Nicely taught the algorithm..Thanks for making learning simple

fit_transform() is used on the training data to learn the scaling or transformation parameters and then applies the same transformation to the training data. transform() is used on new data (e.g. test data) to apply the same transformation that was learned on the training data.

@ritikkohad5045

Жыл бұрын

Thanks brother

@ayushpant8290

11 ай бұрын

thanks bro

@kartiknampalliwar8603

4 ай бұрын

i am unable to do first step i.e load_ boston is showing error can you please help me

@dhmahbub

3 ай бұрын

@@kartiknampalliwar8603 that data is not available in the new version. You can alternatively use "fetch_california_housing" and load it. Probably that is the similar sort of data.

@hiteshkotian6023

2 ай бұрын

@@kartiknampalliwar8603 load_boston is no longer available use some other data like load_diabetes or something

fit transform is used in train data set to predict the value(linear regression) test data set just to se our accuracy with the model.

Sir you explain soo good please continue making videos in hindi

Sir , your effort is really wonderfull and is inspiration. please make a separate playlist for EDA and feature engineering , lakhs of aspirants are wait , please make it on serious note.

Please more videos on machine learning also practical video more Thankyou

why are we using 'neg_mean_squared_error' can u please share link linear regression loss function video ?

fit_transform is used on training data to learn parameters and transform it, while transform is used on new or unseen data to apply previously learned transformations without re-learning the parameters.

Great video sir...just one concern...Why we are not checking VIF?

Very good Sir

what is random state= 42 in that train test split command?

Sir i didnt able to understand what is y the dependent variable . I mean which column is gettimg predicted ?

great job sir jee

here linear regression doing but why taken independent features more than 1 feature can anybody tell me

I am seeing you videos just to similary apply another multivarite problem but when I got the displot(with kind=kde) it came similar but of the rang eof the 1e^9 so How can decrease the error should I use the tunning or what ?

Krish ap kon sa video software use kerto ho recording k liye

sir sklearn na dataset remove kar deya ha. dataset fetch nahi ho raha ha

Good video sir.

sir if my accuracy_score is 0.85 then my predication model is good or bad?

Sir when new batch start for data science?

Do I need to know sklearn before starting this playlist ?

where is the theory playlist? can someone please attach the link in the reply to this comment

Gradient decent iss implementation mein kaise implement kaise hua?? Agar back end mein hua toh alpha value kaha diya?? @krish please explain

kind='kde' not showing that graph

How is it that you are predicting on x_test but calling your y_test as truth value?

nice video sir

Please keep uploading videos in hindi

Simply amazing ❤

Thank You

Thank you sir 🙏

Can anyone please explain displot discussed in this video?

Sir this can be explain in English language some what difficult to understand Hindi

Sir please l1 and L2 k liye bhi video banaiye

The Boston datasathas been removed from sklearn....

what is score?

playlist ke hisab se video dalo sir, theory aur practical implementation ke video mein bahot fark hai , much samjh nhi aya, r2 score, cross validation , xtrain ye sab kya hai theory mein to that he nhi ye sab.

@Techwithgalvan

4 ай бұрын

yoo bhai

bcoz we evaluate our model on test data set

fit transform and transform = we want to keep as a surprise is no longer unknown to our model and we will not get a good estimate of how our model is performing on the test (unseen) data which is the ultimate goal of building a model using machine learning algorithm.

@patelkdhawal

Жыл бұрын

Can you elaborate more. Please

Sir in this where us accuracy

*** please create EDA and Feature engineering playlist in HINDI ***

What value of MSE , RMSE, R-square should be taken into consideration to come to conclusion that model build is accurate one? Is there any range of value for MSE, RMSE and R-square ?

@aftab6417

Жыл бұрын

R2 to be gerater than .70 that is 70%

Boston housing dataset has been removed from scikit-learn. Is there any way to load it as a bunch data??

@ubarbde

5 ай бұрын

from sklearn.datasets import fetch_california_housing housing = fetch_california_housing()

@user-ns2xy1uw6d

2 ай бұрын

from sklearn.datasets import fetch_california_housing use this

Sir i have build the model in linear regression and performance of evaluation metrics are also done. Now additional I want to add one more new row(instance) and find the performance of it how to do can you guide me pl. How to check the performance particularly that single row.

@filmybeats4827

8 ай бұрын

Bro I couldn't understand this to that level how can I understand these concepts as sir is directly implemented it so

problem with the inbuilt datasets is that they are removed often later when new versions is released, like Boston dataset is no more part of sklearn in-built dataset. it would be great if you use independent datasets in future.

@ayushmangupta6682

10 ай бұрын

There is any way to download the boston dataset with the dependent feature I want to implement the linear regression if there is any way please suggest

@shivanshmishra8395

9 ай бұрын

Ya its a problem i am facing as well

@EjazUddin786

3 ай бұрын

​@@shivanshmishra8395 have got solution how to import please let us know to help

@AbhijeetDewangan-gr9sj

3 ай бұрын

@@ayushmangupta6682 please use " from sklearn.datasets import fetch_california_housing " alternative of Boston

But what are we predicting here? Can someone explain please..what does the values in "reg_pred" tell us? what is the difference between values in target features array and "reg_pred" values?

@NavyaTrilokD

2 ай бұрын

so we are predicting the output feature house pricing...for the independent features in x_test, dependent feature or actual values are in y_test. After applying linear regression, predicted values are in reg_pred. In linear regression we find the difference between actual values and predicted values, that is the error. MSE is that error here.

Sir the boston dataset is no more available in the scikit-learn datasets also can't load the boston dataset in juyter notebook can U please provide any solution for that?

@innocentgamer8670

4 ай бұрын

you can use the alternate dataset like california housing , or you can search and save the boston dataset , and use pd.read_csv() method to use that dataset

@AbhijeetDewangan-gr9sj

3 ай бұрын

please use " from sklearn.datasets import fetch_california_housing " alternative of Boston

sklearn removed load_boston which dataset i can use to follow along?

@krishnaikhindi

Жыл бұрын

Fetchcaliforniadataset

@dnswm95

9 ай бұрын

@@krishnaikhindican we use fetch_california_housing ??

How to know model is overfitted or stable model.

@entertainmenthub7700

Жыл бұрын

if it perfectly fits to the traning data in simple meanings if it remember the data instead of learning it overfits

Why Y_train is not standardized? Please answer

@satyamghule147

9 ай бұрын

Standardization is typically applied to the feature variables (X_train) rather than the target variable (Y_train) in machine learning.

subscribed

i use the same model on 'fetch_california_housing' dataset and the mse i got is 0.5.

@shivanshmishra8395

9 ай бұрын

Me too.. and the score is 0.34

@bhuvansagar3294

7 ай бұрын

same, and as mentioned by the other person, my score is 0.33

Kuch samajh nahi aa raha...

boston dataset is been reomved from the kcikit liberary

@azharafridi9619

11 ай бұрын

which one do you use right now. facing the same problem bro

@aishwaryajadhav8736

11 ай бұрын

same issue @@azharafridi9619

sir ap dataset bhi dal diya kro. load_boston to ho ni rha hai hmara. kaise kre ab

@niharikakhanna3585

4 ай бұрын

same problem, kuch solution mila?

@AmitSharma-oh5uw

4 ай бұрын

@kakhanna3585 hi niharika are you a data science student. pehle mene socha inke 38 videos hi hai machine learning ke. and me bhut jldi complete kr lunga. but me 2 video se age hi bdha hi ni abhi tk. ye beginers jaise ni pdha rhe hai. and mujhe ek ek chiz likhna pd rha hai, ki sir kya bol rhe hai video me. and wo atleast definations bhi likhwate to smjne me easy hota. it's difficult to understand. kya apko koi aur playlist pta hai. jisse jo machine learning ke liye ho.

i got a score of "0.017460452225004253" why i got low score?

@himanchalpawar167

Жыл бұрын

Same prob

too complex🙁

Noice

Plz do videos in English🥲

@adila8128

9 ай бұрын

If you're looking for videos in English you can refer to his other channel. You will find all the videos in the English language.

bouncer ho gya ye video

load boston has been removed

@AbhijeetDewangan-gr9sj

3 ай бұрын

please use " from sklearn.datasets import fetch_california_housing " alternative of Boston

@user-iq1yw9bz6f

3 ай бұрын

Bhai fetch wali BHI Nahi Chal rahi

Esi video bnaya naa kro jo kisi k leptop m work naa kre kya jya oerte ho smj nhi atta padhne bethq toh sara dinak khrab ho gya kuch hua nhi

Great thank you

sir aap JSPM Tathwade ke student hai kya

youtube shanel

Boston housing dataset has been removed from scikit-learn. Is there any way to load it as a bunch data??

@ausking3081

6 ай бұрын

import pandas as pd import numpy as np data_url = "lib.stat.cmu.edu/datasets/boston" raw_df = pd.read_csv(data_url, sep="\s+", skiprows=22, header=None) data = np.hstack([raw_df.values[::2, :], raw_df.values[1::2, :2]]) target = raw_df.values[1::2, 2] or install version 1.0.1 pip install scikit-learn==1.0.1