In AI, data correction, also known as data cleaning or data scrubbing, refers to the process of identifying and correcting errors, inconsistencies, and inaccuracies in datasets to improve their quality and reliability. This process is crucial because AI algorithms are trained on data, and if the data is flawed, the AI models will produce inaccurate or biased results.
Category
ЁЯУЪ
LearningTranscript
00:00So, now we are finally talking about six steps, which is the iteration or experimentation.
00:09You can see that we have a total six steps framework, which is the problem defined.
00:17You can see all the steps that you can see in the last lecture.
00:23Now, when you have a fairly good result, then you have one option that you can see that you can see the data and model with some experimentation.
00:36So, if I can improve this as far as you can see.
00:39If I can improve this as far as you can see,
00:43you can see the model of two aspects.
00:45Either you can see the accuracy or the computational cost.
00:50I can see the accuracy and computational cost.
00:56So, you can see the two things.
00:59Either you can see the computational cost or the accuracy.
01:03And then you can see the balance, compromise.
01:06So, this is basically the total process.
01:09This is the total process.
01:10In which you have to see the six steps and repeat the process.
01:13And the accuracy of your model of your accuracy.
01:16Now, you have to see that we have a whole framework and different components.
01:23So, this was the framework.
01:25Now, we have to have some tools.
01:28We have to use this framework to implement this framework.
01:35team and learn them.
01:37Which means that we have to run the market.
01:39In which we have,
01:41anaconda,
01:44this is anaconda.
01:45This is anaconda.
01:47I am powiedzia┼В,
01:48aaconda.
01:49Anaconda.
01:51Is anaconda.
01:52Anaconda.
01:53Anaconda.
01:55Of course,
01:58the idea is called anaconda.
01:59Which means josimo.
02:03So, I'm going to tell you how to do the tools.
02:05I'm going to tell you how to do the tools.
02:07So, I'm going to tell you how to do the tools.
02:10So, we can see how to do the tools.
02:13Let's make it more short.
02:23Okay, so I have to resize this image.
02:24Now, I'm going to give you the framework.
02:26I apply this tool to do it.
02:29Now, I'm going to break down.
02:31this whole framework, this whole framework, we apply to anaconda, anaconda, and it's a
02:42idea called jupiter notebook, jupiter notebook, okay, this thing is done, now in this
02:52case, the data analysis, evaluation, and featuring, this total work, these three steps, we will
03:02do libraries, which are three libraries, the first is panda, the second is mat, plot, library,
03:14lib, lib, it's called 2, and the third is numpy, which is basically data evaluation, numpy,
03:23this is the next feature, which is modeling, modeling, which is machine learning, which
03:34matters, tensor flow, tensor flow, it's called pi torch, pi t o r c h, and then we have
03:45scikit learn, s k i t l e a r n, so we have to fix it in our class, okay, so we have
03:54tools that we use to enable data visualisation, and we will use this tools for the model,
04:00in ьДЬed link рд▓реА рд╣рдордиреЗ рд╣реИ this tool use рдХрд░рдиреЗ рдФрд░ рдпрд╣ рд╕рд╛рд░рд╛ рдХрд╛рдо рд╣рдордиреЗ рдХрд┐рд╕рдореЗрдВ рдХрд░рдирд╛ рд╣реИ anaconda good?
04:07рдмрд╛рдд- рдЪрд▓рд┐рдпреЗ рдЖрдк interesting рд╕рдмреНрд╕рдХреНрд░рд╛рдЗрдм рдХреЛ рдпрд╣ рдмрджрд╛рддрд╛ рд╣реВрдВ рдЖрдкрдиреЗ рдкреНрд░рд┐рд╢рд╛рди рд╣реА рд╣реЛ рдирд╛ рдХрд┐ рдпрд╛рдИ
04:12рдЗрддрдиреА рд╕рд╛рд░реА рд▓рд╛рдЗрдмрд░реЗрд░реА, рдЗрддрдирд╛ рд╕рд╛рд░реЗ рдЯреВрд▓ рдЕрдкрд▓рдм рдпрд╣ рдкрддрд╛ рдирд╣реАрдВ рд╣рдореЗрдВ рдХрд┐рд╕реЗ рдкрддрд╛ рдЪрд▓реЗрдЧрд╛
04:16we don't need to know everything about everything
04:18You don't need to know everything about everything.
04:24We need to help out our goals to achieve our goals.