Introduction to Machine Learning Project Workflow | #10 of 28 | Foundations of ML: The Big Picture скачать в хорошем качестве

Introduction to Machine Learning Project Workflow | #10 of 28 | Foundations of ML: The Big Picture

In this one, let's understand how a typical ml or data science project workflow typically looks like. To understand this, we will look at it from two different perspectives. One is a minimalistic version, a minimalistic version. Complete Machine Learning Course for FREE:    • Foundation of Machine Learning (The Big Pi...   🔹 Introduction to Machine Learning Project Workflow (Step by Step) 2022 We will first look at this, we will understand this first. And then we will go into a more detailed version as if you are actually implementing a machine learning project. And this sort of are the various activities that happen in a real company also, to implement a data science project. Let's look at this in both perspectives. To understand this, let's take a real project we will go through the steps based on this particular real project. Now, the problem statement that we're looking at this you are working for a real estate brokerage firm, you're working as a data scientist. Now this firm is facing a problem. The problem is the sellers who are listing their properties on their hotel, the sellers often quote an inflated price at the beginning for their respective properties. And once they quote, the properties don't sell initially as it is, and then after a certain point in time, they bring the price down to a more realistic level, and then the sales happen, right. Now to make a solution for this to make the sales happen a little bit more faster and make things more realistic. The company wants to build an AI driven tool that the sellers can use to predict a fair quote for their property should be a fair quote, it should be realistic. By providing this tool or a utility, the company wants to provide this tool to the sellers to the various different people who are listing their properties on the portal, they want to make it available to them, so that a realistic code is placed on the portal and the sale happens faster basically. Now, this is the project. So basically, this form wants you to build a Prediction Engine. This Prediction Engine is going to be the workhorse that is going to give the dollar value of a given property. Right this will be shown to the users to this Prediction Engine, you are going to give certain set of inputs. And this is shown to the users. User gives the input in this case. Alright, and your job is to build this engine, what are the typical basic steps involved. Now, let's look at the minimalistic workflow, just three steps, the first step is you will want to collect the data that you are going to require to build that engine data collection that is the first step. Once you collect the data, you train your machine learning model using a machine learning algorithm. This is the second step, just train your model. Once your model is trained, you are going to deploy it somewhere. And you will work probably with the software engineering team to make the solution be available to the users. So you deploy your machine learning solution, just three different steps. Now in this particular case, we are making the engine available to the users in the front end of the website, right available to the end users who are the customers through the front end. Let me know in the comments section if you have any questions! If you enjoyed this video, be sure to throw it a like and make sure to subscribe to not miss any future videos! Thanks for watching! #machinelearningplus #python #machinelearning #datascience