Project 15. Titanic Survival Prediction using Machine Learning in Python | Machine Learning Project

Hi! I will be conducting one-on-one discussion with all channel members. Checkout the perks and Join membership if interested: / @siddhardhan Check membership Perks: / @siddhardhan
. This video is about Titanic Survival Prediction using Machine Learning with Python. This is one of the important and standard Machine Learning Projects. For this Project, I have used Logistic regression model.
All presentation files for the Machine Learning course as PDF for as low as ₹200 (INR): Drop a mail to [e-mail]
Enroll at One Neuron to learn from 100 courses in one subscription with 5% discount: courses.ineuron.ai/neurons/Te...
Machine Learning Projects Playlist: • Machine Learning Projects
Machine Learning Course with Python Playlist: • Machine Learning Cours...
Hello everyone! I am setting up a donation campaign for my KZbin Channel. If you like my videos and wish to support me financially, you can donate through the following means:
From India 👉 UPI ID : siddhardhselvam2317@oksbi
Outside of India? 👉 Paypal id: [e-mail]
(No donation is small. Every penny counts)
Thanks in advance!
Let's build a Community of Machine Learning experts! Kindly Subscribe here👉 tinyurl.com/md0gjbis
I am making a "Hands-on Machine Learning Course with Python" in KZbin. I'll be posting 3 videos per week: Monday Evening; Wednesday Evening; Friday Evening.
Dataset file: www.kaggle.com/c/titanic
Colab File Link: colab.research.google.com/dri...
Download the Course Curriculum File from here: drive.google.com/file/d/17i0c...
LinkedIn: / siddhardhan-s-741652207
Telegram Group: t.me/siddhardhan
Facebook group: groups/49085... Instagram: / siddhardhan23

[@ehtishamkhan421]

One of the finest explanations on the internet on how to replace missing values.

[@Siddhardhan]

thanks 😇

[@rong-hongjhuo3911]

Very useful course ,hoping there will be more and more project cases

[@sonal5051]

Such clear cut explanations.. keep up your good work .. much appreciated 👍

[@Siddhardhan]

thanks a lot for your positive words 😇

[@prateekbhatt5905]

Thanks a ton for the video! One question though, why did we not apply feature scaling before training the dataset?

[@siyays1868]

👏🙏👍The Best , indepth , woderful explaination I've ever seen. Basics Vey much clear. & this is also something i was surfing on youtubee. THANK U so much sir for such a amazing session. I was looking for basics of machine learning project model evaluation part & finally got here & i m so looking forward for such amazing videos. I bacame more confidant about projects ,n model evaluation n training part of a machine learning. Thanku sir once again!

[@Siddhardhan]

😇😇

[@piyushpalsania7685]

Thank you Siddhardhan for the free resource

[@gopikrishna4368]

Now I understood very clearly Thanks Mate

[@Siddhardhan]

You're welcome!😇

[@sejal1996]

Well explained ! Very helpful video.

[@ApnaVlogs-tj7do]

Thanks Siddhardhan, wonderfull session

[@funlearnwithdishu5737]

Thanks for the easy explanation

[@firenoms7809]

Bro thanks to you I could do this.Hope you'll be posting good vedios for us to learn.

[@user-gn6ld5to8m]

Im grateful to you.. thank you so much... ❤

[@navaneethdas8855]

if we use mean for filling in missing values of age the data type will be float. is there any problem with that or can we use the mode

[@growingfire]

Thanks a lot !

[@stargoonmiracleousworld1251]

Thank you so much sir✨

[@bramuelaswani3046]

This helpful. Thanks

[@saketml8676]

I built the complete model myself, thankyou sir, great Explanation

[@Siddhardhan]

Great to hear 😇 all the best for your learning journey!

[@karthiksrinath6558]

Bro can u share me that ...i need to project

[@suvigyasharma3456]

Have you done the predictive system?

[@AnimeManiaa]

When giving data to predict whether a person survived or not , should we include name and ticket data ??

[@yaminigupta1696]

Thank you !

[@ranjanasacademy4941]

You are doing a great service by sharing your knowledge..thank you very much..keep doing and keep inspiring..

[@Siddhardhan]

Thanks a lot for your positive words 😇

[@user-yz4wv9ij4n]

Excellent!

[@subbumv338]

my struggling to solve this titanic survival problem is relieved with your simple and detailed explanation thankyou very much bro. when will you start machine learning MODELS explanation series??? reply plz

[@Siddhardhan]

thanks a lot for your positive words 😇. after mathematics module, there is another module about model training, model selection, etc. after that I'll start with model explanation videos.

[@abhisknowledge5514]

Nice video sir.i have doubt regarding the alogrithm so can i apply random forest algorithm on the dataset

[@mandadirohitreddy8086]

Wonderful job bro

[@lakshmiroja7450]

Hi sir i getting an error of sns that is like sns name is not defined what to do please reply sir

[@mohammathuisath4952]

Excellent 👌

[@TheDevimata]

Hi, So you have replaced missing value with 'S' in embarked column and with mode()[0]. My question is will the mode index is always 0? or will it change? Can I chnage it to 1 or any other number instead of 0?

[@micheleadriaans6688]

The mode index will always be 0

[@archismanbanik6477]

Please can you understand the rainfall prediction using linear regression...

[@devilnk8475]

thank you

[@SampathDiss]

Thank you

[@huzaifakhan4076]

i used the other train dataset from kaggle and follow these step. The accuracy score computed by both test and train model is 1.0

[@kateeileen6840]

Thanks Very much Sir, i have some questions

[@stutysijaria8833]

Hi, I have a question regarding the Ensemble techniques- I am getting different importance weight on the same set of features with Gradient Boost and XGBoost? Why could it be?

[@Siddhardhan]

hi! xgboost is a regularized form of gradient boosting. so there is a difference on how they operate. try to learn the theory of these models and how they work. that will help you to understand it better

[@stutysijaria8833]

@@Siddhardhan thanks.

[@DhikshaShrivruthiG]

wonderful explaination sir but i have a small doubt why cant be use command df.dropna(inplace=True) for handling missing values

[@gilangvirgiawan2780]

brother thanks youuu

[@bobby_6455]

it would have been great if you have shown how to submit it to kaggle, I'm kinda lost in there

[@arahmanniazi3241]

Please complete the project or make sure to mention something that a beginner who wants the project to be pinned in the kaggle dropbox knows that he would not be able to watching this video

[@sachinvithubone4278]

Can we drop the Emarded column if that we are not using for our production?

[@Siddhardhan]

yes. we can drop it.

[@meshmeso]

Best ❤

[@gkakash4789]

Super bro👍👍

[@Siddhardhan]

thanks 😇

[@koppuravurinavya3528]

Post videos for all models sir to get better understanding regarding project i.e. Titanic survival prediction

[@Siddhardhan]

sure! I ll make a separate module for ML models

[@mamidisettyprathyusha]

Kindly post any timeseries analysis project

[@user-yw3kh3lt6c]

I am new to data analysis and have a question about this video: you trained and tested dataset and it showed 80% accuracy rate. However, how can I use this to predict survival chance based on specific parameters. For example, what is a chance to survive for Male 25 years old from First class?

[@ajaypanigrahi5603]

I think it is not possible to tell exactly the probability of someone surviving however for every prediction the chances of it being correct is 80% , Which actually answers your question. This is my understanding so take it with a pinch of salt

[@rishabhsharma2776]

Has anybody done the final predict function? I am not able to do that. Sir, can you please help? can you add it in your collab link? if yes, please share your code or link to the code. Thanks! Wonderful explanation sir. This is the first project I'm doing in machine learning and I understood quite a bit. Thank you so much sir

[@suvigyasharma3456]

Did you do it? I am having same issue

[@srirammurali8224]

can anyone tell me, how to find SURVIVED count of MALE from eany of PCLASS using countplot?

[@alibagheri391]

I wrote my code exactly like yours (in Jupiter notebook). When I wrote "model.fit(X_train, Y_train)" it give me this error: float() argument must be a string or a number, not 'method'. Can you please explain me why I get this error?

[@Siddhardhan]

hi! go through this: stackoverflow.com/questions/41256626/pandas-typeerror-float-argument-must-be-a-string-or-a-number

[@Megness]

@@Siddhardhan Sorry but I still dont understand how to resolve this

[@subhasreebanerjee98]

My accuracy for training data is 95.5% and testing data is 79.3% . Is it good or bad ? If bad how can I rectify that?

[@nguyenhuyhoangk18hcm37]

You need to add the correlations between the predict attributes and other attributes.

[@YellowHeart-tw8eo]

39:33

[@akshatsrivastava7771]

Why i have u taken Logistic Regression? Any specific reason????

[@RaymondStormbl3ssed]

When i enter "titanic_data = titanic_data.drop(columns='Cabin', axis=1)", it is giving KeyError: "['Cabin'] not found in axis" What am i doing wrong here?

[@YellowHeart-tw8eo]

@A A Ron when these errors occur, ('not found') just run by 'restart the kernel' beside the run option

[@shreyaaravind1169]

Can i get solution for prediction of infection type

[@sachinvithubone4278]

Small daubt we are finding person survived or not survived, so this a classification problem.. right?

[@Siddhardhan]

hi! yes it is.

[@TheDiary35]

how to submmit this to kaggle????

[@manjumankoji1448]

Please upload project with feature selection comparing 2 to 3 algorithms

[@Siddhardhan]

hi! sure. will do in the future.

[@rathodgaurav1]

Hi , You have not done any feature engineering .....then the training model would be something different....?

[@Siddhardhan]

hi! I try not to do everything in a single video... It would be hard for beginners and intermediate learners... I'll definitely cover about it in my upcoming project videos...

[@rathodgaurav1]

@@Siddhardhan okay......thanks for replying......

[@drake3745]

I am facing an error in titanic_data['Embarked'].value_counts() saying:= TypeError: unhashable type: 'Series' It would be of great help if anybody could point out the mistake.

[@koustavbarik2129]

Most probably while transforming the dataframe, you missed a square bracket somewhere [ ] and it has become an array instead of dataframe. Its almost impossible to guess without seeing the code though. What it actually means is that your column Embarked is no longer countable.

[@chandrakantapatel9406]

Why did'nt he removed outliers??

[@yogeshwarkethepalli4234]

RecursionError: maximum recursion depth exceeded while calling a Python object X = titanic_data.drop(columns = ['PassengerId','Name', 'Ticket', 'Survived'],axis = 1) Getting the above error message ,what is the solution

[@AliHassan-cu9bi]

Bhai chat gpt pr paste kro

[@ramkoni178]

bro the video is very helpful but i received both score are 1.0 what to do now

[@divyachandel91]

airlines crash prediction using ml can you help me with this question

[@Ankursinghblogs]

I m getting the accuracy is 1.0 Is it Right??

[@sachinvithubone4278]

In Column Sex and Embared.. we are moving char to number right, so this process called as label encoding. Please correct me if I m wrong.

[@Siddhardhan]

yes. it is correct. but we refer the target variable as labels. in this case, survived column is the target variable. so I haven't mentioned it as label Encoding. but the process we are doing is the same.

[@sachinvithubone4278]

@@Siddhardhan thanks for confirmation.

[@loicturounet6533]

How do you know logistic Regression will be the best model? (and not random forest for instance)

[@Siddhardhan]

hi! that's because logistic regression can find the difference between two classes. I meant logistic regression is best for binary classification. and not for all the cases. you can understand this Better once we start with Machine Learning models videos

[@naufalgunaningrat9719]

Next question: how to submit on kaggle?

[@amitkumar5602]

Display is not clearly visible.

[@prachijain133]

What will be the code for confusion matrix?

[@Siddhardhan]

Hi! That will be my next video

[@q2success352]

Dear Sir, after running the command: sns.countplot('Sex', data=titanic_data) or sns.countplot('Survived', data=titanic_data) I am getting a TypeError given below: TypeError: countplot() got multiple values for argument 'data' Kindly assist me regarding the same

[@micheleadriaans6688]

Run this command instead sns.countplot(data=titanic_data, x="Sex") Your code assumes that the variable "Sex" is assigned to the data parameter and on top of that you are assigning the dataframe titanic_data to the data parameter as well => TypeError: countplot() got multiple values for argument 'data'

[@aaditya9030]

@@micheleadriaans6688 thnx buddy!

[@yogeshwarkethepalli4234]

@@micheleadriaans6688 Thank You very much buddy, You really helped me

[@yogeshwarkethepalli4234]

@@micheleadriaans6688 RecursionError: maximum recursion depth exceeded while calling a Python object X = titanic_data.drop(columns = ['PassengerId','Name', 'Ticket', 'Survived'],axis = 1) Getting the above error message ,what is the solution

[@micheleadriaans6688]

​@@yogeshwarkethepalli4234 Are you using the Pycharm IDE by any chance? If you are it seems to be a common problem especially when you set breakpoints.

[@sachinvithubone4278]

If the person survived may be it's male or female and if we want to know the name of the person how we can find out?

[@Siddhardhan]

hi! we are not taking that approach... the Idea here is to check whether the person has survived or not based on some parameter. this is a generalised approach. but your doubt is actually a good one. you can definitely try that out.

[@sachinvithubone4278]

@@Siddhardhan Thanks

[@piyushtale0001]

Logistic = 0.81320 SVM = 0.8314

[@dpmharry]

titanic_data.replace({'Sex':{'male':0,'female':1}, 'Embarked':{'S':0,'C':1,'Q':2}}, inplace=True) is giving me error bro

[@Siddhardhan]

hi! mail the details of the error along with the Colab file link to [e-mail]

[@dpmharry]

@@Siddhardhan thanks for your reply , sent mail

[@AdityaYadav-gn2id]

@@Siddhardhan i also got error do you have the solution

[@nanubgm1438]

google colabirity link saher me plz

[@gkakash4789]

This is project-16..bro...

[@Siddhardhan]

hi! i'll rename the Face recognition project to project 16

[@gkakash4789]

@@Siddhardhan 👍

[@TheGifton36]

38:55-39:03🙄😔

[@sreyachanda2842]

showing this error what to do ?~~~~~ . . . TypeError Traceback (most recent call last) in () 1 #making a count plot for "Survived" column ----> 2 sns.countplot('Survived', data=titanic_data) TypeError: countplot() got multiple values for argument 'data'

[@mayukhchakrabarti1507]

Very well made video Just sns.countplot('Sex',hue='Survived',data =titanic_data) Showing error Count plot() got multiple values for argument 'data'

[@mayukhchakrabarti1507]

@Sidhhardhan

[@CompSciSimplified]

Such clear cut explanations.. keep up your good work .. much appreciated 👍

[@Siddhardhan]

Thanks a ton😇