0,0:09:58.26,0:10:00.48,Default,,0000,0000,0000,,projects and users. Dialogue: 0,0:10:00.48,0:10:03.00,Default,,0000,0000,0000,,A pipeline starts with the data set from Dialogue: 0,0:10:03.00,0:10:04.14,Default,,0000,0000,0000,,which you want to train the model Dialogue: 0,0:10:04.14,0:10:05.88,Default,,0000,0000,0000,,because all begins with data when Dialogue: 0,0:10:05.88,0:10:07.38,Default,,0000,0000,0000,,talking about data science and machine Dialogue: 0,0:10:07.38,0:10:09.54,Default,,0000,0000,0000,,learning. And each time you run a Dialogue: 0,0:10:09.54,0:10:10.98,Default,,0000,0000,0000,,pipeline, the configuration of the Dialogue: 0,0:10:10.98,0:10:12.96,Default,,0000,0000,0000,,pipeline and its results are stored in Dialogue: 0,0:10:12.96,0:10:17.34,Default,,0000,0000,0000,,your workspace as a pipeline job. Dialogue: 0,0:10:17.34,0:10:21.96,Default,,0000,0000,0000,,So the second main concept of Azure Dialogue: 0,0:10:21.96,0:10:25.08,Default,,0000,0000,0000,,Machine Learning is a component. So, going Dialogue: 0,0:10:25.08,0:10:28.44,Default,,0000,0000,0000,,hierarchically from the pipeline, we can Dialogue: 0,0:10:28.44,0:10:30.54,Default,,0000,0000,0000,,say that each building block of a Dialogue: 0,0:10:30.54,0:10:32.92,Default,,0000,0000,0000,,pipeline is called a component. Dialogue: 0,0:10:32.92,0:10:34.12,Default,,0000,0000,0000,,In other words, an Azure Machine Dialogue: 0,0:10:34.12,0:10:36.96,Default,,0000,0000,0000,,Learning component encapsulates one step Dialogue: 0,0:10:36.96,0:10:39.42,Default,,0000,0000,0000,,in a machine learning pipeline. So, it's a Dialogue: 0,0:10:39.42,0:10:41.64,Default,,0000,0000,0000,,reusable piece of code with inputs and Dialogue: 0,0:10:41.64,0:10:44.10,Default,,0000,0000,0000,,outputs, something very similar to a Dialogue: 0,0:10:44.10,0:10:46.50,Default,,0000,0000,0000,,function in any programming language. Dialogue: 0,0:10:46.50,0:10:48.90,Default,,0000,0000,0000,,And in a pipeline project, you can access Dialogue: 0,0:10:48.90,0:10:51.48,Default,,0000,0000,0000,,data assets and components from the left Dialogue: 0,0:10:51.48,0:10:52.70,Default,,0000,0000,0000,,panels Dialogue: 0,0:10:52.70,0:10:56.28,Default,,0000,0000,0000,,Asset Library tab, as you can see Dialogue: 0,0:10:56.28,0:11:00.20,Default,,0000,0000,0000,,here in the screenshot in the deck. Dialogue: 0,0:11:00.30,0:11:03.36,Default,,0000,0000,0000,,So you can create data assets on using Dialogue: 0,0:11:03.36,0:11:08.34,Default,,0000,0000,0000,,an ADOC page called Data Page. And a data Dialogue: 0,0:11:08.34,0:11:11.16,Default,,0000,0000,0000,,asset is a reference to a data source Dialogue: 0,0:11:11.16,0:11:12.48,Default,,0000,0000,0000,,location. Dialogue: 0,0:11:12.48,0:11:15.72,Default,,0000,0000,0000,,So this data source location could be a Dialogue: 0,0:11:15.72,0:11:18.78,Default,,0000,0000,0000,,local file, a data store, a web file or Dialogue: 0,0:11:18.78,0:11:21.66,Default,,0000,0000,0000,,even an Azure open asset. Dialogue: 0,0:11:21.66,0:11:23.88,Default,,0000,0000,0000,,And these data assets will appear along Dialogue: 0,0:11:23.88,0:11:26.46,Default,,0000,0000,0000,,with standard sample data set in the Dialogue: 0,0:11:26.46,0:11:30.02,Default,,0000,0000,0000,,designers Asset Library. Dialogue: 0,0:11:30.08,0:11:31.56,Default,,0000,0000,0000,,Um. Dialogue: 0,0:11:31.56,0:11:36.96,Default,,0000,0000,0000,,Another basic concept of Azure ML is Dialogue: 0,0:11:36.96,0:11:38.88,Default,,0000,0000,0000,,Azure Machine Learning jobs. Dialogue: 0,0:11:38.88,0:11:43.52,Default,,0000,0000,0000,,So, basically, when you submit a pipeline, Dialogue: 0,0:11:43.52,0:11:47.04,Default,,0000,0000,0000,,you create a job which will run all the Dialogue: 0,0:11:47.04,0:11:49.92,Default,,0000,0000,0000,,steps in your pipeline. So a job executes Dialogue: 0,0:11:49.92,0:11:52.80,Default,,0000,0000,0000,,a task against a specified compute Dialogue: 0,0:11:52.80,0:11:53.76,Default,,0000,0000,0000,,target. Dialogue: 0,0:11:53.76,0:11:56.64,Default,,0000,0000,0000,,Jobs enable systematic tracking for your Dialogue: 0,0:11:56.64,0:11:58.56,Default,,0000,0000,0000,,machine learning experimentation in Dialogue: 0,0:11:58.56,0:11:59.88,Default,,0000,0000,0000,,Azure ML. Dialogue: 0,0:11:59.88,0:12:02.40,Default,,0000,0000,0000,,And once a job is created, Azure ML Dialogue: 0,0:12:02.40,0:12:05.46,Default,,0000,0000,0000,,maintains a run record, uh, for the Dialogue: 0,0:12:05.46,0:12:07.64,Default,,0000,0000,0000,,job. Dialogue: 0,0:12:07.88,0:12:12.18,Default,,0000,0000,0000,,Um, but, let's move to the classification Dialogue: 0,0:12:12.18,0:12:14.04,Default,,0000,0000,0000,,steps. So, Dialogue: 0,0:12:14.04,0:12:17.16,Default,,0000,0000,0000,,um, let's introduce how to create a Dialogue: 0,0:12:17.16,0:12:21.36,Default,,0000,0000,0000,,classification model in Azure ML, but you Dialogue: 0,0:12:21.36,0:12:23.64,Default,,0000,0000,0000,,will see it in more details in a Dialogue: 0,0:12:23.64,0:12:26.34,Default,,0000,0000,0000,,handsome demo that John will guide Dialogue: 0,0:12:26.34,0:12:29.46,Default,,0000,0000,0000,,through in a few minutes. Dialogue: 0,0:12:29.46,0:12:32.22,Default,,0000,0000,0000,,So, you can think of the steps to train Dialogue: 0,0:12:32.22,0:12:33.72,Default,,0000,0000,0000,,and evaluate a classification machine Dialogue: 0,0:12:33.72,0:12:36.66,Default,,0000,0000,0000,,learning model as four main steps. So Dialogue: 0,0:12:36.66,0:12:38.46,Default,,0000,0000,0000,,first of all, you need to prepare your Dialogue: 0,0:12:38.46,0:12:41.10,Default,,0000,0000,0000,,data. So, you need to identify the Dialogue: 0,0:12:41.10,0:12:43.14,Default,,0000,0000,0000,,features and the label in your data set, Dialogue: 0,0:12:43.14,0:12:46.14,Default,,0000,0000,0000,,you need to pre-process, so you need to Dialogue: 0,0:12:46.14,0:12:48.84,Default,,0000,0000,0000,,clean and transform the data as needed. Dialogue: 0,0:12:48.84,0:12:51.12,Default,,0000,0000,0000,,Then, the second step, of course, is Dialogue: 0,0:12:51.12,0:12:52.74,Default,,0000,0000,0000,,training the model. Dialogue: 0,0:12:52.74,0:12:54.60,Default,,0000,0000,0000,,And for training the model, you need to Dialogue: 0,0:12:54.60,0:12:57.06,Default,,0000,0000,0000,,split the data into two groups: a Dialogue: 0,0:12:57.06,0:12:59.52,Default,,0000,0000,0000,,training and a validation set. Dialogue: 0,0:12:59.52,0:13:01.32,Default,,0000,0000,0000,,Then you train a machine learning model Dialogue: 0,0:13:01.32,0:13:03.54,Default,,0000,0000,0000,,using the training data set and you test Dialogue: 0,0:13:03.54,0:13:05.04,Default,,0000,0000,0000,,the machine learning model for Dialogue: 0,0:13:05.04,0:13:06.89,Default,,0000,0000,0000,,performance using the validation data Dialogue: 0,0:13:06.89,0:13:08.10,Default,,0000,0000,0000,,set. Dialogue: 0,0:13:08.10,0:13:12.18,Default,,0000,0000,0000,,The third step is performance evaluation, Dialogue: 0,0:13:12.18,0:13:14.52,Default,,0000,0000,0000,,which means comparing how close the Dialogue: 0,0:13:14.52,0:13:16.14,Default,,0000,0000,0000,,model's predictions are to the known Dialogue: 0,0:13:16.14,0:13:20.52,Default,,0000,0000,0000,,labels and these lead us to compute some Dialogue: 0,0:13:20.52,0:13:23.28,Default,,0000,0000,0000,,evaluation performance metrics. Dialogue: 0,0:13:23.28,0:13:25.74,Default,,0000,0000,0000,,And then finally... Dialogue: 0,0:13:25.74,0:13:29.05,Default,,0000,0000,0000,,So, these three steps are not, Dialogue: 0,0:13:29.05,0:13:32.77,Default,,0000,0000,0000,,um, not performed every time in a Dialogue: 0,0:13:32.77,0:13:35.46,Default,,0000,0000,0000,,linear manner. It's more an iterative Dialogue: 0,0:13:35.46,0:13:39.42,Default,,0000,0000,0000,,process. But once you obtain, you achieve Dialogue: 0,0:13:39.42,0:13:42.96,Default,,0000,0000,0000,,a performance with which you are Dialogue: 0,0:13:42.96,0:13:45.78,Default,,0000,0000,0000,,satisfied, so you are ready to, let's say Dialogue: 0,0:13:45.78,0:13:48.66,Default,,0000,0000,0000,,go into production, and you can deploy Dialogue: 0,0:13:48.66,0:13:51.92,Default,,0000,0000,0000,,your train model as a predictive service Dialogue: 0,0:13:51.92,0:13:55.98,Default,,0000,0000,0000,,into a real-time, uh, to a real-time Dialogue: 0,0:13:55.98,0:13:58.02,Default,,0000,0000,0000,,endpoint. And to do so, you need to Dialogue: 0,0:13:58.02,0:14:00.24,Default,,0000,0000,0000,,convert the training pipeline into a Dialogue: 0,0:14:00.24,0:14:02.82,Default,,0000,0000,0000,,real-time inference pipeline, and then Dialogue: 0,0:14:02.82,0:14:04.26,Default,,0000,0000,0000,,you can deploy the model as an Dialogue: 0,0:14:04.26,0:14:06.78,Default,,0000,0000,0000,,application on a server or device so Dialogue: 0,0:14:06.78,0:14:11.42,Default,,0000,0000,0000,,that others can consume this model. Dialogue: 0,0:14:11.46,0:14:14.28,Default,,0000,0000,0000,,So let's start with the first step, which Dialogue: 0,0:14:14.28,0:14:17.70,Default,,0000,0000,0000,,is prepare data. Real-world data can contain Dialogue: 0,0:14:17.70,0:14:19.92,Default,,0000,0000,0000,,many different issues that can affect Dialogue: 0,0:14:19.92,0:14:22.32,Default,,0000,0000,0000,,the utility of the data and our Dialogue: 0,0:14:22.32,0:14:24.96,Default,,0000,0000,0000,,interpretation of the results. So also Dialogue: 0,0:14:24.96,0:14:26.58,Default,,0000,0000,0000,,the machine learning model that you Dialogue: 0,0:14:26.58,0:14:29.28,Default,,0000,0000,0000,,train using this data. For example, real- Dialogue: 0,0:14:29.28,0:14:31.44,Default,,0000,0000,0000,,world data can be affected by a bad Dialogue: 0,0:14:31.44,0:14:34.08,Default,,0000,0000,0000,,recording or a bad measurement, and it Dialogue: 0,0:14:34.08,0:14:36.48,Default,,0000,0000,0000,,can also contain missing values for some Dialogue: 0,0:14:36.48,0:14:38.88,Default,,0000,0000,0000,,parameters. And Azure Machine Learning Dialogue: 0,0:14:38.88,0:14:40.86,Default,,0000,0000,0000,,designer has several pre-built Dialogue: 0,0:14:40.86,0:14:43.02,Default,,0000,0000,0000,,components that can be used to prepare Dialogue: 0,0:14:43.02,0:14:46.08,Default,,0000,0000,0000,,data for training. These components Dialogue: 0,0:14:46.08,0:14:48.30,Default,,0000,0000,0000,,enable you to clean data, normalize Dialogue: 0,0:14:48.30,0:14:52.94,Default,,0000,0000,0000,,features, join tables, and more. Dialogue: 0,0:14:53.00,0:14:57.12,Default,,0000,0000,0000,,Let's come to training. So, to train a Dialogue: 0,0:14:57.12,0:14:59.22,Default,,0000,0000,0000,,classification model you need a data set Dialogue: 0,0:14:59.22,0:15:02.16,Default,,0000,0000,0000,,that includes historical features, so the Dialogue: 0,0:15:02.16,0:15:03.90,Default,,0000,0000,0000,,characteristics of the entity for which Dialogue: 0,0:15:03.90,0:15:06.90,Default,,0000,0000,0000,,one to make a prediction, and known label Dialogue: 0,0:15:06.90,0:15:09.78,Default,,0000,0000,0000,,values. The label is the class indicator Dialogue: 0,0:15:09.78,0:15:11.82,Default,,0000,0000,0000,,we want to train a model to predict. Dialogue: 0,0:15:11.82,0:15:13.92,Default,,0000,0000,0000,,And it's common practice to train a Dialogue: 0,0:15:13.92,0:15:16.20,Default,,0000,0000,0000,,model using a subset of the data while Dialogue: 0,0:15:16.20,0:15:18.30,Default,,0000,0000,0000,,holding back some data with which to Dialogue: 0,0:15:18.30,0:15:20.76,Default,,0000,0000,0000,,test the train model. And this enables Dialogue: 0,0:15:20.76,0:15:22.44,Default,,0000,0000,0000,,you to compare the labels that the model Dialogue: 0,0:15:22.44,0:15:25.38,Default,,0000,0000,0000,,predicts with the actual known labels in Dialogue: 0,0:15:25.38,0:15:27.42,Default,,0000,0000,0000,,the original data set. Dialogue: 0,0:15:27.42,0:15:29.88,Default,,0000,0000,0000,,This operation can be performed in the Dialogue: 0,0:15:29.88,0:15:32.10,Default,,0000,0000,0000,,designer using the split data component Dialogue: 0,0:15:32.10,0:15:34.74,Default,,0000,0000,0000,,as shown by the screenshot here in the... Dialogue: 0,0:15:34.74,0:15:36.66,Default,,0000,0000,0000,,in the deck. Dialogue: 0,0:15:36.66,0:15:39.54,Default,,0000,0000,0000,,There's also another component that you Dialogue: 0,0:15:39.54,0:15:40.98,Default,,0000,0000,0000,,should use, which is the score model Dialogue: 0,0:15:40.98,0:15:43.14,Default,,0000,0000,0000,,component to generate the predicted Dialogue: 0,0:15:43.14,0:15:45.36,Default,,0000,0000,0000,,class label value using the validation Dialogue: 0,0:15:45.36,0:15:48.06,Default,,0000,0000,0000,,data as input. So once you connect all Dialogue: 0,0:15:48.06,0:15:49.80,Default,,0000,0000,0000,,these components, Dialogue: 0,0:15:49.80,0:15:52.44,Default,,0000,0000,0000,,the component specifying the Dialogue: 0,0:15:52.44,0:15:54.96,Default,,0000,0000,0000,,model we are going to use, the split data Dialogue: 0,0:15:54.96,0:15:57.06,Default,,0000,0000,0000,,component, the trained model component, Dialogue: 0,0:15:57.06,0:16:00.30,Default,,0000,0000,0000,,and the score model component, you want Dialogue: 0,0:16:00.30,0:16:02.64,Default,,0000,0000,0000,,to run a new experiment in Dialogue: 0,0:16:02.64,0:16:05.76,Default,,0000,0000,0000,,Azure ML, which will use the data set Dialogue: 0,0:16:05.76,0:16:09.60,Default,,0000,0000,0000,,on the canvas to train and score a model. Dialogue: 0,0:16:09.60,0:16:12.00,Default,,0000,0000,0000,,After training a model, it is important, Dialogue: 0,0:16:12.00,0:16:14.64,Default,,0000,0000,0000,,we say, to evaluate its performance, to Dialogue: 0,0:16:14.64,0:16:17.06,Default,,0000,0000,0000,,understand how bad-how good sorry Dialogue: 0,0:16:17.06,0:16:20.76,Default,,0000,0000,0000,,our model is performing. Dialogue: 0,0:16:20.76,0:16:22.68,Default,,0000,0000,0000,,And there are many performance metrics Dialogue: 0,0:16:22.68,0:16:24.60,Default,,0000,0000,0000,,and methodologies for evaluating how Dialogue: 0,0:16:24.60,0:16:27.00,Default,,0000,0000,0000,,well a model makes predictions. The Dialogue: 0,0:16:27.00,0:16:29.16,Default,,0000,0000,0000,,component to use to perform evaluation Dialogue: 0,0:16:29.16,0:16:32.22,Default,,0000,0000,0000,,in Azure ML designer is called, as Dialogue: 0,0:16:32.22,0:16:35.06,Default,,0000,0000,0000,,intuitive as it is, Evaluate Model. Dialogue: 0,0:16:35.06,0:16:38.34,Default,,0000,0000,0000,,Once the job of training and evaluation Dialogue: 0,0:16:38.34,0:16:40.74,Default,,0000,0000,0000,,of the model is completed, you can review Dialogue: 0,0:16:40.74,0:16:42.96,Default,,0000,0000,0000,,evaluation metrics on the completed job Dialogue: 0,0:16:42.96,0:16:45.86,Default,,0000,0000,0000,,page by right clicking on the component. Dialogue: 0,0:16:45.86,0:16:48.48,Default,,0000,0000,0000,,In the evaluation results, you can also Dialogue: 0,0:16:48.48,0:16:51.00,Default,,0000,0000,0000,,find the so-called confusion Matrix that Dialogue: 0,0:16:51.00,0:16:53.40,Default,,0000,0000,0000,,you can see here in the right side of Dialogue: 0,0:16:53.40,0:16:55.08,Default,,0000,0000,0000,,this deck Dialogue: 0,0:16:55.08,0:16:57.42,Default,,0000,0000,0000,,A confusion matrix shows cases where Dialogue: 0,0:16:57.42,0:16:59.22,Default,,0000,0000,0000,,both the predicted and actual values Dialogue: 0,0:16:59.22,0:17:01.98,Default,,0000,0000,0000,,were one, the so-called true positives Dialogue: 0,0:17:01.98,0:17:04.50,Default,,0000,0000,0000,,at the top left and also cases where Dialogue: 0,0:17:04.50,0:17:06.60,Default,,0000,0000,0000,,both the predicted and the actual values Dialogue: 0,0:17:06.60,0:17:08.46,Default,,0000,0000,0000,,were zero, the so-called true negatives Dialogue: 0,0:17:08.46,0:17:10.92,Default,,0000,0000,0000,,at the bottom right. While the other Dialogue: 0,0:17:10.92,0:17:13.68,Default,,0000,0000,0000,,cells show cases where the predicting Dialogue: 0,0:17:13.68,0:17:15.38,Default,,0000,0000,0000,,and actual values differ, Dialogue: 0,0:17:15.38,0:17:17.94,Default,,0000,0000,0000,,called false positive and false Dialogue: 0,0:17:17.94,0:17:19.92,Default,,0000,0000,0000,,negatives, and this is an example of a Dialogue: 0,0:17:19.92,0:17:23.58,Default,,0000,0000,0000,,confusion matrix for a binary classifier. Dialogue: 0,0:17:23.58,0:17:25.56,Default,,0000,0000,0000,,While for a multi-class classification Dialogue: 0,0:17:25.56,0:17:28.08,Default,,0000,0000,0000,,model the same approach is used to Dialogue: 0,0:17:28.08,0:17:30.12,Default,,0000,0000,0000,,tabulate each possible combination of Dialogue: 0,0:17:30.12,0:17:32.94,Default,,0000,0000,0000,,actual and predictive value counts. So Dialogue: 0,0:17:32.94,0:17:34.74,Default,,0000,0000,0000,,for example, a model with three possible Dialogue: 0,0:17:34.74,0:17:37.56,Default,,0000,0000,0000,,classes would result in three times Dialogue: 0,0:17:37.56,0:17:39.12,Default,,0000,0000,0000,,three matrix. Dialogue: 0,0:17:39.12,0:17:41.88,Default,,0000,0000,0000,,The confusion matrix is also useful for Dialogue: 0,0:17:41.88,0:17:43.86,Default,,0000,0000,0000,,the matrix that can be derived from it, Dialogue: 0,0:17:43.86,0:17:48.26,Default,,0000,0000,0000,,like accuracy, recall, or precision. Dialogue: 0,0:17:49.32,0:17:52.08,Default,,0000,0000,0000,,We say that the last step is Dialogue: 0,0:17:52.08,0:17:55.62,Default,,0000,0000,0000,,deploying the train model to a real-time Dialogue: 0,0:17:55.62,0:17:59.28,Default,,0000,0000,0000,,endpoint as a predictive service. And in Dialogue: 0,0:17:59.28,0:18:00.90,Default,,0000,0000,0000,,order to automate your model into a Dialogue: 0,0:18:00.90,0:18:02.76,Default,,0000,0000,0000,,service that makes continuous Dialogue: 0,0:18:02.76,0:18:04.98,Default,,0000,0000,0000,,predictions, you need, first of all, to Dialogue: 0,0:18:04.98,0:18:08.04,Default,,0000,0000,0000,,create and then deploy an Dialogue: 0,0:18:08.04,0:18:10.08,Default,,0000,0000,0000,,inference pipeline. The process of Dialogue: 0,0:18:10.08,0:18:11.94,Default,,0000,0000,0000,,converting the training pipeline into a Dialogue: 0,0:18:11.94,0:18:13.98,Default,,0000,0000,0000,,real-time inference pipeline removes Dialogue: 0,0:18:13.98,0:18:16.26,Default,,0000,0000,0000,,training components and adds web service Dialogue: 0,0:18:16.26,0:18:18.96,Default,,0000,0000,0000,,inputs and outputs to handle requests. Dialogue: 0,0:18:18.96,0:18:21.24,Default,,0000,0000,0000,,And the inference pipeline performs...they Dialogue: 0,0:18:21.24,0:18:22.68,Default,,0000,0000,0000,,seem that the transformation is the Dialogue: 0,0:18:22.68,0:18:26.16,Default,,0000,0000,0000,,first pipeline, but for new data. Then it Dialogue: 0,0:18:26.16,0:18:28.68,Default,,0000,0000,0000,,uses the train model to infer or predict Dialogue: 0,0:18:28.68,0:18:32.54,Default,,0000,0000,0000,,label values based on its feature. Dialogue: 0,0:18:32.82,0:18:36.12,Default,,0000,0000,0000,,So, I think I've talked a lot for now Dialogue: 0,0:18:36.12,0:18:40.38,Default,,0000,0000,0000,,I would like to let John show us Dialogue: 0,0:18:40.38,0:18:44.34,Default,,0000,0000,0000,,something in practice with Dialogue: 0,0:18:44.34,0:18:47.28,Default,,0000,0000,0000,,the hands-on demo, so please, John, go Dialogue: 0,0:18:47.28,0:18:49.86,Default,,0000,0000,0000,,ahead, share your screen and guide us Dialogue: 0,0:18:49.86,0:18:52.38,Default,,0000,0000,0000,,through this demo of creating a Dialogue: 0,0:18:52.38,0:18:53.76,Default,,0000,0000,0000,,classification with the Azure Machine Dialogue: 0,0:18:53.76,0:18:55.86,Default,,0000,0000,0000,,Learning designer. Dialogue: 0,0:18:55.86,0:18:58.92,Default,,0000,0000,0000,,Thank you so much Carlotta for this Dialogue: 0,0:18:58.92,0:19:01.38,Default,,0000,0000,0000,,interesting explanation of the Azure ML Dialogue: 0,0:19:01.38,0:19:03.81,Default,,0000,0000,0000,,designer. And now, Dialogue: 0,0:19:03.81,0:19:07.50,Default,,0000,0000,0000,,um, I'm going to start with you in the Dialogue: 0,0:19:07.50,0:19:10.20,Default,,0000,0000,0000,,practical demo part, so if you want to Dialogue: 0,0:19:10.20,0:19:13.32,Default,,0000,0000,0000,,follow along, go to the link that Carlotta Dialogue: 0,0:19:13.32,0:19:18.38,Default,,0000,0000,0000,,sent in the chat so you can do Dialogue: 0,0:19:18.38,0:19:21.84,Default,,0000,0000,0000,,the demo or the practical part with me. Dialogue: 0,0:19:21.84,0:19:25.26,Default,,0000,0000,0000,,I'm just going to share my screen... Dialogue: 0,0:19:25.26,0:19:27.14,Default,,0000,0000,0000,,and... Dialogue: 0,0:19:27.14,0:19:31.56,Default,,0000,0000,0000,,...go here. So, uh... Dialogue: 0,0:19:31.56,0:19:34.32,Default,,0000,0000,0000,,Where am I right now? I'm inside the Dialogue: 0,0:19:34.32,0:19:36.96,Default,,0000,0000,0000,,Microsoft Learn documentation. This is Dialogue: 0,0:19:36.96,0:19:40.26,Default,,0000,0000,0000,,the exercise part of this module, and we Dialogue: 0,0:19:40.26,0:19:43.08,Default,,0000,0000,0000,,will start by setting two things, which Dialogue: 0,0:19:43.08,0:19:45.30,Default,,0000,0000,0000,,are a prequisite for us to work inside Dialogue: 0,0:19:45.30,0:19:49.92,Default,,0000,0000,0000,,this module, which are the users group Dialogue: 0,0:19:49.92,0:19:52.40,Default,,0000,0000,0000,,and the Azure Machine Learning workspace, Dialogue: 0,0:19:52.40,0:19:55.62,Default,,0000,0000,0000,,and something extra which is the compute Dialogue: 0,0:19:55.62,0:19:59.76,Default,,0000,0000,0000,,cluster that Carlotta talked about. So I Dialogue: 0,0:19:59.76,0:20:02.10,Default,,0000,0000,0000,,just want to make sure that you all have Dialogue: 0,0:20:02.10,0:20:05.66,Default,,0000,0000,0000,,a resource group created inside your Dialogue: 0,0:20:05.66,0:20:08.04,Default,,0000,0000,0000,,portal inside your Microsoft Azure Dialogue: 0,0:20:08.04,0:20:11.10,Default,,0000,0000,0000,,platform. So this is my resource group. Dialogue: 0,0:20:11.10,0:20:14.64,Default,,0000,0000,0000,,Inside this is this Resource Group. I Dialogue: 0,0:20:14.64,0:20:17.30,Default,,0000,0000,0000,,have created an Azure Machine Learning Dialogue: 0,0:20:17.30,0:20:21.54,Default,,0000,0000,0000,,workspace. So I'm just going to access Dialogue: 0,0:20:21.54,0:20:24.00,Default,,0000,0000,0000,,the workspace that I have created Dialogue: 0,0:20:24.00,0:20:27.00,Default,,0000,0000,0000,,already from this link. I am going to Dialogue: 0,0:20:27.00,0:20:30.24,Default,,0000,0000,0000,,open it, which is the studio web URL, and Dialogue: 0,0:20:30.24,0:20:33.00,Default,,0000,0000,0000,,I will follow the steps. So what is this? Dialogue: 0,0:20:33.00,0:20:35.76,Default,,0000,0000,0000,,This is your machine learning workspace, Dialogue: 0,0:20:35.76,0:20:38.22,Default,,0000,0000,0000,,or machine learning studio. You can do a Dialogue: 0,0:20:38.22,0:20:40.08,Default,,0000,0000,0000,,lot of things here, but we are going to Dialogue: 0,0:20:40.08,0:20:42.42,Default,,0000,0000,0000,,focus mainly on the designer and the Dialogue: 0,0:20:42.42,0:20:46.08,Default,,0000,0000,0000,,data and the compute. So another Dialogue: 0,0:20:46.08,0:20:49.14,Default,,0000,0000,0000,,prerequisite here, as Carlotta told you, Dialogue: 0,0:20:49.14,0:20:51.48,Default,,0000,0000,0000,,we need some resources to power up the Dialogue: 0,0:20:51.48,0:20:54.30,Default,,0000,0000,0000,,classification, the processes that Dialogue: 0,0:20:54.30,0:20:55.14,Default,,0000,0000,0000,,will happen. Dialogue: 0,0:20:55.14,0:20:58.08,Default,,0000,0000,0000,,So, we have created this computing Dialogue: 0,0:20:58.08,0:20:59.10,Default,,0000,0000,0000,,cluster, Dialogue: 0,0:20:59.10,0:21:02.88,Default,,0000,0000,0000,,and we have set some presets for Dialogue: 0,0:21:02.88,0:21:04.14,Default,,0000,0000,0000,,it. So Dialogue: 0,0:21:04.14,0:21:07.08,Default,,0000,0000,0000,,where can you find this preset? You go Dialogue: 0,0:21:07.08,0:21:10.20,Default,,0000,0000,0000,,here. Under the create compute, you'll Dialogue: 0,0:21:10.20,0:21:13.22,Default,,0000,0000,0000,,find everything that you need to do. So Dialogue: 0,0:21:13.22,0:21:16.74,Default,,0000,0000,0000,,the size is the Standard DS11 Version 2, Dialogue: 0,0:21:16.74,0:21:19.80,Default,,0000,0000,0000,,and it's a CPU not GPU, because we don't Dialogue: 0,0:21:19.80,0:21:22.50,Default,,0000,0000,0000,,know the GPU, and we don't need a GPU. Dialogue: 0,0:21:22.50,0:21:25.80,Default,,0000,0000,0000,,Uh, it is ready for us to use. Dialogue: 0,0:21:25.80,0:21:30.90,Default,,0000,0000,0000,,The next thing which we will look into Dialogue: 0,0:21:30.90,0:21:33.60,Default,,0000,0000,0000,,is the designer. How can you access the Dialogue: 0,0:21:33.60,0:21:35.10,Default,,0000,0000,0000,,designer? Dialogue: 0,0:21:35.10,0:21:37.68,Default,,0000,0000,0000,,You can either click on this icon or Dialogue: 0,0:21:37.68,0:21:40.02,Default,,0000,0000,0000,,click on the navigation menu and click Dialogue: 0,0:21:40.02,0:21:42.30,Default,,0000,0000,0000,,on the designer for me. Dialogue: 0,0:21:42.90,0:21:45.78,Default,,0000,0000,0000,,Now I am inside my designer. Dialogue: 0,0:21:45.78,0:21:47.64,Default,,0000,0000,0000,,What we are going to do now is the Dialogue: 0,0:21:47.64,0:21:50.28,Default,,0000,0000,0000,,pipeline that Carlotta told you about. Dialogue: 0,0:21:50.28,0:21:54.36,Default,,0000,0000,0000,,And from where can I know these steps? If Dialogue: 0,0:21:54.36,0:21:57.12,Default,,0000,0000,0000,,you follow along in the learn module, you Dialogue: 0,0:21:57.12,0:21:58.74,Default,,0000,0000,0000,,will find everything that I'm doing Dialogue: 0,0:21:58.74,0:22:02.34,Default,,0000,0000,0000,,right now in detail, with screenshots Dialogue: 0,0:22:02.34,0:22:05.82,Default,,0000,0000,0000,,of course. So I'm going to create a new Dialogue: 0,0:22:05.82,0:22:09.12,Default,,0000,0000,0000,,pipeline, and I can do so by clicking on Dialogue: 0,0:22:09.12,0:22:10.98,Default,,0000,0000,0000,,this plus button. Dialogue: 0,0:22:10.98,0:22:13.74,Default,,0000,0000,0000,,It's going to redirect me to the Dialogue: 0,0:22:13.74,0:22:17.10,Default,,0000,0000,0000,,designer authoring the pipeline, uh, where Dialogue: 0,0:22:17.10,0:22:19.50,Default,,0000,0000,0000,,I can drag and drop data and components Dialogue: 0,0:22:19.50,0:22:21.78,Default,,0000,0000,0000,,that Carlotta told you the difference Dialogue: 0,0:22:21.78,0:22:22.98,Default,,0000,0000,0000,,between. Dialogue: 0,0:22:22.98,0:22:26.34,Default,,0000,0000,0000,,And here I am going to do some changes Dialogue: 0,0:22:26.34,0:22:29.10,Default,,0000,0000,0000,,to the settings. I am going to connect Dialogue: 0,0:22:29.10,0:22:31.86,Default,,0000,0000,0000,,this with my compute cluster that I Dialogue: 0,0:22:31.86,0:22:35.12,Default,,0000,0000,0000,,created previously so I can utilize it. Dialogue: 0,0:22:35.12,0:22:38.10,Default,,0000,0000,0000,,From here I'm going to choose this Dialogue: 0,0:22:38.10,0:22:40.38,Default,,0000,0000,0000,,compute cluster demo that I have showed Dialogue: 0,0:22:40.38,0:22:42.60,Default,,0000,0000,0000,,you before in the clusters here, Dialogue: 0,0:22:42.60,0:22:45.90,Default,,0000,0000,0000,,and I am going to change the name to Dialogue: 0,0:22:45.90,0:22:47.82,Default,,0000,0000,0000,,something more meaningful. Instead of Dialogue: 0,0:22:47.82,0:22:50.58,Default,,0000,0000,0000,,byline and the date of today I'm going Dialogue: 0,0:22:50.58,0:22:53.76,Default,,0000,0000,0000,,to name it Diabetes... Dialogue: 0,0:22:53.76,0:22:56.12,Default,,0000,0000,0000,,uh... Dialogue: 0,0:22:56.12,0:23:00.02,Default,,0000,0000,0000,,let's just check this training. Dialogue: 0,0:23:00.02,0:23:05.10,Default,,0000,0000,0000,,Let's say Training 0.1 or 01, okay? Dialogue: 0,0:23:05.10,0:23:09.36,Default,,0000,0000,0000,,And I am going to close this tab in Dialogue: 0,0:23:09.36,0:23:12.00,Default,,0000,0000,0000,,order to have a bigger place to work Dialogue: 0,0:23:12.00,0:23:14.70,Default,,0000,0000,0000,,inside because this is where we will Dialogue: 0,0:23:14.70,0:23:17.22,Default,,0000,0000,0000,,work, where everything will happen. So I Dialogue: 0,0:23:17.22,0:23:19.56,Default,,0000,0000,0000,,will click on close from here, Dialogue: 0,0:23:19.56,0:23:23.46,Default,,0000,0000,0000,,and I will go to the data and I will Dialogue: 0,0:23:23.46,0:23:25.62,Default,,0000,0000,0000,,create a new data set. Dialogue: 0,0:23:25.62,0:23:27.90,Default,,0000,0000,0000,,How can I create a new data set? There is Dialogue: 0,0:23:27.90,0:23:29.88,Default,,0000,0000,0000,,multiple options here you can find, from Dialogue: 0,0:23:29.88,0:23:31.80,Default,,0000,0000,0000,,local files, from data store, from web Dialogue: 0,0:23:31.80,0:23:34.02,Default,,0000,0000,0000,,files, from open data set, but I'm going Dialogue: 0,0:23:34.02,0:23:36.54,Default,,0000,0000,0000,,to choose from web files, as this is the Dialogue: 0,0:23:36.54,0:23:40.28,Default,,0000,0000,0000,,way we're going to create our data. Dialogue: 0,0:23:40.28,0:23:43.38,Default,,0000,0000,0000,,From here, the information of my data set Dialogue: 0,0:23:43.38,0:23:47.34,Default,,0000,0000,0000,,I'm going to get them from the Microsoft Dialogue: 0,0:23:47.34,0:23:50.82,Default,,0000,0000,0000,,Learn module. So if we go to the step Dialogue: 0,0:23:50.82,0:23:52.86,Default,,0000,0000,0000,,that says "Create a dataset", Dialogue: 0,0:23:52.86,0:23:55.02,Default,,0000,0000,0000,,under it, it illustrates that you can Dialogue: 0,0:23:55.02,0:23:57.72,Default,,0000,0000,0000,,access the data from inside the asset Dialogue: 0,0:23:57.72,0:23:59.76,Default,,0000,0000,0000,,library, and inside your asset library, Dialogue: 0,0:23:59.76,0:24:01.68,Default,,0000,0000,0000,,you'll find the data and find the Dialogue: 0,0:24:01.68,0:24:05.54,Default,,0000,0000,0000,,component. And I'm going to select Dialogue: 0,0:24:05.54,0:24:09.00,Default,,0000,0000,0000,,this link because this is where my data Dialogue: 0,0:24:09.00,0:24:12.00,Default,,0000,0000,0000,,is stored. If you open this link, you will Dialogue: 0,0:24:12.00,0:24:14.82,Default,,0000,0000,0000,,find this is a CSV file, I think. Dialogue: 0,0:24:14.82,0:24:17.40,Default,,0000,0000,0000,,Yeah. And you can...like, all the data are Dialogue: 0,0:24:17.40,0:24:18.36,Default,,0000,0000,0000,,here. Dialogue: 0,0:24:18.36,0:24:21.08,Default,,0000,0000,0000,,Now let's get back.. Dialogue: 0,0:24:21.08,0:24:22.15,Default,,0000,0000,0000,,Um... Dialogue: 0,0:24:26.88,0:24:28.20,Default,,0000,0000,0000,,And you are going to do something Dialogue: 0,0:24:28.20,0:24:29.88,Default,,0000,0000,0000,,meaningful, but because I have already Dialogue: 0,0:24:29.88,0:24:31.82,Default,,0000,0000,0000,,created it before twice, so I'm gonna Dialogue: 0,0:24:31.82,0:24:34.98,Default,,0000,0000,0000,,add a number to the name Dialogue: 0,0:24:34.98,0:24:37.56,Default,,0000,0000,0000,,The data set is tabular and there is Dialogue: 0,0:24:37.56,0:24:39.36,Default,,0000,0000,0000,,the file, but this is a table, so we're Dialogue: 0,0:24:39.36,0:24:40.76,Default,,0000,0000,0000,,going to choose the table. Dialogue: 0,0:24:40.76,0:24:42.24,Default,,0000,0000,0000,,Data type Dialogue: 0,0:24:42.24,0:24:43.74,Default,,0000,0000,0000,,for data set type. Dialogue: 0,0:24:43.74,0:24:46.26,Default,,0000,0000,0000,,Now we will click on "Next". That's gonna Dialogue: 0,0:24:46.26,0:24:51.18,Default,,0000,0000,0000,,review, or display for you the content Dialogue: 0,0:24:51.18,0:24:54.02,Default,,0000,0000,0000,,of this file that you have Dialogue: 0,0:24:54.02,0:24:57.42,Default,,0000,0000,0000,,imported to this workspace. Dialogue: 0,0:24:57.42,0:25:01.56,Default,,0000,0000,0000,,And for these settings, these are Dialogue: 0,0:25:01.56,0:25:03.72,Default,,0000,0000,0000,,related to our file format. Dialogue: 0,0:25:03.72,0:25:08.28,Default,,0000,0000,0000,,So this is a delimited file, and it's not Dialogue: 0,0:25:08.28,0:25:11.40,Default,,0000,0000,0000,,plain text, it's not a Jason. The delimiter Dialogue: 0,0:25:11.40,0:25:14.16,Default,,0000,0000,0000,,is common, as we have seen that they Dialogue: 0,0:25:14.16,0:25:26.70,Default,,0000,0000,0000,,[INDISTINGUISHABLE] Dialogue: 0,0:25:26.70,0:25:29.04,Default,,0000,0000,0000,,So I'm choosing common Dialogue: 0,0:25:29.04,0:25:32.90,Default,,0000,0000,0000,,errors because the only the first five... Dialogue: 0,0:25:32.90,0:25:34.88,Default,,0000,0000,0000,,[INDISTINGUISHABLE] Dialogue: 0,0:25:34.88,0:25:38.16,Default,,0000,0000,0000,,...for example. Okay, uh, if you have any Dialogue: 0,0:25:38.16,0:25:39.96,Default,,0000,0000,0000,,doubts, if you have any problems, please Dialogue: 0,0:25:39.96,0:25:42.96,Default,,0000,0000,0000,,don't hesitate to write me Dialogue: 0,0:25:42.96,0:25:45.02,Default,,0000,0000,0000,,in the chat, Dialogue: 0,0:25:45.02,0:25:48.48,Default,,0000,0000,0000,,like, what is blocking you, and Dialogue: 0,0:25:48.48,0:25:50.94,Default,,0000,0000,0000,,me and Carlotta will try to help you, Dialogue: 0,0:25:50.94,0:25:53.22,Default,,0000,0000,0000,,like whenever possible. Dialogue: 0,0:25:53.22,0:25:55.66,Default,,0000,0000,0000,,And now this is the new preview for my Dialogue: 0,0:25:55.66,0:25:57.84,Default,,0000,0000,0000,,data set. I can see that I have an ID, I Dialogue: 0,0:25:57.84,0:25:59.70,Default,,0000,0000,0000,,have patient ID, I have pregnancies, I Dialogue: 0,0:25:59.70,0:26:02.22,Default,,0000,0000,0000,,have the age of the people, Dialogue: 0,0:26:02.22,0:26:05.72,Default,,0000,0000,0000,,I have the body mass, I think Dialogue: 0,0:26:05.72,0:26:08.46,Default,,0000,0000,0000,,whether they have diabetes or not, as a Dialogue: 0,0:26:08.46,0:26:10.68,Default,,0000,0000,0000,,zero and one. Zero indicates a negative, Dialogue: 0,0:26:10.68,0:26:14.16,Default,,0000,0000,0000,,the person doesn't have diabetes, and one Dialogue: 0,0:26:14.16,0:26:16.08,Default,,0000,0000,0000,,indicates a positive, that this person Dialogue: 0,0:26:16.08,0:26:18.30,Default,,0000,0000,0000,,has diabetes. Okay. Dialogue: 0,0:26:18.30,0:26:20.52,Default,,0000,0000,0000,,Now I'm going to click on "Next". Here I am Dialogue: 0,0:26:20.52,0:26:23.40,Default,,0000,0000,0000,,defining my schema. All the data types Dialogue: 0,0:26:23.40,0:26:25.38,Default,,0000,0000,0000,,inside my columns, the column names, which Dialogue: 0,0:26:25.38,0:26:28.76,Default,,0000,0000,0000,,columns to include, which to exclude. And Dialogue: 0,0:26:28.76,0:26:31.50,Default,,0000,0000,0000,,here we will include everything except Dialogue: 0,0:26:31.50,0:26:35.58,Default,,0000,0000,0000,,the path of the bath color. And we are Dialogue: 0,0:26:35.58,0:26:37.86,Default,,0000,0000,0000,,going to review the data types of each Dialogue: 0,0:26:37.86,0:26:40.44,Default,,0000,0000,0000,,column. So let's review this first one. Dialogue: 0,0:26:40.44,0:26:43.32,Default,,0000,0000,0000,,This is numbers, numbers, numbers, then it's the Dialogue: 0,0:26:43.32,0:26:45.78,Default,,0000,0000,0000,,integer. And this is, Dialogue: 0,0:26:45.78,0:26:48.68,Default,,0000,0000,0000,,um, like decimal.. Dialogue: 0,0:26:48.68,0:26:50.90,Default,,0000,0000,0000,,...dotted... Dialogue: 0,0:26:50.90,0:26:53.58,Default,,0000,0000,0000,,decimal number. So we are going to choose Dialogue: 0,0:26:53.58,0:26:55.02,Default,,0000,0000,0000,,this data type. Dialogue: 0,0:26:55.02,0:26:57.20,Default,,0000,0000,0000,,And for this one Dialogue: 0,0:26:57.20,0:27:01.20,Default,,0000,0000,0000,,it says diabetic, and it's a zero under Dialogue: 0,0:27:01.20,0:27:02.46,Default,,0000,0000,0000,,one, and we are going to make it as Dialogue: 0,0:27:02.46,0:27:04.46,Default,,0000,0000,0000,,integers. Dialogue: 0,0:27:04.46,0:27:07.98,Default,,0000,0000,0000,,Now we are going to click on "Next" and Dialogue: 0,0:27:07.98,0:27:09.78,Default,,0000,0000,0000,,move to reviewing everything. This is Dialogue: 0,0:27:09.78,0:27:11.57,Default,,0000,0000,0000,,everything that we have defined together. Dialogue: 0,0:27:11.57,0:27:13.50,Default,,0000,0000,0000,,I will click on "Create". Dialogue: 0,0:27:13.50,0:27:15.18,Default,,0000,0000,0000,,And... Dialogue: 0,0:27:15.18,0:27:17.94,Default,,0000,0000,0000,,now the first step has ended. We have Dialogue: 0,0:27:17.94,0:27:19.92,Default,,0000,0000,0000,,gotten our data ready. Dialogue: 0,0:27:19.92,0:27:22.44,Default,,0000,0000,0000,,Now...what now? We're going to utilize the Dialogue: 0,0:27:22.44,0:27:23.47,Default,,0000,0000,0000,,designer... Dialogue: 0,0:27:23.47,0:27:26.82,Default,,0000,0000,0000,,um...power. We're going to drag and drop Dialogue: 0,0:27:26.82,0:27:29.82,Default,,0000,0000,0000,,our data set to create the pipeline. Dialogue: 0,0:27:29.82,0:27:33.18,Default,,0000,0000,0000,,So I have clicked on it and dragged it Dialogue: 0,0:27:33.18,0:27:35.64,Default,,0000,0000,0000,,to this space. It's gonna appear to you. Dialogue: 0,0:27:35.64,0:27:39.66,Default,,0000,0000,0000,,And we can inspect it by right clicking and Dialogue: 0,0:27:39.66,0:27:42.18,Default,,0000,0000,0000,,choose "Preview data" Dialogue: 0,0:27:42.18,0:27:46.20,Default,,0000,0000,0000,,to see what we have created together. Dialogue: 0,0:27:46.20,0:27:48.90,Default,,0000,0000,0000,,From here, you can see everything that we Dialogue: 0,0:27:48.90,0:27:50.70,Default,,0000,0000,0000,,have seen previously, but in more Dialogue: 0,0:27:50.70,0:27:53.10,Default,,0000,0000,0000,,details. And we are just going to close Dialogue: 0,0:27:53.10,0:27:56.58,Default,,0000,0000,0000,,this. Now what? Now we are gonna do the Dialogue: 0,0:27:56.58,0:28:00.80,Default,,0000,0000,0000,,processing that Carlota mentioned. Dialogue: 0,0:28:00.80,0:28:03.66,Default,,0000,0000,0000,,These are some instructions about the Dialogue: 0,0:28:03.66,0:28:05.46,Default,,0000,0000,0000,,data, about how you can look at them, how you Dialogue: 0,0:28:05.46,0:28:07.14,Default,,0000,0000,0000,,can open them but we are going to move Dialogue: 0,0:28:07.14,0:28:09.72,Default,,0000,0000,0000,,to the transformation or the processing. Dialogue: 0,0:28:09.72,0:28:13.50,Default,,0000,0000,0000,,So as Carlotta told you, like any data Dialogue: 0,0:28:13.50,0:28:15.48,Default,,0000,0000,0000,,for us to work on we have to do some Dialogue: 0,0:28:15.48,0:28:17.30,Default,,0000,0000,0000,,processing to it Dialogue: 0,0:28:17.30,0:28:20.16,Default,,0000,0000,0000,,to make it easy easier for the model to Dialogue: 0,0:28:20.16,0:28:23.28,Default,,0000,0000,0000,,be trained and easier to work with. So, uh, Dialogue: 0,0:28:23.28,0:28:25.86,Default,,0000,0000,0000,,we're gonna do the normalization. And Dialogue: 0,0:28:25.86,0:28:29.16,Default,,0000,0000,0000,,normalization meaning is, uh, Dialogue: 0,0:28:29.16,0:28:33.54,Default,,0000,0000,0000,,to scale our data, either down or up, but Dialogue: 0,0:28:33.54,0:28:35.40,Default,,0000,0000,0000,,we're going to scale them down, Dialogue: 0,0:28:35.40,0:28:38.82,Default,,0000,0000,0000,,and we are going to decrease, uh, Dialogue: 0,0:28:38.82,0:28:40.80,Default,,0000,0000,0000,,relatively decrease Dialogue: 0,0:28:40.80,0:28:44.64,Default,,0000,0000,0000,,the values, all the values, to work Dialogue: 0,0:28:44.64,0:28:48.12,Default,,0000,0000,0000,,with lower numbers. And if we are working Dialogue: 0,0:28:48.12,0:28:49.56,Default,,0000,0000,0000,,with larger numbers, it's going to take Dialogue: 0,0:28:49.56,0:28:52.50,Default,,0000,0000,0000,,more time. If we're working with smaller Dialogue: 0,0:28:52.50,0:28:54.78,Default,,0000,0000,0000,,numbers, it's going to take less time to Dialogue: 0,0:28:54.78,0:28:59.16,Default,,0000,0000,0000,,calculate them, and that's it. So Dialogue: 0,0:28:59.16,0:29:02.16,Default,,0000,0000,0000,,where can I find the normalized data? I Dialogue: 0,0:29:02.16,0:29:04.26,Default,,0000,0000,0000,,can find it inside my component. Dialogue: 0,0:29:04.26,0:29:06.72,Default,,0000,0000,0000,,So I will choose the component and Dialogue: 0,0:29:06.72,0:29:09.66,Default,,0000,0000,0000,,search for "Normalized data". Dialogue: 0,0:29:09.66,0:29:12.36,Default,,0000,0000,0000,,I will drag and drop it as usual and I Dialogue: 0,0:29:12.36,0:29:14.82,Default,,0000,0000,0000,,will connect between these two things Dialogue: 0,0:29:14.82,0:29:18.36,Default,,0000,0000,0000,,by clicking on this spot, this, uh, Dialogue: 0,0:29:18.36,0:29:20.16,Default,,0000,0000,0000,,circuit, and Dialogue: 0,0:29:20.16,0:29:23.16,Default,,0000,0000,0000,,drag and drop onto the next circuit. Dialogue: 0,0:29:23.16,0:29:24.90,Default,,0000,0000,0000,,Now we are going to define our Dialogue: 0,0:29:24.90,0:29:27.42,Default,,0000,0000,0000,,normalization method. Dialogue: 0,0:29:27.42,0:29:31.08,Default,,0000,0000,0000,,So I'm going to double click on the Dialogue: 0,0:29:31.08,0:29:32.64,Default,,0000,0000,0000,,normalized data. Dialogue: 0,0:29:32.64,0:29:34.86,Default,,0000,0000,0000,,It's going to open the settings for the Dialogue: 0,0:29:34.86,0:29:36.48,Default,,0000,0000,0000,,normalization Dialogue: 0,0:29:36.48,0:29:38.82,Default,,0000,0000,0000,,as a better transformation method, which is Dialogue: 0,0:29:38.82,0:29:40.50,Default,,0000,0000,0000,,a mathematical way Dialogue: 0,0:29:40.50,0:29:42.30,Default,,0000,0000,0000,,that is going to scale our data Dialogue: 0,0:29:42.30,0:29:44.52,Default,,0000,0000,0000,,according to. Dialogue: 0,0:29:44.52,0:29:47.76,Default,,0000,0000,0000,,We're going to choose min-max, and for Dialogue: 0,0:29:47.76,0:29:51.54,Default,,0000,0000,0000,,this one, we are going to choose "Use Zero", Dialogue: 0,0:29:51.54,0:29:53.10,Default,,0000,0000,0000,,for constant column we are going to Dialogue: 0,0:29:53.10,0:29:54.48,Default,,0000,0000,0000,,choose "True", Dialogue: 0,0:29:54.48,0:29:56.88,Default,,0000,0000,0000,,and we are going to define which columns Dialogue: 0,0:29:56.88,0:29:58.86,Default,,0000,0000,0000,,to normalize. So we are not going to Dialogue: 0,0:29:58.86,0:30:01.08,Default,,0000,0000,0000,,normalize the whole data set. We are Dialogue: 0,0:30:01.08,0:30:02.76,Default,,0000,0000,0000,,going to choose a subset from the data Dialogue: 0,0:30:02.76,0:30:04.56,Default,,0000,0000,0000,,set to normalize. So we're going to Dialogue: 0,0:30:04.56,0:30:07.02,Default,,0000,0000,0000,,choose everything except for the patient Dialogue: 0,0:30:07.02,0:30:09.00,Default,,0000,0000,0000,,ID and the diabetic, because the patient Dialogue: 0,0:30:09.00,0:30:10.92,Default,,0000,0000,0000,,ID is a number, but it's a categorical Dialogue: 0,0:30:10.92,0:30:13.74,Default,,0000,0000,0000,,data. It describes a patient, it's not a Dialogue: 0,0:30:13.74,0:30:17.46,Default,,0000,0000,0000,,number that I can sum. I can't say "patient Dialogue: 0,0:30:17.46,0:30:20.16,Default,,0000,0000,0000,,ID number one plus patient ID number two". Dialogue: 0,0:30:20.16,0:30:21.72,Default,,0000,0000,0000,,No, this is a patient and another Dialogue: 0,0:30:21.72,0:30:23.40,Default,,0000,0000,0000,,patient, it's not a number that I can do Dialogue: 0,0:30:23.40,0:30:25.74,Default,,0000,0000,0000,,mathematical operations on, so I'm not Dialogue: 0,0:30:25.74,0:30:28.20,Default,,0000,0000,0000,,going to choose it. So we will choose Dialogue: 0,0:30:28.20,0:30:30.54,Default,,0000,0000,0000,,everything as I said, except for the Dialogue: 0,0:30:30.54,0:30:33.48,Default,,0000,0000,0000,,diabetic and the patient ID. I will Dialogue: 0,0:30:33.48,0:30:34.86,Default,,0000,0000,0000,,click on "Save". Dialogue: 0,0:30:34.86,0:30:37.74,Default,,0000,0000,0000,,And it's not showing me a warning again, Dialogue: 0,0:30:37.74,0:30:39.48,Default,,0000,0000,0000,,everything is good. Dialogue: 0,0:30:39.48,0:30:41.88,Default,,0000,0000,0000,,Now I can click on "Submit" Dialogue: 0,0:30:41.88,0:30:46.80,Default,,0000,0000,0000,,and review my normalization output. Dialogue: 0,0:30:46.80,0:30:48.24,Default,,0000,0000,0000,,Um. Dialogue: 0,0:30:48.24,0:30:51.66,Default,,0000,0000,0000,,So, if you click on "Submit" here, Dialogue: 0,0:30:51.66,0:30:54.66,Default,,0000,0000,0000,,you will choose "Create new" and Dialogue: 0,0:30:54.66,0:30:56.46,Default,,0000,0000,0000,,set the name that is mentioned here Dialogue: 0,0:30:56.46,0:30:59.90,Default,,0000,0000,0000,,inside the notebook. So it tells you Dialogue: 0,0:30:59.90,0:31:03.42,Default,,0000,0000,0000,,to create a job and name it, name Dialogue: 0,0:31:03.42,0:31:05.46,Default,,0000,0000,0000,,the experiment "MS Learn Diabetes Dialogue: 0,0:31:05.46,0:31:06.72,Default,,0000,0000,0000,,Training", because you will continue Dialogue: 0,0:31:06.72,0:31:10.16,Default,,0000,0000,0000,,working on and building component later. Dialogue: 0,0:31:10.16,0:31:13.02,Default,,0000,0000,0000,,I have it already created, I am the, uh, Dialogue: 0,0:31:13.02,0:31:16.92,Default,,0000,0000,0000,,we can review it together. So let Dialogue: 0,0:31:16.92,0:31:19.86,Default,,0000,0000,0000,,me just open this in another tab. I think Dialogue: 0,0:31:19.86,0:31:21.00,Default,,0000,0000,0000,,I have it... Dialogue: 0,0:31:21.00,0:31:23.66,Default,,0000,0000,0000,,here. Dialogue: 0,0:31:25.68,0:31:28.22,Default,,0000,0000,0000,,Okay. Dialogue: 0,0:31:30.72,0:31:34.74,Default,,0000,0000,0000,,So, these are all the jobs that I have Dialogue: 0,0:31:34.74,0:31:37.34,Default,,0000,0000,0000,,created. Dialogue: 0,0:31:37.86,0:31:40.12,Default,,0000,0000,0000,,All the jobs there. Let's do this over. Dialogue: 0,0:31:40.12,0:31:42.06,Default,,0000,0000,0000,,These are all the jobs that I have Dialogue: 0,0:31:42.06,0:31:43.68,Default,,0000,0000,0000,,submitted previously. Dialogue: 0,0:31:43.68,0:31:45.84,Default,,0000,0000,0000,,And I think this one is the Dialogue: 0,0:31:45.84,0:31:48.36,Default,,0000,0000,0000,,normalization job, so let's see the Dialogue: 0,0:31:48.36,0:31:50.10,Default,,0000,0000,0000,,output of it. Dialogue: 0,0:31:50.10,0:31:54.12,Default,,0000,0000,0000,,As you can see, it says, uh, "Check mark", yes, Dialogue: 0,0:31:54.12,0:31:56.64,Default,,0000,0000,0000,,which means that it worked, and we can Dialogue: 0,0:31:56.64,0:31:59.40,Default,,0000,0000,0000,,preview it. How can I do that? Right click Dialogue: 0,0:31:59.40,0:32:02.54,Default,,0000,0000,0000,,on it, choose "Preview data", Dialogue: 0,0:32:02.54,0:32:06.66,Default,,0000,0000,0000,,and as you can see all the data are Dialogue: 0,0:32:06.66,0:32:08.40,Default,,0000,0000,0000,,scaled down Dialogue: 0,0:32:08.40,0:32:10.98,Default,,0000,0000,0000,,so everything is between zero Dialogue: 0,0:32:10.98,0:32:15.86,Default,,0000,0000,0000,,and, uh, one I think. Dialogue: 0,0:32:15.86,0:32:18.90,Default,,0000,0000,0000,,So everything is good for us. Now we Dialogue: 0,0:32:18.90,0:32:21.84,Default,,0000,0000,0000,,can move forward to the next step Dialogue: 0,0:32:21.84,0:32:26.94,Default,,0000,0000,0000,,which is to create the whole pipeline. Dialogue: 0,0:32:26.94,0:32:30.84,Default,,0000,0000,0000,,So, uh, Carlota told you that Dialogue: 0,0:32:30.84,0:32:33.18,Default,,0000,0000,0000,,we're going to use a classification Dialogue: 0,0:32:33.18,0:32:37.26,Default,,0000,0000,0000,,model to create this data set, so let Dialogue: 0,0:32:37.26,0:32:40.62,Default,,0000,0000,0000,,me just drag and drop everything Dialogue: 0,0:32:40.62,0:32:43.14,Default,,0000,0000,0000,,to get runtime and we're doing Dialogue: 0,0:32:43.14,0:32:46.49,Default,,0000,0000,0000,,[INDISTINGUISHABLE] Dialogue: 0,0:32:46.49,0:32:48.47,Default,,0000,0000,0000,,about everything by Dialogue: 0,0:32:48.47,0:32:51.42,Default,,0000,0000,0000,,[INDISTINGUISHABLE] Dialogue: 0,0:32:51.42,0:32:52.92,Default,,0000,0000,0000,,So, Dialogue: 0,0:32:52.92,0:32:55.59,Default,,0000,0000,0000,,as a result, we are going to explain Dialogue: 0,0:32:55.59,0:32:59.76,Default,,0000,0000,0000,,[INDISTINGUISHABLE] Dialogue: 0,0:32:59.76,0:33:03.60,Default,,0000,0000,0000,,Yeah. So, I'm going to give this split Dialogue: 0,0:33:03.60,0:33:06.07,Default,,0000,0000,0000,,data. I'm going to take the Dialogue: 0,0:33:06.07,0:33:08.88,Default,,0000,0000,0000,,transformation data to split data and Dialogue: 0,0:33:08.88,0:33:10.38,Default,,0000,0000,0000,,connect it like that. Dialogue: 0,0:33:10.38,0:33:12.30,Default,,0000,0000,0000,,I'm going to get three model Dialogue: 0,0:33:12.30,0:33:15.24,Default,,0000,0000,0000,,components because I want to train my Dialogue: 0,0:33:15.24,0:33:16.68,Default,,0000,0000,0000,,model, Dialogue: 0,0:33:16.68,0:33:19.74,Default,,0000,0000,0000,,and I'm going to put it right here. Dialogue: 0,0:33:19.74,0:33:21.74,Default,,0000,0000,0000,,Okay. Dialogue: 0,0:33:21.74,0:33:24.42,Default,,0000,0000,0000,,Let's just move it down there. Okay. Dialogue: 0,0:33:24.42,0:33:27.06,Default,,0000,0000,0000,,And we are going to use a classification Dialogue: 0,0:33:27.06,0:33:28.62,Default,,0000,0000,0000,,model, Dialogue: 0,0:33:28.62,0:33:31.88,Default,,0000,0000,0000,,a two class Dialogue: 0,0:33:32.24,0:33:35.40,Default,,0000,0000,0000,,logistic regression model. Dialogue: 0,0:33:35.40,0:33:38.16,Default,,0000,0000,0000,,So I'm going to give this algorithm to Dialogue: 0,0:33:38.16,0:33:41.48,Default,,0000,0000,0000,,enable my model to work Dialogue: 0,0:33:41.82,0:33:45.96,Default,,0000,0000,0000,,This is the untrained model, this is... Dialogue: 0,0:33:45.96,0:33:48.06,Default,,0000,0000,0000,,here. Dialogue: 0,0:33:48.06,0:33:51.12,Default,,0000,0000,0000,,The left... Dialogue: 0,0:33:51.12,0:33:52.86,Default,,0000,0000,0000,,the left, uh, circuit, I'm going to Dialogue: 0,0:33:52.86,0:33:54.82,Default,,0000,0000,0000,,connect it to the data set, and the right Dialogue: 0,0:33:54.82,0:33:56.94,Default,,0000,0000,0000,,one, we are going to connect it to Dialogue: 0,0:33:56.94,0:33:59.70,Default,,0000,0000,0000,,evaluate model. Dialogue: 0,0:33:59.70,0:34:02.64,Default,,0000,0000,0000,,Evaluate model...so let's search for Dialogue: 0,0:34:02.64,0:34:05.22,Default,,0000,0000,0000,,"Evaluate model" here. Dialogue: 0,0:34:05.22,0:34:07.44,Default,,0000,0000,0000,,So because we want to do what...we want to Dialogue: 0,0:34:07.44,0:34:10.80,Default,,0000,0000,0000,,evaluate our model and see how it it has Dialogue: 0,0:34:10.80,0:34:13.79,Default,,0000,0000,0000,,been doing. Is it good, is it bad? Dialogue: 0,0:34:13.79,0:34:18.20,Default,,0000,0000,0000,,Um, sorry... Dialogue: 0,0:34:19.98,0:34:22.82,Default,,0000,0000,0000,,This is... Dialogue: 0,0:34:23.46,0:34:25.56,Default,,0000,0000,0000,,this is down there Dialogue: 0,0:34:25.56,0:34:28.14,Default,,0000,0000,0000,,after the score model. Dialogue: 0,0:34:28.14,0:34:31.32,Default,,0000,0000,0000,,So we have to get the score model first, Dialogue: 0,0:34:31.32,0:34:33.96,Default,,0000,0000,0000,,so let's get it. Dialogue: 0,0:34:33.96,0:34:36.12,Default,,0000,0000,0000,,And this will take the trained model and Dialogue: 0,0:34:36.12,0:34:37.26,Default,,0000,0000,0000,,the data set Dialogue: 0,0:34:37.26,0:34:39.42,Default,,0000,0000,0000,,to score our model and see if it's Dialogue: 0,0:34:39.42,0:34:42.18,Default,,0000,0000,0000,,performing good or bad. Dialogue: 0,0:34:42.18,0:34:44.41,Default,,0000,0000,0000,,And... Dialogue: 0,0:34:44.41,0:34:47.16,Default,,0000,0000,0000,,um... Dialogue: 0,0:34:47.16,0:34:49.08,Default,,0000,0000,0000,,after that, we have finished Dialogue: 0,0:34:49.08,0:34:51.92,Default,,0000,0000,0000,,everything. Now, we are going to do the what? Dialogue: 0,0:34:52.14,0:34:54.36,Default,,0000,0000,0000,,The presets for everything. Dialogue: 0,0:34:54.36,0:34:56.82,Default,,0000,0000,0000,,As a starter, we will be splitting our Dialogue: 0,0:34:56.82,0:34:58.92,Default,,0000,0000,0000,,data. So Dialogue: 0,0:34:58.92,0:35:01.14,Default,,0000,0000,0000,,how are we going to do this, according to Dialogue: 0,0:35:01.14,0:35:03.78,Default,,0000,0000,0000,,what? To the split rules. So I'm going to Dialogue: 0,0:35:03.78,0:35:05.94,Default,,0000,0000,0000,,double-click on it and choose "Split rules". Dialogue: 0,0:35:05.94,0:35:09.42,Default,,0000,0000,0000,,And the percentage is Dialogue: 0,0:35:09.42,0:35:11.78,Default,,0000,0000,0000,,70 percent for the [INSISTINGUASHABLE] Dialogue: 0,0:35:11.78,0:35:12.78,Default,,0000,0000,0000,,and 30 percent of the Dialogue: 0,0:35:12.78,0:35:14.82,Default,,0000,0000,0000,,data for Dialogue: 0,0:35:14.82,0:35:18.42,Default,,0000,0000,0000,,the valuation or for the scoring, okay? Dialogue: 0,0:35:18.42,0:35:20.88,Default,,0000,0000,0000,,I'm going to make it a randomization, so Dialogue: 0,0:35:20.88,0:35:22.98,Default,,0000,0000,0000,,I'm going to split data randomly and the Dialogue: 0,0:35:22.98,0:35:26.06,Default,,0000,0000,0000,,seat is, uh, Dialogue: 0,0:35:26.06,0:35:29.34,Default,,0000,0000,0000,,132, uh 23 I think...yeah. Dialogue: 0,0:35:29.34,0:35:32.52,Default,,0000,0000,0000,,And I think that's it. Dialogue: 0,0:35:32.52,0:35:35.04,Default,,0000,0000,0000,,The split says why this holds, and that's Dialogue: 0,0:35:35.04,0:35:36.24,Default,,0000,0000,0000,,good. Dialogue: 0,0:35:36.24,0:35:39.54,Default,,0000,0000,0000,,Now for the next one, which is the train Dialogue: 0,0:35:39.54,0:35:42.00,Default,,0000,0000,0000,,model we are going to connect it as Dialogue: 0,0:35:42.00,0:35:43.50,Default,,0000,0000,0000,,mentioned here. Dialogue: 0,0:35:43.50,0:35:48.66,Default,,0000,0000,0000,,And we have done that and...then why Dialogue: 0,0:35:48.66,0:35:50.70,Default,,0000,0000,0000,,am I having here? Let's double click Dialogue: 0,0:35:50.70,0:35:54.66,Default,,0000,0000,0000,,on it...yeah. It has...it needs the Dialogue: 0,0:35:54.66,0:35:57.18,Default,,0000,0000,0000,,label column that I am trying to predict. Dialogue: 0,0:35:57.18,0:35:58.68,Default,,0000,0000,0000,,So from here, I'm going to choose Dialogue: 0,0:35:58.68,0:36:01.38,Default,,0000,0000,0000,,diabetic. I'm going to save. Dialogue: 0,0:36:01.38,0:36:05.18,Default,,0000,0000,0000,,I'm going to close this one. Dialogue: 0,0:36:05.52,0:36:07.38,Default,,0000,0000,0000,,So it says here, Dialogue: 0,0:36:07.38,0:36:10.62,Default,,0000,0000,0000,,the diabetic label, the model, it will Dialogue: 0,0:36:10.62,0:36:12.30,Default,,0000,0000,0000,,predict the zero and one, because this is Dialogue: 0,0:36:12.30,0:36:14.70,Default,,0000,0000,0000,,a binary classification algorithm, so Dialogue: 0,0:36:14.70,0:36:16.26,Default,,0000,0000,0000,,it's going to predict either this or Dialogue: 0,0:36:16.26,0:36:17.52,Default,,0000,0000,0000,,that. Dialogue: 0,0:36:17.52,0:36:18.46,Default,,0000,0000,0000,,And... Dialogue: 0,0:36:18.46,0:36:20.16,Default,,0000,0000,0000,,um... Dialogue: 0,0:36:20.16,0:36:23.88,Default,,0000,0000,0000,,I think that's everything to run the the Dialogue: 0,0:36:23.88,0:36:25.86,Default,,0000,0000,0000,,pipeline. Dialogue: 0,0:36:25.86,0:36:29.04,Default,,0000,0000,0000,,So everything is done, everything is good Dialogue: 0,0:36:29.04,0:36:31.20,Default,,0000,0000,0000,,for this one. We're just gonna leave it Dialogue: 0,0:36:31.20,0:36:34.14,Default,,0000,0000,0000,,for now, because this is the next Dialogue: 0,0:36:34.14,0:36:35.62,Default,,0000,0000,0000,,step. Dialogue: 0,0:36:35.62,0:36:39.84,Default,,0000,0000,0000,,Um, this will be put instead of the Dialogue: 0,0:36:39.84,0:36:43.52,Default,,0000,0000,0000,,score model, but let's... Dialogue: 0,0:36:44.10,0:36:46.92,Default,,0000,0000,0000,,let's delete it for now. Dialogue: 0,0:36:46.92,0:36:49.50,Default,,0000,0000,0000,,Okay. Dialogue: 0,0:36:49.50,0:36:52.92,Default,,0000,0000,0000,,Now we have to submit the job in order Dialogue: 0,0:36:52.92,0:36:55.68,Default,,0000,0000,0000,,to see the output of it. So I can click Dialogue: 0,0:36:55.68,0:36:59.28,Default,,0000,0000,0000,,on "Submit" and choose the previous job Dialogue: 0,0:36:59.28,0:37:01.20,Default,,0000,0000,0000,,which is the one that I have showed you Dialogue: 0,0:37:01.20,0:37:02.46,Default,,0000,0000,0000,,before. Dialogue: 0,0:37:02.46,0:37:05.46,Default,,0000,0000,0000,,And then let's review its output Dialogue: 0,0:37:05.46,0:37:06.96,Default,,0000,0000,0000,,together here. Dialogue: 0,0:37:06.96,0:37:09.96,Default,,0000,0000,0000,,So if I go to the jobs, Dialogue: 0,0:37:09.96,0:37:15.12,Default,,0000,0000,0000,,if I go to MS Learn, maybe it is training? Dialogue: 0,0:37:15.12,0:37:18.18,Default,,0000,0000,0000,,I think it's the one that lasted the Dialogue: 0,0:37:18.18,0:37:20.64,Default,,0000,0000,0000,,longest, this one here. Dialogue: 0,0:37:20.64,0:37:23.70,Default,,0000,0000,0000,,So here I can see Dialogue: 0,0:37:23.70,0:37:27.08,Default,,0000,0000,0000,,the job output, what happened inside Dialogue: 0,0:37:27.08,0:37:30.42,Default,,0000,0000,0000,,the model, as you can see. Dialogue: 0,0:37:30.42,0:37:33.84,Default,,0000,0000,0000,,So the normalization we have seen Dialogue: 0,0:37:33.84,0:37:36.54,Default,,0000,0000,0000,,before, the split data, I can preview it. Dialogue: 0,0:37:36.54,0:37:39.36,Default,,0000,0000,0000,,The result one or the result two as it Dialogue: 0,0:37:39.36,0:37:41.76,Default,,0000,0000,0000,,splits the data to 70 here and Dialogue: 0,0:37:41.76,0:37:43.64,Default,,0000,0000,0000,,thirty percent here. Dialogue: 0,0:37:43.64,0:37:46.86,Default,,0000,0000,0000,,Um, I can see the score model, which is Dialogue: 0,0:37:46.86,0:37:49.14,Default,,0000,0000,0000,,something that we need Dialogue: 0,0:37:49.14,0:37:51.53,Default,,0000,0000,0000,,to review. Dialogue: 0,0:37:51.53,0:37:56.82,Default,,0000,0000,0000,,Inside the scroll model, uh, from Dialogue: 0,0:37:56.82,0:37:57.96,Default,,0000,0000,0000,,here, Dialogue: 0,0:37:57.96,0:38:00.96,Default,,0000,0000,0000,,we can see that... Dialogue: 0,0:38:00.96,0:38:04.46,Default,,0000,0000,0000,,let's get back here. Dialogue: 0,0:38:05.94,0:38:08.22,Default,,0000,0000,0000,,This is the data that the model has Dialogue: 0,0:38:08.22,0:38:11.58,Default,,0000,0000,0000,,been scored and this is a scoring output. Dialogue: 0,0:38:11.58,0:38:15.30,Default,,0000,0000,0000,,So it says "code label true", and he is Dialogue: 0,0:38:15.30,0:38:17.37,Default,,0000,0000,0000,,not diabetic, so this is, Dialogue: 0,0:38:17.37,0:38:19.20,Default,,0000,0000,0000,,um, Dialogue: 0,0:38:19.20,0:38:21.84,Default,,0000,0000,0000,,a wrong prediction, let's say. Dialogue: 0,0:38:21.84,0:38:23.88,Default,,0000,0000,0000,,For this one it's true and true, and this Dialogue: 0,0:38:23.88,0:38:26.88,Default,,0000,0000,0000,,is a good, like, what do you say, Dialogue: 0,0:38:26.88,0:38:29.46,Default,,0000,0000,0000,,prediction, and the probabilities of this Dialogue: 0,0:38:29.46,0:38:30.42,Default,,0000,0000,0000,,score, Dialogue: 0,0:38:30.42,0:38:33.12,Default,,0000,0000,0000,,which means the certainty of our model Dialogue: 0,0:38:33.12,0:38:36.62,Default,,0000,0000,0000,,of that this is really true. It's 80 percent. Dialogue: 0,0:38:36.62,0:38:38.78,Default,,0000,0000,0000,,For this one it's 75 percent. Dialogue: 0,0:38:38.78,0:38:42.60,Default,,0000,0000,0000,,So these are some cool metrics that we Dialogue: 0,0:38:42.60,0:38:45.36,Default,,0000,0000,0000,,can review to understand how our model Dialogue: 0,0:38:45.36,0:38:47.58,Default,,0000,0000,0000,,is performing. It's performing good for Dialogue: 0,0:38:47.58,0:38:48.54,Default,,0000,0000,0000,,now. Dialogue: 0,0:38:48.54,0:38:53.18,Default,,0000,0000,0000,,Let's check our evaluation model. Dialogue: 0,0:38:53.18,0:38:56.70,Default,,0000,0000,0000,,So this is the extra one that I told you Dialogue: 0,0:38:56.70,0:38:59.58,Default,,0000,0000,0000,,about. Instead of the Dialogue: 0,0:38:59.58,0:39:01.80,Default,,0000,0000,0000,,score model only, we are going to add Dialogue: 0,0:39:01.80,0:39:04.26,Default,,0000,0000,0000,,what evaluate model Dialogue: 0,0:39:04.26,0:39:06.90,Default,,0000,0000,0000,,after it. So here Dialogue: 0,0:39:06.90,0:39:09.42,Default,,0000,0000,0000,,we're going to go to our Asset Library Dialogue: 0,0:39:09.42,0:39:12.18,Default,,0000,0000,0000,,and we are going to choose the evaluate Dialogue: 0,0:39:12.18,0:39:14.94,Default,,0000,0000,0000,,model, Dialogue: 0,0:39:14.94,0:39:17.76,Default,,0000,0000,0000,,and we are going to put it here, and we Dialogue: 0,0:39:17.76,0:39:20.22,Default,,0000,0000,0000,,are going to connect it, and we are going Dialogue: 0,0:39:20.22,0:39:23.10,Default,,0000,0000,0000,,to submit the job using the same name of Dialogue: 0,0:39:23.10,0:39:25.14,Default,,0000,0000,0000,,the job that we used previously. Dialogue: 0,0:39:25.14,0:39:29.52,Default,,0000,0000,0000,,Let's review it. Also, so, after it Dialogue: 0,0:39:29.52,0:39:33.30,Default,,0000,0000,0000,,finishes, you will find it here. So I have Dialogue: 0,0:39:33.30,0:39:35.28,Default,,0000,0000,0000,,already done it before, this is how I'm Dialogue: 0,0:39:35.28,0:39:37.38,Default,,0000,0000,0000,,able to see the output. Dialogue: 0,0:39:37.38,0:39:40.32,Default,,0000,0000,0000,,So let's see Dialogue: 0,0:39:40.32,0:39:43.28,Default,,0000,0000,0000,,what is the output of this Dialogue: 0,0:39:43.28,0:39:45.66,Default,,0000,0000,0000,,evaluation process. Dialogue: 0,0:39:45.66,0:39:49.80,Default,,0000,0000,0000,,Here it mentioned to you that there are Dialogue: 0,0:39:49.80,0:39:51.48,Default,,0000,0000,0000,,some matrix, Dialogue: 0,0:39:51.48,0:39:54.84,Default,,0000,0000,0000,,like the confusion matrix, which Carlotta Dialogue: 0,0:39:54.84,0:39:57.06,Default,,0000,0000,0000,,told you about, there is the accuracy, the Dialogue: 0,0:39:57.06,0:39:59.76,Default,,0000,0000,0000,,precision, the recall, and F1 Score. Dialogue: 0,0:39:59.76,0:40:02.34,Default,,0000,0000,0000,,Every matrix gives us some insight about Dialogue: 0,0:40:02.34,0:40:04.92,Default,,0000,0000,0000,,our model. It helps us to understand it Dialogue: 0,0:40:04.92,0:40:08.58,Default,,0000,0000,0000,,more, and, um, Dialogue: 0,0:40:08.58,0:40:10.56,Default,,0000,0000,0000,,understand if it's overfitting, if Dialogue: 0,0:40:10.56,0:40:12.24,Default,,0000,0000,0000,,it's good, if it's bad, and really really, Dialogue: 0,0:40:12.24,0:40:16.34,Default,,0000,0000,0000,,like, understand how it's working. Dialogue: 0,0:40:17.06,0:40:20.40,Default,,0000,0000,0000,,Now I'm just waiting for the job to load. Dialogue: 0,0:40:20.40,0:40:22.71,Default,,0000,0000,0000,,Until it loads, Dialogue: 0,0:40:22.71,0:40:23.64,Default,,0000,0000,0000,,um, Dialogue: 0,0:40:23.64,0:40:26.04,Default,,0000,0000,0000,,we can continue Dialogue: 0,0:40:26.04,0:40:28.74,Default,,0000,0000,0000,,to work on our Dialogue: 0,0:40:28.74,0:40:31.80,Default,,0000,0000,0000,,model. So I will go to my designer. I'm Dialogue: 0,0:40:31.80,0:40:34.74,Default,,0000,0000,0000,,just going to confirm this. Dialogue: 0,0:40:34.74,0:40:38.28,Default,,0000,0000,0000,,And I'm going to continue working on it Dialogue: 0,0:40:38.28,0:40:39.78,Default,,0000,0000,0000,,from Dialogue: 0,0:40:39.78,0:40:42.12,Default,,0000,0000,0000,,where we have stopped. Where have we Dialogue: 0,0:40:42.12,0:40:43.56,Default,,0000,0000,0000,,stopped? Dialogue: 0,0:40:43.56,0:40:46.44,Default,,0000,0000,0000,,we have stopped on the evaluate model. So Dialogue: 0,0:40:46.44,0:40:48.96,Default,,0000,0000,0000,,I'm going to choose this one. Dialogue: 0,0:40:48.96,0:40:53.42,Default,,0000,0000,0000,,And it says here Dialogue: 0,0:40:54.18,0:40:56.94,Default,,0000,0000,0000,,"select experiment", "create inference Dialogue: 0,0:40:56.94,0:40:58.20,Default,,0000,0000,0000,,pipeline", so Dialogue: 0,0:40:58.20,0:41:01.08,Default,,0000,0000,0000,,I am going to go to the jobs, Dialogue: 0,0:41:01.08,0:41:04.68,Default,,0000,0000,0000,,I'm going to select my experiment. Dialogue: 0,0:41:04.68,0:41:06.66,Default,,0000,0000,0000,,I hope this works. Dialogue: 0,0:41:06.66,0:41:09.72,Default,,0000,0000,0000,,Okay. Finally, now we have our Dialogue: 0,0:41:09.72,0:41:12.18,Default,,0000,0000,0000,,evaluate model output. Dialogue: 0,0:41:12.18,0:41:15.48,Default,,0000,0000,0000,,Let's preview evaluation results Dialogue: 0,0:41:15.48,0:41:18.66,Default,,0000,0000,0000,,and, uh... Dialogue: 0,0:41:18.66,0:41:22.22,Default,,0000,0000,0000,,come on. Dialogue: 0,0:41:25.50,0:41:28.02,Default,,0000,0000,0000,,Finally. Now we can create our inference Dialogue: 0,0:41:28.02,0:41:31.02,Default,,0000,0000,0000,,pipeline. So, Dialogue: 0,0:41:31.02,0:41:33.51,Default,,0000,0000,0000,,I think it says that... Dialogue: 0,0:41:33.51,0:41:35.28,Default,,0000,0000,0000,,um... Dialogue: 0,0:41:35.28,0:41:38.16,Default,,0000,0000,0000,,select the experiment, then select MS Dialogue: 0,0:41:38.16,0:41:39.36,Default,,0000,0000,0000,,Learn. So, Dialogue: 0,0:41:39.36,0:41:43.32,Default,,0000,0000,0000,,I am just going to select it, Dialogue: 0,0:41:43.32,0:41:48.30,Default,,0000,0000,0000,,and finally. Now we can, the ROC curve, we Dialogue: 0,0:41:48.30,0:41:51.00,Default,,0000,0000,0000,,can see it, that the true positive rate Dialogue: 0,0:41:51.00,0:41:53.76,Default,,0000,0000,0000,,and the force was integrate. The false Dialogue: 0,0:41:53.76,0:41:56.66,Default,,0000,0000,0000,,positive rate is increasing with time, Dialogue: 0,0:41:56.66,0:42:01.02,Default,,0000,0000,0000,,and also the true positive rate. True Dialogue: 0,0:42:01.02,0:42:03.54,Default,,0000,0000,0000,,positive is something that it predicted, Dialogue: 0,0:42:03.54,0:42:06.96,Default,,0000,0000,0000,,that it is, uh, positive it has diabetes, Dialogue: 0,0:42:06.96,0:42:09.48,Default,,0000,0000,0000,,and it's really...it's really true. Dialogue: 0,0:42:09.48,0:42:12.60,Default,,0000,0000,0000,,The person really has diabetes. Okay. And Dialogue: 0,0:42:12.60,0:42:14.76,Default,,0000,0000,0000,,for the false positive, it predicted that Dialogue: 0,0:42:14.76,0:42:17.58,Default,,0000,0000,0000,,someone has diabetes and someone doesn't Dialogue: 0,0:42:17.58,0:42:20.96,Default,,0000,0000,0000,,have it. This is what true position and Dialogue: 0,0:42:20.96,0:42:24.96,Default,,0000,0000,0000,,false positive means. This is the record Dialogue: 0,0:42:24.96,0:42:28.02,Default,,0000,0000,0000,,curve, so we can review the metrics Dialogue: 0,0:42:28.02,0:42:32.16,Default,,0000,0000,0000,,of our model. This is the lift curve. I Dialogue: 0,0:42:32.16,0:42:36.00,Default,,0000,0000,0000,,can change the threshold of my confusion Dialogue: 0,0:42:36.00,0:42:37.74,Default,,0000,0000,0000,,matrix here Dialogue: 0,0:42:37.74,0:42:39.12,Default,,0000,0000,0000,,and this could [...] don't want to add Dialogue: 0,0:42:39.12,0:42:43.92,Default,,0000,0000,0000,,anything about the...the graphs, Dialogue: 0,0:42:43.92,0:42:47.00,Default,,0000,0000,0000,,you can do so. Dialogue: 0,0:42:50.46,0:42:51.00,Default,,0000,0000,0000,,Um, Dialogue: 0,0:42:51.00,0:42:54.72,Default,,0000,0000,0000,,yeah, so just wanted to if you go yeah I Dialogue: 0,0:42:54.72,0:42:57.12,Default,,0000,0000,0000,,just wanted to comment comment for the Dialogue: 0,0:42:57.12,0:43:00.48,Default,,0000,0000,0000,,RSC curve uh that actually from this Dialogue: 0,0:43:00.48,0:43:03.90,Default,,0000,0000,0000,,graph the metric which uh usually we're Dialogue: 0,0:43:03.90,0:43:06.96,Default,,0000,0000,0000,,going to compute is the end area under Dialogue: 0,0:43:06.96,0:43:09.90,Default,,0000,0000,0000,,under the curve and this coefficient or Dialogue: 0,0:43:09.90,0:43:12.24,Default,,0000,0000,0000,,metric Dialogue: 0,0:43:12.24,0:43:15.06,Default,,0000,0000,0000,,um it's a confusion Dialogue: 0,0:43:15.06,0:43:18.42,Default,,0000,0000,0000,,um is a value that could span from from Dialogue: 0,0:43:18.42,0:43:22.92,Default,,0000,0000,0000,,zero to one and the the highest is Dialogue: 0,0:43:22.92,0:43:23.48,Default,,0000,0000,0000,,um Dialogue: 0,0:43:23.48,0:43:26.70,Default,,0000,0000,0000,,this the highest is the the score so the Dialogue: 0,0:43:26.70,0:43:29.22,Default,,0000,0000,0000,,the closest one Dialogue: 0,0:43:29.22,0:43:32.76,Default,,0000,0000,0000,,um so the the highest is the amount of Dialogue: 0,0:43:32.76,0:43:35.28,Default,,0000,0000,0000,,area under this curve Dialogue: 0,0:43:35.28,0:43:40.50,Default,,0000,0000,0000,,um the the the highest performance uh we Dialogue: 0,0:43:40.50,0:43:43.32,Default,,0000,0000,0000,,we've got from from our model and Dialogue: 0,0:43:43.32,0:43:46.44,Default,,0000,0000,0000,,another thing is what John is Dialogue: 0,0:43:46.44,0:43:49.68,Default,,0000,0000,0000,,um playing with so this threshold for Dialogue: 0,0:43:49.68,0:43:51.38,Default,,0000,0000,0000,,the logistic Dialogue: 0,0:43:51.38,0:43:55.92,Default,,0000,0000,0000,,regression is the threshold used by the Dialogue: 0,0:43:55.92,0:43:57.18,Default,,0000,0000,0000,,model Dialogue: 0,0:43:57.18,0:43:58.74,Default,,0000,0000,0000,,um to Dialogue: 0,0:43:58.74,0:43:59.52,Default,,0000,0000,0000,,um Dialogue: 0,0:43:59.52,0:44:02.94,Default,,0000,0000,0000,,to predict uh if the category is zero or Dialogue: 0,0:44:02.94,0:44:05.22,Default,,0000,0000,0000,,one so if the probability the Dialogue: 0,0:44:05.22,0:44:08.60,Default,,0000,0000,0000,,probability score is above the threshold Dialogue: 0,0:44:08.60,0:44:11.58,Default,,0000,0000,0000,,then the category will be predicted as Dialogue: 0,0:44:11.58,0:44:15.36,Default,,0000,0000,0000,,one while if the the probability is Dialogue: 0,0:44:15.36,0:44:17.46,Default,,0000,0000,0000,,below the threshold in this case for Dialogue: 0,0:44:17.46,0:44:21.30,Default,,0000,0000,0000,,example 0.5 the category is predicted as Dialogue: 0,0:44:21.30,0:44:23.58,Default,,0000,0000,0000,,as zero so that's why it's very Dialogue: 0,0:44:23.58,0:44:26.10,Default,,0000,0000,0000,,important to um to choose the the Dialogue: 0,0:44:26.10,0:44:27.84,Default,,0000,0000,0000,,threshold because the performance really Dialogue: 0,0:44:27.84,0:44:29.52,Default,,0000,0000,0000,,can vary Dialogue: 0,0:44:29.52,0:44:30.56,Default,,0000,0000,0000,,um Dialogue: 0,0:44:30.56,0:44:34.38,Default,,0000,0000,0000,,with this threshold value Dialogue: 0,0:44:34.38,0:44:41.10,Default,,0000,0000,0000,,uh thank you uh so much uh kellota and Dialogue: 0,0:44:41.40,0:44:44.40,Default,,0000,0000,0000,,as I mentioned now we are going to like Dialogue: 0,0:44:44.40,0:44:46.56,Default,,0000,0000,0000,,create our inference pipeline so we are Dialogue: 0,0:44:46.56,0:44:48.54,Default,,0000,0000,0000,,going to select the latest one which I Dialogue: 0,0:44:48.54,0:44:50.82,Default,,0000,0000,0000,,already have it opened here this is the Dialogue: 0,0:44:50.82,0:44:52.86,Default,,0000,0000,0000,,one that we were reviewing together this Dialogue: 0,0:44:52.86,0:44:55.50,Default,,0000,0000,0000,,is where we have stopped and we're going Dialogue: 0,0:44:55.50,0:44:57.60,Default,,0000,0000,0000,,to create an inference pipeline we are Dialogue: 0,0:44:57.60,0:44:59.52,Default,,0000,0000,0000,,going to choose a real-time inference Dialogue: 0,0:44:59.52,0:45:02.52,Default,,0000,0000,0000,,pipeline okay Dialogue: 0,0:45:02.52,0:45:05.16,Default,,0000,0000,0000,,um from where I can find this here as it Dialogue: 0,0:45:05.16,0:45:08.10,Default,,0000,0000,0000,,says real-time inference pipeline Dialogue: 0,0:45:08.10,0:45:10.68,Default,,0000,0000,0000,,so it's gonna add some things to my Dialogue: 0,0:45:10.68,0:45:12.42,Default,,0000,0000,0000,,workspace it's going to add the web Dialogue: 0,0:45:12.42,0:45:13.98,Default,,0000,0000,0000,,service inboard it's going to have the Dialogue: 0,0:45:13.98,0:45:15.78,Default,,0000,0000,0000,,web service output because we will be Dialogue: 0,0:45:15.78,0:45:18.18,Default,,0000,0000,0000,,creating it as a web service to access Dialogue: 0,0:45:18.18,0:45:19.74,Default,,0000,0000,0000,,it from the internet Dialogue: 0,0:45:19.74,0:45:21.90,Default,,0000,0000,0000,,uh what are we going to do we're going Dialogue: 0,0:45:21.90,0:45:24.72,Default,,0000,0000,0000,,to remove this diabetes data okay Dialogue: 0,0:45:24.72,0:45:27.54,Default,,0000,0000,0000,,and we are going to get a component Dialogue: 0,0:45:27.54,0:45:29.36,Default,,0000,0000,0000,,called Web Dialogue: 0,0:45:29.36,0:45:33.18,Default,,0000,0000,0000,,input and what's up let me check Dialogue: 0,0:45:33.18,0:45:35.94,Default,,0000,0000,0000,,it's enter data manually Dialogue: 0,0:45:35.94,0:45:38.40,Default,,0000,0000,0000,,we have we already have the with input Dialogue: 0,0:45:38.40,0:45:39.54,Default,,0000,0000,0000,,present Dialogue: 0,0:45:39.54,0:45:42.12,Default,,0000,0000,0000,,so we are going to get the entire data Dialogue: 0,0:45:42.12,0:45:43.20,Default,,0000,0000,0000,,manually Dialogue: 0,0:45:43.20,0:45:45.42,Default,,0000,0000,0000,,and we're going to collect it to connect Dialogue: 0,0:45:45.42,0:45:49.56,Default,,0000,0000,0000,,it as it was connected before like that Dialogue: 0,0:45:49.56,0:45:53.04,Default,,0000,0000,0000,,and also I am not going to directly take Dialogue: 0,0:45:53.04,0:45:55.26,Default,,0000,0000,0000,,the web service sorry escort model to Dialogue: 0,0:45:55.26,0:45:57.84,Default,,0000,0000,0000,,the web service output like that Dialogue: 0,0:45:57.84,0:46:00.24,Default,,0000,0000,0000,,I'm going to delete this Dialogue: 0,0:46:00.24,0:46:03.96,Default,,0000,0000,0000,,and I'm going to execute a python script Dialogue: 0,0:46:03.96,0:46:05.88,Default,,0000,0000,0000,,before Dialogue: 0,0:46:05.88,0:46:09.50,Default,,0000,0000,0000,,I display my result Dialogue: 0,0:46:10.68,0:46:12.06,Default,,0000,0000,0000,,so Dialogue: 0,0:46:12.06,0:46:17.48,Default,,0000,0000,0000,,this will be connected like okay but Dialogue: 0,0:46:19.26,0:46:20.40,Default,,0000,0000,0000,,so Dialogue: 0,0:46:20.40,0:46:23.60,Default,,0000,0000,0000,,the other way around Dialogue: 0,0:46:23.60,0:46:27.66,Default,,0000,0000,0000,,and from here I am going to connect this Dialogue: 0,0:46:27.66,0:46:30.96,Default,,0000,0000,0000,,with that and there is some data uh that Dialogue: 0,0:46:30.96,0:46:33.48,Default,,0000,0000,0000,,we will be getting from the node or from Dialogue: 0,0:46:33.48,0:46:37.68,Default,,0000,0000,0000,,the the examination here and this is the Dialogue: 0,0:46:37.68,0:46:40.74,Default,,0000,0000,0000,,data that will be entered like to our Dialogue: 0,0:46:40.74,0:46:44.40,Default,,0000,0000,0000,,website manually okay this is instead of Dialogue: 0,0:46:44.40,0:46:47.46,Default,,0000,0000,0000,,the data that we have been getting from Dialogue: 0,0:46:47.46,0:46:49.74,Default,,0000,0000,0000,,our data set that we created so I'm just Dialogue: 0,0:46:49.74,0:46:51.96,Default,,0000,0000,0000,,going to double click on it and choose Dialogue: 0,0:46:51.96,0:46:55.58,Default,,0000,0000,0000,,CSV and I will choose it has headers Dialogue: 0,0:46:55.58,0:47:00.84,Default,,0000,0000,0000,,and I will take or copy this content and Dialogue: 0,0:47:00.84,0:47:02.82,Default,,0000,0000,0000,,put it there okay Dialogue: 0,0:47:02.82,0:47:05.70,Default,,0000,0000,0000,,so let's do it Dialogue: 0,0:47:05.70,0:47:07.92,Default,,0000,0000,0000,,I think I have to click on edit code now Dialogue: 0,0:47:07.92,0:47:10.68,Default,,0000,0000,0000,,I can click on Save and I can close it Dialogue: 0,0:47:10.68,0:47:13.08,Default,,0000,0000,0000,,another thing which is the python script Dialogue: 0,0:47:13.08,0:47:16.70,Default,,0000,0000,0000,,that we will be executing Dialogue: 0,0:47:17.10,0:47:19.38,Default,,0000,0000,0000,,um yeah we are going to remove this also Dialogue: 0,0:47:19.38,0:47:21.14,Default,,0000,0000,0000,,we don't need the evaluate model anymore Dialogue: 0,0:47:21.14,0:47:24.32,Default,,0000,0000,0000,,so we are going to remove Dialogue: 0,0:47:24.32,0:47:28.58,Default,,0000,0000,0000,,script that I will be executing okay Dialogue: 0,0:47:28.58,0:47:32.60,Default,,0000,0000,0000,,I can find it here Dialogue: 0,0:47:33.54,0:47:34.62,Default,,0000,0000,0000,,um Dialogue: 0,0:47:34.62,0:47:35.76,Default,,0000,0000,0000,,yeah Dialogue: 0,0:47:35.76,0:47:38.64,Default,,0000,0000,0000,,this is the python script that we will Dialogue: 0,0:47:38.64,0:47:41.52,Default,,0000,0000,0000,,execute and it says to you that this Dialogue: 0,0:47:41.52,0:47:43.62,Default,,0000,0000,0000,,code selects only the patient's ID Dialogue: 0,0:47:43.62,0:47:45.00,Default,,0000,0000,0000,,that's correct label the school Dialogue: 0,0:47:45.00,0:47:47.70,Default,,0000,0000,0000,,probability and return returns them to Dialogue: 0,0:47:47.70,0:47:49.98,Default,,0000,0000,0000,,the web service output so we don't want Dialogue: 0,0:47:49.98,0:47:51.96,Default,,0000,0000,0000,,to return all the columns as we have Dialogue: 0,0:47:51.96,0:47:53.34,Default,,0000,0000,0000,,seen previously Dialogue: 0,0:47:53.34,0:47:55.56,Default,,0000,0000,0000,,uh the determines everything Dialogue: 0,0:47:55.56,0:47:56.94,Default,,0000,0000,0000,,so Dialogue: 0,0:47:56.94,0:47:59.04,Default,,0000,0000,0000,,we want to return certain stuff the Dialogue: 0,0:47:59.04,0:48:02.94,Default,,0000,0000,0000,,stuff that we will use inside our Dialogue: 0,0:48:02.94,0:48:05.64,Default,,0000,0000,0000,,endpoint so I'm just going to select Dialogue: 0,0:48:05.64,0:48:07.98,Default,,0000,0000,0000,,everything and delete it and Dialogue: 0,0:48:07.98,0:48:11.06,Default,,0000,0000,0000,,paste the code that I have gotten from Dialogue: 0,0:48:11.06,0:48:14.28,Default,,0000,0000,0000,,the uh Dialogue: 0,0:48:14.28,0:48:16.50,Default,,0000,0000,0000,,the Microsoft learn Docs Dialogue: 0,0:48:16.50,0:48:19.08,Default,,0000,0000,0000,,now I can click on Save and I can close Dialogue: 0,0:48:19.08,0:48:20.28,Default,,0000,0000,0000,,this Dialogue: 0,0:48:20.28,0:48:21.96,Default,,0000,0000,0000,,let me check something I don't think Dialogue: 0,0:48:21.96,0:48:25.02,Default,,0000,0000,0000,,it's saved it's saved but the display is Dialogue: 0,0:48:25.02,0:48:26.16,Default,,0000,0000,0000,,wrong okay Dialogue: 0,0:48:26.16,0:48:30.30,Default,,0000,0000,0000,,and now I think everything is good to go Dialogue: 0,0:48:30.30,0:48:32.64,Default,,0000,0000,0000,,I'm just gonna double check everything Dialogue: 0,0:48:32.64,0:48:36.36,Default,,0000,0000,0000,,so uh yeah we are gonna change the name Dialogue: 0,0:48:36.36,0:48:38.64,Default,,0000,0000,0000,,of this uh Dialogue: 0,0:48:38.64,0:48:40.80,Default,,0000,0000,0000,,Pipeline and we are gonna call it Dialogue: 0,0:48:40.80,0:48:42.78,Default,,0000,0000,0000,,predict Dialogue: 0,0:48:42.78,0:48:46.32,Default,,0000,0000,0000,,diabetes okay Dialogue: 0,0:48:46.32,0:48:50.34,Default,,0000,0000,0000,,now let's close it and Dialogue: 0,0:48:50.34,0:48:57.12,Default,,0000,0000,0000,,I think that we are good to go so Dialogue: 0,0:48:57.12,0:48:59.30,Default,,0000,0000,0000,,um Dialogue: 0,0:48:59.72,0:49:04.46,Default,,0000,0000,0000,,okay I think everything is good for us Dialogue: 0,0:49:06.42,0:49:08.34,Default,,0000,0000,0000,,I just want to make sure of something is Dialogue: 0,0:49:08.34,0:49:12.42,Default,,0000,0000,0000,,the data is correct the data is uh yeah Dialogue: 0,0:49:12.42,0:49:13.56,Default,,0000,0000,0000,,it's correct Dialogue: 0,0:49:13.56,0:49:16.32,Default,,0000,0000,0000,,okay now I can run the pipeline let's Dialogue: 0,0:49:16.32,0:49:17.64,Default,,0000,0000,0000,,submit Dialogue: 0,0:49:17.64,0:49:21.00,Default,,0000,0000,0000,,select an existing Pipeline and we're Dialogue: 0,0:49:21.00,0:49:22.74,Default,,0000,0000,0000,,going to choose the MS layer and Dialogue: 0,0:49:22.74,0:49:24.60,Default,,0000,0000,0000,,diabetes training which is the pipeline Dialogue: 0,0:49:24.60,0:49:27.06,Default,,0000,0000,0000,,that we have been working on Dialogue: 0,0:49:27.06,0:49:31.62,Default,,0000,0000,0000,,from the beginning of this module Dialogue: 0,0:49:31.68,0:49:33.84,Default,,0000,0000,0000,,I don't think that this is going to take Dialogue: 0,0:49:33.84,0:49:36.06,Default,,0000,0000,0000,,much time so we have submitted the job Dialogue: 0,0:49:36.06,0:49:37.32,Default,,0000,0000,0000,,and it's running Dialogue: 0,0:49:37.32,0:49:40.14,Default,,0000,0000,0000,,until the job ends we are going to set Dialogue: 0,0:49:40.14,0:49:41.72,Default,,0000,0000,0000,,everything Dialogue: 0,0:49:41.72,0:49:45.60,Default,,0000,0000,0000,,and for deploying a service Dialogue: 0,0:49:45.60,0:49:49.56,Default,,0000,0000,0000,,in order to deploy a service okay Dialogue: 0,0:49:49.56,0:49:50.52,Default,,0000,0000,0000,,um Dialogue: 0,0:49:50.52,0:49:54.00,Default,,0000,0000,0000,,I have to have the job ready so Dialogue: 0,0:49:54.00,0:49:56.04,Default,,0000,0000,0000,,until it's ready or you can deploy it so Dialogue: 0,0:49:56.04,0:49:58.32,Default,,0000,0000,0000,,let's go to the job the job details from Dialogue: 0,0:49:58.32,0:50:01.32,Default,,0000,0000,0000,,here okay Dialogue: 0,0:50:01.32,0:50:05.12,Default,,0000,0000,0000,,and until it finishes Dialogue: 0,0:50:05.12,0:50:07.26,Default,,0000,0000,0000,,Carlotta do you think that we can have Dialogue: 0,0:50:07.26,0:50:09.24,Default,,0000,0000,0000,,the questions and then we can get back Dialogue: 0,0:50:09.24,0:50:12.86,Default,,0000,0000,0000,,to the job I'm deploying it Dialogue: 0,0:50:13.70,0:50:17.58,Default,,0000,0000,0000,,yeah yeah yeah so yeah yeah guys if you Dialogue: 0,0:50:17.58,0:50:18.98,Default,,0000,0000,0000,,have any questions Dialogue: 0,0:50:18.98,0:50:24.12,Default,,0000,0000,0000,,uh on on what you just uh just saw here Dialogue: 0,0:50:24.12,0:50:26.94,Default,,0000,0000,0000,,or into introductions feel free this is Dialogue: 0,0:50:26.94,0:50:30.30,Default,,0000,0000,0000,,a good moment we can uh we can discuss Dialogue: 0,0:50:30.30,0:50:33.90,Default,,0000,0000,0000,,now while we wait for this job to to Dialogue: 0,0:50:33.90,0:50:36.26,Default,,0000,0000,0000,,finish Dialogue: 0,0:50:36.30,0:50:38.76,Default,,0000,0000,0000,,uh and the Dialogue: 0,0:50:38.76,0:50:40.22,Default,,0000,0000,0000,,can can Dialogue: 0,0:50:40.22,0:50:45.00,Default,,0000,0000,0000,,we have the energy check one or like Dialogue: 0,0:50:45.00,0:50:47.70,Default,,0000,0000,0000,,what do you think uh yeah we can also go Dialogue: 0,0:50:47.70,0:50:49.68,Default,,0000,0000,0000,,to the knowledge check Dialogue: 0,0:50:49.68,0:50:50.94,Default,,0000,0000,0000,,um Dialogue: 0,0:50:50.94,0:50:56.34,Default,,0000,0000,0000,,yeah okay so let me share my screen Dialogue: 0,0:50:56.34,0:50:58.98,Default,,0000,0000,0000,,yeah so if you have not any questions Dialogue: 0,0:50:58.98,0:51:01.62,Default,,0000,0000,0000,,for us we can maybe propose some Dialogue: 0,0:51:01.62,0:51:05.34,Default,,0000,0000,0000,,questions to to you that you can Dialogue: 0,0:51:05.34,0:51:06.24,Default,,0000,0000,0000,,um Dialogue: 0,0:51:06.24,0:51:09.66,Default,,0000,0000,0000,,uh to check our knowledge so far and you Dialogue: 0,0:51:09.66,0:51:12.90,Default,,0000,0000,0000,,can uh maybe answer to these questions Dialogue: 0,0:51:12.90,0:51:15.42,Default,,0000,0000,0000,,uh via chat Dialogue: 0,0:51:15.42,0:51:18.30,Default,,0000,0000,0000,,um so we have do you see my screen can Dialogue: 0,0:51:18.30,0:51:19.86,Default,,0000,0000,0000,,you see my screen Dialogue: 0,0:51:19.86,0:51:22.02,Default,,0000,0000,0000,,yes Dialogue: 0,0:51:22.02,0:51:25.44,Default,,0000,0000,0000,,um so John I think I will read this Dialogue: 0,0:51:25.44,0:51:29.04,Default,,0000,0000,0000,,question loud and ask it to you okay so Dialogue: 0,0:51:29.04,0:51:32.04,Default,,0000,0000,0000,,are you ready to transfer Dialogue: 0,0:51:32.04,0:51:33.66,Default,,0000,0000,0000,,yes I am Dialogue: 0,0:51:33.66,0:51:35.46,Default,,0000,0000,0000,,so Dialogue: 0,0:51:35.46,0:51:37.26,Default,,0000,0000,0000,,um you're using Azure machine learning Dialogue: 0,0:51:37.26,0:51:39.78,Default,,0000,0000,0000,,designer to create a training pipeline Dialogue: 0,0:51:39.78,0:51:42.54,Default,,0000,0000,0000,,for a binary classification model so Dialogue: 0,0:51:42.54,0:51:45.30,Default,,0000,0000,0000,,what what we were doing in our demo Dialogue: 0,0:51:45.30,0:51:48.06,Default,,0000,0000,0000,,right and you have added a data set Dialogue: 0,0:51:48.06,0:51:51.66,Default,,0000,0000,0000,,containing features and labels uh a true Dialogue: 0,0:51:51.66,0:51:54.36,Default,,0000,0000,0000,,class decision Forest module so we used Dialogue: 0,0:51:54.36,0:51:56.82,Default,,0000,0000,0000,,a logistic regression model our Dialogue: 0,0:51:56.82,0:51:59.10,Default,,0000,0000,0000,,um in our example here we're using A2 Dialogue: 0,0:51:59.10,0:52:01.26,Default,,0000,0000,0000,,class decision force model Dialogue: 0,0:52:01.26,0:52:04.50,Default,,0000,0000,0000,,and of course a trained model model you Dialogue: 0,0:52:04.50,0:52:07.20,Default,,0000,0000,0000,,plan now to use score model and evaluate Dialogue: 0,0:52:07.20,0:52:09.48,Default,,0000,0000,0000,,model modules to test the train model Dialogue: 0,0:52:09.48,0:52:11.64,Default,,0000,0000,0000,,with the subset of the data set that Dialogue: 0,0:52:11.64,0:52:13.50,Default,,0000,0000,0000,,wasn't used for training Dialogue: 0,0:52:13.50,0:52:15.96,Default,,0000,0000,0000,,but what are we missing so what's Dialogue: 0,0:52:15.96,0:52:18.78,Default,,0000,0000,0000,,another model you should add and we have Dialogue: 0,0:52:18.78,0:52:21.66,Default,,0000,0000,0000,,three options we have join data we have Dialogue: 0,0:52:21.66,0:52:25.20,Default,,0000,0000,0000,,split data or we have select columns in Dialogue: 0,0:52:25.20,0:52:26.82,Default,,0000,0000,0000,,in that set Dialogue: 0,0:52:26.82,0:52:28.26,Default,,0000,0000,0000,,so Dialogue: 0,0:52:28.26,0:52:32.04,Default,,0000,0000,0000,,um while John thinks about the answer uh Dialogue: 0,0:52:32.04,0:52:33.84,Default,,0000,0000,0000,,go ahead and Dialogue: 0,0:52:33.84,0:52:34.80,Default,,0000,0000,0000,,um Dialogue: 0,0:52:34.80,0:52:37.80,Default,,0000,0000,0000,,answer yourself so give us your your Dialogue: 0,0:52:37.80,0:52:39.54,Default,,0000,0000,0000,,guess Dialogue: 0,0:52:39.54,0:52:41.94,Default,,0000,0000,0000,,put in the chat or just come off mute Dialogue: 0,0:52:41.94,0:52:44.90,Default,,0000,0000,0000,,and announcer Dialogue: 0,0:52:46.74,0:52:48.96,Default,,0000,0000,0000,,a b yes Dialogue: 0,0:52:48.96,0:52:50.58,Default,,0000,0000,0000,,yeah what do you think is the correct Dialogue: 0,0:52:50.58,0:52:53.58,Default,,0000,0000,0000,,answer for this one I need something to Dialogue: 0,0:52:53.58,0:52:56.58,Default,,0000,0000,0000,,uh like I have to score my model and I Dialogue: 0,0:52:56.58,0:53:00.36,Default,,0000,0000,0000,,have to evaluate it so I I like I need Dialogue: 0,0:53:00.36,0:53:03.12,Default,,0000,0000,0000,,something to enable me to do these two Dialogue: 0,0:53:03.12,0:53:05.36,Default,,0000,0000,0000,,things Dialogue: 0,0:53:06.66,0:53:09.12,Default,,0000,0000,0000,,I think it's something you showed us in Dialogue: 0,0:53:09.12,0:53:12.98,Default,,0000,0000,0000,,in your pipeline right John Dialogue: 0,0:53:13.26,0:53:16.82,Default,,0000,0000,0000,,of course I did Dialogue: 0,0:53:23.46,0:53:28.02,Default,,0000,0000,0000,,uh we have no guests yeah Dialogue: 0,0:53:28.02,0:53:32.28,Default,,0000,0000,0000,,can someone like someone want to guess Dialogue: 0,0:53:32.28,0:53:35.58,Default,,0000,0000,0000,,uh we have a b yeah Dialogue: 0,0:53:35.58,0:53:38.76,Default,,0000,0000,0000,,uh maybe Dialogue: 0,0:53:38.76,0:53:43.26,Default,,0000,0000,0000,,so uh in order to do this in order to do Dialogue: 0,0:53:43.26,0:53:46.20,Default,,0000,0000,0000,,this I mentioned the Dialogue: 0,0:53:46.20,0:53:49.38,Default,,0000,0000,0000,,the module that is going to help me to Dialogue: 0,0:53:49.38,0:53:53.82,Default,,0000,0000,0000,,to divide my data into two things 70 for Dialogue: 0,0:53:53.82,0:53:56.22,Default,,0000,0000,0000,,the training and thirty percent for the Dialogue: 0,0:53:56.22,0:53:59.34,Default,,0000,0000,0000,,evaluation so what did I use I used Dialogue: 0,0:53:59.34,0:54:01.86,Default,,0000,0000,0000,,split data because this is what is going Dialogue: 0,0:54:01.86,0:54:05.28,Default,,0000,0000,0000,,to split my data randomly into training Dialogue: 0,0:54:05.28,0:54:08.58,Default,,0000,0000,0000,,data and validation data so the correct Dialogue: 0,0:54:08.58,0:54:12.24,Default,,0000,0000,0000,,answer is B and good job eek thank you Dialogue: 0,0:54:12.24,0:54:13.98,Default,,0000,0000,0000,,for participating Dialogue: 0,0:54:13.98,0:54:17.40,Default,,0000,0000,0000,,next question please Dialogue: 0,0:54:17.40,0:54:19.34,Default,,0000,0000,0000,,yes Dialogue: 0,0:54:19.34,0:54:22.56,Default,,0000,0000,0000,,answer so thanks John Dialogue: 0,0:54:22.56,0:54:26.04,Default,,0000,0000,0000,,uh for uh explaining us the the correct Dialogue: 0,0:54:26.04,0:54:26.94,Default,,0000,0000,0000,,one Dialogue: 0,0:54:26.94,0:54:30.42,Default,,0000,0000,0000,,and we want to go with question two Dialogue: 0,0:54:30.42,0:54:33.18,Default,,0000,0000,0000,,yeah so uh I'm going to ask you now Dialogue: 0,0:54:33.18,0:54:35.88,Default,,0000,0000,0000,,karnata you use Azure machine learning Dialogue: 0,0:54:35.88,0:54:38.28,Default,,0000,0000,0000,,designer to create a training pipeline Dialogue: 0,0:54:38.28,0:54:40.50,Default,,0000,0000,0000,,for your classification model Dialogue: 0,0:54:40.50,0:54:44.10,Default,,0000,0000,0000,,what must you do before you deploy this Dialogue: 0,0:54:44.10,0:54:45.96,Default,,0000,0000,0000,,model as a service you have to do Dialogue: 0,0:54:45.96,0:54:47.58,Default,,0000,0000,0000,,something before you deploy it what do Dialogue: 0,0:54:47.58,0:54:49.74,Default,,0000,0000,0000,,you think is the correct answer Dialogue: 0,0:54:49.74,0:54:52.74,Default,,0000,0000,0000,,is it a b or c Dialogue: 0,0:54:52.74,0:54:55.02,Default,,0000,0000,0000,,share your thoughts without in touch Dialogue: 0,0:54:55.02,0:54:58.38,Default,,0000,0000,0000,,with us in the chat and Dialogue: 0,0:54:58.38,0:55:00.18,Default,,0000,0000,0000,,um and I'm also going to give you some Dialogue: 0,0:55:00.18,0:55:02.94,Default,,0000,0000,0000,,like minutes to think of it before I Dialogue: 0,0:55:02.94,0:55:06.02,Default,,0000,0000,0000,,like tell you about Dialogue: 0,0:55:06.60,0:55:09.00,Default,,0000,0000,0000,,yeah so let me go through the possible Dialogue: 0,0:55:09.00,0:55:12.36,Default,,0000,0000,0000,,answers right so we have a uh create an Dialogue: 0,0:55:12.36,0:55:14.94,Default,,0000,0000,0000,,inference pipeline from the training Dialogue: 0,0:55:14.94,0:55:16.02,Default,,0000,0000,0000,,pipeline Dialogue: 0,0:55:16.02,0:55:19.26,Default,,0000,0000,0000,,uh B we have ADD and evaluate model Dialogue: 0,0:55:19.26,0:55:22.38,Default,,0000,0000,0000,,module to the training Pipeline and then Dialogue: 0,0:55:22.38,0:55:25.08,Default,,0000,0000,0000,,three we have uh clone the training Dialogue: 0,0:55:25.08,0:55:29.48,Default,,0000,0000,0000,,Pipeline with a different name Dialogue: 0,0:55:29.52,0:55:31.56,Default,,0000,0000,0000,,so what do you think is the correct Dialogue: 0,0:55:31.56,0:55:33.96,Default,,0000,0000,0000,,answer a b or c Dialogue: 0,0:55:33.96,0:55:36.66,Default,,0000,0000,0000,,uh also this time I think it's something Dialogue: 0,0:55:36.66,0:55:39.30,Default,,0000,0000,0000,,we mentioned both in the decks and in Dialogue: 0,0:55:39.30,0:55:41.96,Default,,0000,0000,0000,,the demo right Dialogue: 0,0:55:42.60,0:55:44.82,Default,,0000,0000,0000,,yes it is Dialogue: 0,0:55:44.82,0:55:48.72,Default,,0000,0000,0000,,it's something that I have done like two Dialogue: 0,0:55:48.72,0:55:51.80,Default,,0000,0000,0000,,like five minutes ago Dialogue: 0,0:55:51.80,0:55:57.20,Default,,0000,0000,0000,,it's real time real time what's Dialogue: 0,0:55:58.02,0:55:58.76,Default,,0000,0000,0000,,um Dialogue: 0,0:55:58.76,0:56:02.04,Default,,0000,0000,0000,,yeah so think about you need to deploy Dialogue: 0,0:56:02.04,0:56:05.46,Default,,0000,0000,0000,,uh the model as a service so uh if I'm Dialogue: 0,0:56:05.46,0:56:07.98,Default,,0000,0000,0000,,going to deploy model Dialogue: 0,0:56:07.98,0:56:10.38,Default,,0000,0000,0000,,um I cannot like evaluate the model Dialogue: 0,0:56:10.38,0:56:12.84,Default,,0000,0000,0000,,after deploying it right because I Dialogue: 0,0:56:12.84,0:56:14.94,Default,,0000,0000,0000,,cannot go into production if I'm not Dialogue: 0,0:56:14.94,0:56:17.58,Default,,0000,0000,0000,,sure I'm not satisfied over my model and Dialogue: 0,0:56:17.58,0:56:19.50,Default,,0000,0000,0000,,I'm not sure that my model is performing Dialogue: 0,0:56:19.50,0:56:20.28,Default,,0000,0000,0000,,well Dialogue: 0,0:56:20.28,0:56:23.46,Default,,0000,0000,0000,,so that's why I would go with Dialogue: 0,0:56:23.46,0:56:24.32,Default,,0000,0000,0000,,um Dialogue: 0,0:56:24.32,0:56:30.48,Default,,0000,0000,0000,,I would like exclude B from from my from Dialogue: 0,0:56:30.48,0:56:31.52,Default,,0000,0000,0000,,my answer Dialogue: 0,0:56:31.52,0:56:33.60,Default,,0000,0000,0000,,uh while Dialogue: 0,0:56:33.60,0:56:36.96,Default,,0000,0000,0000,,um thinking about C uh I don't see you I Dialogue: 0,0:56:36.96,0:56:39.48,Default,,0000,0000,0000,,didn't see you John cloning uh the Dialogue: 0,0:56:39.48,0:56:41.42,Default,,0000,0000,0000,,training Pipeline with a different name Dialogue: 0,0:56:41.42,0:56:44.64,Default,,0000,0000,0000,,uh so I I don't think this is the the Dialogue: 0,0:56:44.64,0:56:46.92,Default,,0000,0000,0000,,right answer Dialogue: 0,0:56:46.92,0:56:49.62,Default,,0000,0000,0000,,um while I've seen you creating an Dialogue: 0,0:56:49.62,0:56:52.86,Default,,0000,0000,0000,,inference pipeline uh yeah from the Dialogue: 0,0:56:52.86,0:56:55.02,Default,,0000,0000,0000,,training Pipeline and you just converted Dialogue: 0,0:56:55.02,0:56:59.28,Default,,0000,0000,0000,,it using uh a one-click button right Dialogue: 0,0:56:59.28,0:57:03.30,Default,,0000,0000,0000,,yeah that's correct so uh this is the Dialogue: 0,0:57:03.30,0:57:04.28,Default,,0000,0000,0000,,right answer Dialogue: 0,0:57:04.28,0:57:07.46,Default,,0000,0000,0000,,uh good job so I created an inference Dialogue: 0,0:57:07.46,0:57:11.28,Default,,0000,0000,0000,,real-time Pipeline and it has done it Dialogue: 0,0:57:11.28,0:57:13.44,Default,,0000,0000,0000,,like it finished it finished the job is Dialogue: 0,0:57:13.44,0:57:18.00,Default,,0000,0000,0000,,finished so uh we can now deploy Dialogue: 0,0:57:18.00,0:57:19.40,Default,,0000,0000,0000,,ment Dialogue: 0,0:57:19.40,0:57:21.50,Default,,0000,0000,0000,,yeah Dialogue: 0,0:57:21.50,0:57:25.34,Default,,0000,0000,0000,,exactly like on time Dialogue: 0,0:57:25.38,0:57:27.84,Default,,0000,0000,0000,,I like it finished like two seconds Dialogue: 0,0:57:27.84,0:57:30.86,Default,,0000,0000,0000,,three three four seconds ago Dialogue: 0,0:57:30.86,0:57:33.12,Default,,0000,0000,0000,,so uh Dialogue: 0,0:57:33.12,0:57:36.48,Default,,0000,0000,0000,,until like um Dialogue: 0,0:57:36.48,0:57:39.84,Default,,0000,0000,0000,,this is my job review so Dialogue: 0,0:57:39.84,0:57:43.26,Default,,0000,0000,0000,,uh like this is the job details that I Dialogue: 0,0:57:43.26,0:57:45.54,Default,,0000,0000,0000,,have already submitted it's just opening Dialogue: 0,0:57:45.54,0:57:48.12,Default,,0000,0000,0000,,and once it opens Dialogue: 0,0:57:48.12,0:57:50.18,Default,,0000,0000,0000,,um Dialogue: 0,0:57:50.40,0:57:52.74,Default,,0000,0000,0000,,like I don't know why it's so heavy Dialogue: 0,0:57:52.74,0:57:56.78,Default,,0000,0000,0000,,today it's not like that usually Dialogue: 0,0:57:58.74,0:58:01.02,Default,,0000,0000,0000,,yeah it's probably because you are also Dialogue: 0,0:58:01.02,0:58:06.00,Default,,0000,0000,0000,,showing your your screen on teams Dialogue: 0,0:58:06.00,0:58:08.16,Default,,0000,0000,0000,,okay so that's the bandwidth of your Dialogue: 0,0:58:08.16,0:58:10.74,Default,,0000,0000,0000,,connection is exactly do something here Dialogue: 0,0:58:10.74,0:58:13.74,Default,,0000,0000,0000,,because yeah finally Dialogue: 0,0:58:13.74,0:58:16.44,Default,,0000,0000,0000,,I can switch to my mobile internet if it Dialogue: 0,0:58:16.44,0:58:18.60,Default,,0000,0000,0000,,did it again so I will click on deploy Dialogue: 0,0:58:18.60,0:58:20.70,Default,,0000,0000,0000,,it's that simple I'll just click on Dialogue: 0,0:58:20.70,0:58:23.04,Default,,0000,0000,0000,,deploy and Dialogue: 0,0:58:23.04,0:58:25.62,Default,,0000,0000,0000,,I am going to deploy a new real-time Dialogue: 0,0:58:25.62,0:58:27.96,Default,,0000,0000,0000,,endpoint Dialogue: 0,0:58:27.96,0:58:30.30,Default,,0000,0000,0000,,so what I'm going to name it I'm Dialogue: 0,0:58:30.30,0:58:31.74,Default,,0000,0000,0000,,description and the compute type Dialogue: 0,0:58:31.74,0:58:33.72,Default,,0000,0000,0000,,everything is already mentioned for me Dialogue: 0,0:58:33.72,0:58:36.24,Default,,0000,0000,0000,,here so I'm just gonna copy and paste it Dialogue: 0,0:58:36.24,0:58:38.94,Default,,0000,0000,0000,,because we like we have we are running Dialogue: 0,0:58:38.94,0:58:41.28,Default,,0000,0000,0000,,out of time Dialogue: 0,0:58:41.28,0:58:45.68,Default,,0000,0000,0000,,so it's all Azure container instance Dialogue: 0,0:58:45.68,0:58:48.72,Default,,0000,0000,0000,,which is a containerization service also Dialogue: 0,0:58:48.72,0:58:51.06,Default,,0000,0000,0000,,both are for containerization but this Dialogue: 0,0:58:51.06,0:58:52.44,Default,,0000,0000,0000,,gives you something and this gives you Dialogue: 0,0:58:52.44,0:58:54.96,Default,,0000,0000,0000,,something else for the advanced options Dialogue: 0,0:58:54.96,0:58:57.42,Default,,0000,0000,0000,,it doesn't say for us to do anything so Dialogue: 0,0:58:57.42,0:59:00.42,Default,,0000,0000,0000,,we are just gonna click on deploy Dialogue: 0,0:59:00.42,0:59:05.22,Default,,0000,0000,0000,,and now we can test our endpoint from Dialogue: 0,0:59:05.22,0:59:07.86,Default,,0000,0000,0000,,the endpoints that we can find here so Dialogue: 0,0:59:07.86,0:59:11.46,Default,,0000,0000,0000,,it's in progress if I go here Dialogue: 0,0:59:11.46,0:59:13.80,Default,,0000,0000,0000,,under the assets I can find something Dialogue: 0,0:59:13.80,0:59:16.68,Default,,0000,0000,0000,,called endpoints and I can find the Dialogue: 0,0:59:16.68,0:59:18.60,Default,,0000,0000,0000,,real-time ones and the batch endpoints Dialogue: 0,0:59:18.60,0:59:22.02,Default,,0000,0000,0000,,and we have created a real-time endpoint Dialogue: 0,0:59:22.02,0:59:25.26,Default,,0000,0000,0000,,so we are going to find it under this uh Dialogue: 0,0:59:25.26,0:59:29.76,Default,,0000,0000,0000,,title so if I like click on it I should Dialogue: 0,0:59:29.76,0:59:32.64,Default,,0000,0000,0000,,be able to test it once it's ready Dialogue: 0,0:59:32.64,0:59:37.20,Default,,0000,0000,0000,,it's still like loading but this is the Dialogue: 0,0:59:37.20,0:59:40.98,Default,,0000,0000,0000,,input and this is the output that we Dialogue: 0,0:59:40.98,0:59:45.20,Default,,0000,0000,0000,,will get back so if I click on test and Dialogue: 0,0:59:45.20,0:59:49.92,Default,,0000,0000,0000,,from here I will input some data to the Dialogue: 0,0:59:49.92,0:59:50.90,Default,,0000,0000,0000,,endpoint Dialogue: 0,0:59:50.90,0:59:54.60,Default,,0000,0000,0000,,which are the patient information The Dialogue: 0,0:59:54.60,0:59:57.12,Default,,0000,0000,0000,,Columns that we have already seen in our Dialogue: 0,0:59:57.12,1:00:00.38,Default,,0000,0000,0000,,data set the patient ID the pregnancies Dialogue: 0,1:00:00.38,1:00:03.96,Default,,0000,0000,0000,,and of course of course I'm not gonna Dialogue: 0,1:00:03.96,1:00:05.94,Default,,0000,0000,0000,,enter the label that I'm trying to Dialogue: 0,1:00:05.94,1:00:08.10,Default,,0000,0000,0000,,predict so I'm not going to give him if Dialogue: 0,1:00:08.10,1:00:10.68,Default,,0000,0000,0000,,the patient is diabetic or not this end Dialogue: 0,1:00:10.68,1:00:13.20,Default,,0000,0000,0000,,point is to tell me this is the end Dialogue: 0,1:00:13.20,1:00:15.60,Default,,0000,0000,0000,,point or the URL is going to give me Dialogue: 0,1:00:15.60,1:00:17.64,Default,,0000,0000,0000,,back this information whether someone Dialogue: 0,1:00:17.64,1:00:22.68,Default,,0000,0000,0000,,has diabetes or he doesn't so if I input Dialogue: 0,1:00:22.68,1:00:24.78,Default,,0000,0000,0000,,these this data I'm just going to copy Dialogue: 0,1:00:24.78,1:00:27.78,Default,,0000,0000,0000,,it and go to my endpoint and click on Dialogue: 0,1:00:27.78,1:00:30.18,Default,,0000,0000,0000,,test I'm gonna give the result pack Dialogue: 0,1:00:30.18,1:00:32.36,Default,,0000,0000,0000,,which are the three columns that we have Dialogue: 0,1:00:32.36,1:00:35.52,Default,,0000,0000,0000,,defined inside our python script the Dialogue: 0,1:00:35.52,1:00:37.86,Default,,0000,0000,0000,,patient ID the diabetic prediction and Dialogue: 0,1:00:37.86,1:00:41.04,Default,,0000,0000,0000,,the probability the certainty of whether Dialogue: 0,1:00:41.04,1:00:45.72,Default,,0000,0000,0000,,someone is diabetic or not based on the Dialogue: 0,1:00:45.72,1:00:50.66,Default,,0000,0000,0000,,uh based on the prediction so that's it Dialogue: 0,1:00:50.66,1:00:54.36,Default,,0000,0000,0000,,and like uh I think that this is really Dialogue: 0,1:00:54.36,1:00:56.82,Default,,0000,0000,0000,,simple step to do you can do it on your Dialogue: 0,1:00:56.82,1:00:58.38,Default,,0000,0000,0000,,own you can test it Dialogue: 0,1:00:58.38,1:01:01.14,Default,,0000,0000,0000,,and I think that I have finished so Dialogue: 0,1:01:01.14,1:01:03.02,Default,,0000,0000,0000,,thank you Dialogue: 0,1:01:03.02,1:01:06.60,Default,,0000,0000,0000,,uh yes we are running out of time I I Dialogue: 0,1:01:06.60,1:01:09.78,Default,,0000,0000,0000,,just wanted to uh thank you John for for Dialogue: 0,1:01:09.78,1:01:12.30,Default,,0000,0000,0000,,this demo for going through all these Dialogue: 0,1:01:12.30,1:01:14.10,Default,,0000,0000,0000,,steps to Dialogue: 0,1:01:14.10,1:01:16.74,Default,,0000,0000,0000,,um create train a classification model Dialogue: 0,1:01:16.74,1:01:19.68,Default,,0000,0000,0000,,and also deploy it as a predictive Dialogue: 0,1:01:19.68,1:01:23.04,Default,,0000,0000,0000,,service and I encourage you all to go Dialogue: 0,1:01:23.04,1:01:25.08,Default,,0000,0000,0000,,back to the learn module Dialogue: 0,1:01:25.08,1:01:28.26,Default,,0000,0000,0000,,um and uh like depend all these topics Dialogue: 0,1:01:28.26,1:01:31.76,Default,,0000,0000,0000,,at your at your own pace and also maybe Dialogue: 0,1:01:31.76,1:01:34.80,Default,,0000,0000,0000,,uh do this demo on your own on your Dialogue: 0,1:01:34.80,1:01:37.14,Default,,0000,0000,0000,,subscription on your Azure for student Dialogue: 0,1:01:37.14,1:01:39.36,Default,,0000,0000,0000,,subscription Dialogue: 0,1:01:39.36,1:01:43.20,Default,,0000,0000,0000,,um and I would also like to recall that Dialogue: 0,1:01:43.20,1:01:46.26,Default,,0000,0000,0000,,this is part of a series of study Dialogue: 0,1:01:46.26,1:01:49.50,Default,,0000,0000,0000,,sessions of cloud skill challenge study Dialogue: 0,1:01:49.50,1:01:51.06,Default,,0000,0000,0000,,sessions Dialogue: 0,1:01:51.06,1:01:54.06,Default,,0000,0000,0000,,um so you will have more in the in the Dialogue: 0,1:01:54.06,1:01:57.54,Default,,0000,0000,0000,,in the following days and this is for Dialogue: 0,1:01:57.54,1:02:00.48,Default,,0000,0000,0000,,you to prepare let's say to to help you Dialogue: 0,1:02:00.48,1:02:04.88,Default,,0000,0000,0000,,in taking the a cloud skills challenge Dialogue: 0,1:02:04.88,1:02:07.04,Default,,0000,0000,0000,,which collect Dialogue: 0,1:02:07.04,1:02:10.80,Default,,0000,0000,0000,,a very interesting learn module that you Dialogue: 0,1:02:10.80,1:02:14.54,Default,,0000,0000,0000,,can use to scale up on various topics Dialogue: 0,1:02:14.54,1:02:18.36,Default,,0000,0000,0000,,and some of them are focused on AI and Dialogue: 0,1:02:18.36,1:02:20.82,Default,,0000,0000,0000,,ml so if you are interested in these Dialogue: 0,1:02:20.82,1:02:23.10,Default,,0000,0000,0000,,topics you can select these these learn Dialogue: 0,1:02:23.10,1:02:24.78,Default,,0000,0000,0000,,modules Dialogue: 0,1:02:24.78,1:02:27.66,Default,,0000,0000,0000,,um so let me also copy Dialogue: 0,1:02:27.66,1:02:29.82,Default,,0000,0000,0000,,um the link the short link to the Dialogue: 0,1:02:29.82,1:02:32.70,Default,,0000,0000,0000,,challenge in the chat uh remember that Dialogue: 0,1:02:32.70,1:02:34.98,Default,,0000,0000,0000,,you have time until the 13th of Dialogue: 0,1:02:34.98,1:02:37.98,Default,,0000,0000,0000,,September to take the challenge and also Dialogue: 0,1:02:37.98,1:02:40.44,Default,,0000,0000,0000,,remember that in October on the 7th of Dialogue: 0,1:02:40.44,1:02:43.02,Default,,0000,0000,0000,,October you have the you can join the Dialogue: 0,1:02:43.02,1:02:46.62,Default,,0000,0000,0000,,student the the student developer Summit Dialogue: 0,1:02:46.62,1:02:50.64,Default,,0000,0000,0000,,which is uh which will be a virtual or Dialogue: 0,1:02:50.64,1:02:53.22,Default,,0000,0000,0000,,in for some for some cases and hybrid Dialogue: 0,1:02:53.22,1:02:56.04,Default,,0000,0000,0000,,event so stay tuned because you will Dialogue: 0,1:02:56.04,1:02:58.56,Default,,0000,0000,0000,,have some surprises in the following Dialogue: 0,1:02:58.56,1:03:01.26,Default,,0000,0000,0000,,days and if you want to learn more about Dialogue: 0,1:03:01.26,1:03:03.48,Default,,0000,0000,0000,,this event you can check the Microsoft Dialogue: 0,1:03:03.48,1:03:08.10,Default,,0000,0000,0000,,Imaging cap Twitter page and stay tuned Dialogue: 0,1:03:08.10,1:03:11.46,Default,,0000,0000,0000,,so thank you everyone for uh for joining Dialogue: 0,1:03:11.46,1:03:13.08,Default,,0000,0000,0000,,this session today and thank you very Dialogue: 0,1:03:13.08,1:03:16.50,Default,,0000,0000,0000,,much Sean for co-hosting with with this Dialogue: 0,1:03:16.50,1:03:20.36,Default,,0000,0000,0000,,session with me it was a pleasure Dialogue: 0,1:03:21.84,1:03:24.12,Default,,0000,0000,0000,,thank you so much Carlotta for having me Dialogue: 0,1:03:24.12,1:03:26.58,Default,,0000,0000,0000,,with you today and thank you like for Dialogue: 0,1:03:26.58,1:03:28.08,Default,,0000,0000,0000,,giving me this opportunity to be with Dialogue: 0,1:03:28.08,1:03:30.18,Default,,0000,0000,0000,,you here Dialogue: 0,1:03:30.18,1:03:33.48,Default,,0000,0000,0000,,great I hope that uh yeah I hope that we Dialogue: 0,1:03:33.48,1:03:36.48,Default,,0000,0000,0000,,work again in the future sure I I hope Dialogue: 0,1:03:36.48,1:03:38.16,Default,,0000,0000,0000,,so as well Dialogue: 0,1:03:38.16,1:03:40.76,Default,,0000,0000,0000,,um so Dialogue: 0,1:03:44.10,1:03:46.50,Default,,0000,0000,0000,,bye bye speak to you soon Dialogue: 0,1:03:46.50,1:03:48.92,Default,,0000,0000,0000,,bye