Hindi Machine Learning Tutorial 10 - Decision Tree

Decision tree algorithm is used to solve classification problem in machine learning domain. In this tutorial we will solve employee salary prediction problem using decision tree. First we will go over some theory and then do coding practice. In the end I've a very interesting exercise for you to solve.
#MachineLearningHindi #PythonMachineLearning #MachineLearningTutorial #Python #PythonTutorial #PythonTraining #MachineLearningCource #DecisionTree
Code: github.com/cod...
To download csv and code for all tutorials: go to github.com/cod..., click on a green button to clone or download the entire repository and then go to relevant folder to get access to that specific file.
Website: codebasicshub.com/
Facebook: / codebasicshub
Twitter: / codebasicshub

[@poojabehera8675]

sir, please make separate videos on Info gini, Gini Impurity & Information gain

[@muhammadjaffarrazadar967]

waiting for you to upload Deep learning tutorials in hindi

[@codebasicsHindi]

Sure Muhammad, I have noted this down in my Todo list.

[@muhammadjaffarrazadar967]

​@@codebasicsHindi Desperately waiting.

[@ajaykushwaha4233]

Sir aap itna accha explain kerte hain ki kya batayein. Sab kuch samajh aa jata hai.

[@codebasicsHindi]

ajay bhai, dhanyavad.

[@DEV-un3ok]

For NaN(missing values) values I calculated Mean of Age column and replaced the NaN values in the Age column with the mean of Age column and when I split (train_test_split method)the data I got 80.44 % accuracy but when I did not split the data I got 97.41 % accuracy. But when I visualise my Decision tree then the one with the train_test_split method looks more accurate so which one is correct?

[@Gemreviewer]

Hello sir, just wanted to confirm that, in the 4th step in the Jupiter notebook, instead of creating 3 objects for Label Encoder, we could have created just one object and the used it for the 3 columns. Was there any particular reason to use it 3 times? Please let me know

[@niku237]

Sir... ANN algorithm bhi kra dijiye hindi mei... hmare liye kaafi benefit hoga..

[@rushikeshdargeai5154]

97.97% Thank you!!!

[@ashishmachcha8626]

Can you make more video of machine learning in python in hindi....it is very helping me.

[@ashkraze]

i read that LabelEncoder should only be used for dependent variables only , can we use it for independent values also??

[@farman4164]

Make a video on Deep learning pls..

[@nidhivishwakarma1339]

bro you are just amazing..

[@kainaatmakhani6550]

good job.

[@Ishowmeat8]

Everyday i visit this channel and i watched videos after liked it .

[@abhasharma9455]

Sir according to initial 1st minute how can we check data is distributed like this so that we will apply DT. Thank u very much Sir for great lectures.

[@govind1937]

sir firstly i did without splitting into train and test, then it gave score 0.9777 and after splitting it gives 0.822222 taking .05 as test_size. and there is a strong correlation between Pclass and Fare, so we can use only one.

[@varshaasabe1192]

Hello sir, i am seen all your vedio and thanks for that , I understand but I am not understand how import csv file in R. Can me explain I am doing data science. T

[@mazharbukhari786]

Sir, one question, from where you learnt?

[@VishalMalik2023]

my model predict without using train_test_split so they predict score is 97% , then i using train_test_split with test_size = 0.2 then they predict score is 82%

[@abhasharma9455]

Sir u explain great.

[@codebasicsHindi]

Glad it was helpful!

[@ashwiniawati2297]

Hi Sir, you are doing really a great job. Kudos to you 🙌. I just solved the Titanic exercise with inputs Age, sex, Pclass and Fare. I got my score 0.75. Please advise about the score. Is it recommended to choose single column as an input and then check the score?

[@owais20017]

please send source code

[@sagar8460830871]

how to drop variable please explain correlation and p-value. i am totaly confuse

[@debdyutimohanta9007]

There are lots of Nan values in age column

[@apnadesicricket]

Titanic program ka score aaya 0.98765 is it correct please tell first time i done the program from my own

[@codebasicsHindi]

IT is correct

[@apnadesicricket]

@@codebasicsHindi thanks so much Sir, watching your videos great help for me, then i understand the ML in simpler way. Thanks again.

[@neerajrokaya1132]

Sir, I got 81.6% with test_size = 0.25 and random_state = 10 . I checked with your github repo and found that you have taken mean to fill the nan value in age column and got ans as 79.3%. I did the box plot and found out that there are many outliers. So, will you please explain in short about the mean and median selection.

[@data_science_06]

mean is used for when data is normally distributed... median is used for when some outliers are existing in data..... mod is used .... when your data has category..... like:- 0,1,2,0,1,1 and male , female and so on....

[@akshatsingh6036]

sir tensorflow kb aaegi

[@ashmitsingh5473]

I got 1.0 is it fine..??

[@me_debankan4178]

I have tried the exercise and each time i am getting more accuracy using Logistic regression than decision trees

[@as-it-is6627]

Wow, what an explanation.

[@poojabehera8675]

why my fit.transform() throwing error as: transform() missing 1 required positional argument: 'y' I have written corect code

[@fooglefirst9738]

fit_transform() ye wala correct karloo aap ko aagayegaa

[@oldisgold5803]

mine too is throwing the same error..... Did you find way out of it?

[@DEV-un3ok]

df['column'] = column.fit_transform(df['column'])

[@anantsharma5330]

I am getting 98.59% Model Score. I have Label Encoded all columns of inputs. Is it wrong?

[@manu-prakash-choudhary]

you have probably not used train test split method

[@prathameshmore5262]

Got 0.97 score on titanic dataset using decision tree

[@DopeGaming719]

got a score of 79.10 is it ok??

[@preetiranichaudhary9758]

my score is 0.7867(input column is pclass and sex and target variable is survived) then i got score is 0.7867..is it correct or not???

[@codebasicsHindi]

yes it is correct. 78% is a pretty good score

[@preetiranichaudhary9758]

@@codebasicsHindi thanku sir 4 answering

[@dheerajyadav3675]

@@codebasicsHindi can i have it's code?

[@mashalamjadkhan9351]

Sir please share csv file argent

[@DharmendraKumar-pf4fs]

My score is 0.977 is this correct for Titanic exercise .???

[@codebasicsHindi]

Yup.97.7 is a pretty good score. Good job 👏👍

[@karmakarishan56]

result is 1

[@anilkumarsharma8901]

Apney subscribe waley ko artificially intelligent ka support dilva do phir Duniya following karegi

[@ajmeershahtricks6666]

sir i got 1.0 score (accuracy) but im confused that what does it means ? is this the accuracy in %?

[@mohammedmouizuddin4403]

1.0 means 100% accuracy rate and 0.98... , 0.95 ... means 98% , 95% accuracy rate ....

[@ajmeershahtricks6666]

@@mohammedmouizuddin4403 thank you bro❤

[@owais20017]

please @Ajmeer Shah tricks send me sorce code

[@RahulVerma-ud1rr]

:D 97.75% accuracy

[@kamilazim5498]

For Titanic Dataset Test_size = 0.3 model.score = 0.8059701492537313

[@darshankachhadiya8014]

My model score is 0.81

[@gaurangsharma5428]

My score is 0.997755310886644

[@uddiptasaikia3121]

Sir , after coverting "male" and "female" to 1 & 2 respectively , it is not showing 1 and 2 while i type inputs.head()...it is showing NaN....please help.

[@uddiptasaikia3121]

@codebasicsHindi

[@NationalistNewsNetwork]

How it behave on irrational input. pls reply

[@waidapapa1514]

If we see the actual documentation of scitkit learn of label encoder we will know that label encoder ia used for target or output variable only.

[@izainonline]

Many Thanks finally I found the clear guidance for example how and why transform the data into numeric and how to train,Inplace?,drop or add column reason. Thanks again

[@sanjayrajbanshi7698]

I got 80% accuracy by doing train_test_split method (Changed null value in 'Age' column and Converted 'Sex' column into dummy variable column).

[@musakhan9779]

Why is Decision Tree a regression model if it classifies. Isn't Regression model expected to have continuous data and prediction.

[@PankajVerma-p8y]

my mode score is 98% .. Really amazing videos sir. Aap jo samjhate ho an maza aa jata hai and concept clear ho jata hai...

[@subhamsaha2235]

Sir aapne to isme decision tree banaya hi nhi????????????????

[@shivam7164]

some values in age are missing.

[@sudarshandev6369]

lajwaab sir awesome explanation

[@mashalamjadkhan9351]

please share the csv file

[@rockfighter9974]

Really very nice video

[@shahidmalik6107]

very well explaind

[@soumyaprakashdasmohapatra7717]

got a score of 74.8% is it correct?

[@muhammadjaffarrazadar967]

I tried to do it by my own and did it with 0.9797979797979798 test scores. but I didn't bread the data into test and train. maybe that's why.

[@codebasicsHindi]

Yes splitting data in test and train is important otherwise your model is biased and you get a high score

[@shivanshsingh5555]

0.82