Project 2: Diabetes Prediction using Machine Learning with Python | End To End Python ML Project
HTML-код
- Опубликовано: 5 сен 2024
- 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
. In this video, we are building a system that can predict whether a person has diabetes or not with the help of Machine Learning. This project is done in Python. In this project, we use Support Vector Machine model for the prediction.
All presentation files for the Machine Learning course as PDF for as low as ₹200 (INR): Drop a mail to siddhardhans2317@gmail.com
Enroll at One Neuron to learn from 100 courses in one subscription with 5% discount: courses.ineuro...
Hi guys! I am Siddhardhan. I work in the field of Data Science and Machine Learning. It all started with my curiosity to learn about Artificial Intelligence and the ability of AI to solve several Real Life Problems. I worked on several Machine Learning & Deep Learning projects involving Computer Vision.
I am on this journey to empower as many students & working professionals as possible with the knowledge of Machine Learning and Artificial Intelligence.
Hello everyone! I am setting up a donation campaign for my RUclips 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: siddhardhselvam2317@gmail.com
(No donation is small. Every penny counts)
Thanks in advance!
Let's build a Community of Machine Learning experts! Kindly Subscribe here👉 tinyurl.com/md...
I am making a "Hands-on Machine Learning Course with Python" in RUclips. I'll be posting 3 videos per week. 2 videos on Machine Learning basics (Monday & Wednesday Evening). 1 video on a Machine Learning project (Friday Evening).
Dataset File link: www.dropbox.co...
Colab File Link: colab.research...
Download the Course Curriculum File from here: drive.google.c...
LinkedIn: / siddhardhan-s-741652207
Telegram Group: t.me/siddhardhan
Facebook group: www.facebook.c...
Machine Learning Project
Python Project
Diabetes Prediction using Machine Learning with Python
Supervised Learning
Supervised Learning Algorithm
Deep Learning
artificial intelligence
#machinelearning #deeplearning #python #pythonproject Instagram: / siddhardhan23 Instagram: / siddhardhan23
Hi! You can join this Telegram group for regular updates about my videos: t.me/siddhardhan
Thank you!
Hi Siddhardhan, I have few doubts in machine learning. I would like to connect with you to clarify those.
Not able to join the group
sir can you plss upload viodeos on ranom forest algorithm and naivebay's algorithm
Not working
bro u teaching style is amazing,
i have stuck in a problem when ever i try to diploy my project by putting location of file on terminal ,i get this error
'c:\users\dell\desktop\multi_disease' is not recognized as an internal or external command,
operable program or batch file.
plz help me as soon as possible
I usually don't comment like this in public. But I honestly feel this type of works must be appreciated! Thank you so much!
thanks a lot for your positive words 😇
@@Siddhardhan nyger
Finally, someone cared to explain to me like a 5 year old. Thanks brother
I am a postgraduate student in China and I was struggling to find a project to do on ML. Now, thanks to you my brother, I did some good works. Thank you 🙏
Your course is most underrated course in ML
I am lucky i found your channel
This was my first project with Machine Learning and i it was way more than i was expecting, thank you so much
like in machine learning in which ide we want to write code i am new to machine learning
It's not enough to tell thanks for you sir, your videos are very helpful for me, you will explain every bit of coding, just because of your videos, i am learning ML projects. Thank you so much sir🙏🙏. keep uploading more ML projects sir.
You're most welcome 😇
I am from Pakistan, I have learned a lot from Indian tutorials, you guys are doing great. I am in love with India and really want to visit or work there in tech companies
Well done brother, even experienced professionals can't details like you .. keep it up.
Thank you so much brother 😇
@@Siddhardhan @munna I agree with you; I was also struggling with project ideas, and I had more confusion in ML. the way he had explained, hats off, and thank you for teaching and making us watch more videos. This is my first public comment.
Let's make a community of Machine Learning Experts! Subscribe here👉 tinyurl.com/md0gjbis
me
usually don't comment like this in public. But I honestly feel this type of works must be appreciated! Thank you so much
Much much much better neat and detailed explanation of every detail. Far better from my university teachers.
this channel is a gold mine sir, these videos are really helping us to apply the ML algorithms
Glad to hear it!😇
@@Siddhardhan can you suggest a good project, which i can use for my semester project?
@@deltechdiaries5907 can I use this project it self??
@@kamathprajna i did the same 🙂
@@deltechdiaries5907 project yahi kiya kyaa?ye kar sakte hai?
Pls, You should cover data visualization in the project because it helps that which algo is suitable for particular dataset.
Hi Sir,
Thank you for your tutorial.
Can you please share how you decided with finalizing SVM as the most suited algorithm for building the ML model?
Thank you
Thank you very much Sir. This is a complete workflow to understand ML for beginner.
Your expression is excellent
Thank you and congratulations because you teach very well
Please continue strong
Am 9yrs in IT and feeling of starting to love coding
That's great! You will do great things definitely.
Hi @Siddhardhan
I was searching for end to end project in machine learning and just landed on your RUclips channel.
This is my second video and i feel you should also add feature selection , scaling , feature engineering and outliers handling so that we can learn complete process of model creation because right know I personally feel we are just creating test and train set and making our models.
hi! I'll make end to end projects once i complete ML & DL concepts.
@@Siddhardhan ohhh great , Thank you
I can't thank you enough
You have done a great help🙏
damn man! this was recommended by my bro! & this is full of worth it.
Thank you very much Bro you have helped me to build my first project for my Internship
Appropriate , structure wise step by step learning approach is best way
Hello, what about if I have categorical features in my model? how can I add those kind of variables in a output of streamlit?
Thanks a lot! really an amazing video👍👍👍👍....need more videos from you to enrich the ML skills
I really loved and understood the model very quickly thanky you for making such videos😍😍🤗
tried predicting inputs (3,78,50,32,88,31,0.248,26) or the 7th row in the csv file
should return 1 but returned 0,
is this normal?
Hats off sir! Great explanation❤
How can we improve the accuracy of our model here?
I have literally binged-watch your series on ML! I'm going through the project videos one by one. If you already have a video that deals with improving the accuracy, please point me to it. Thank you so much!
I looove your content!
hi! thanks for your positive words 😇 we can optimise our model to get better performance. those topics will be covered in upcoming modules. kindly stay tuned.
very nice explanation thankyou so much
Really great work.From Kenya thank you just began today and was amazing
Do you got correct predictions value?
I got the wrong predictions but it doesn't show any errors,when I try to use different input datas it always show the person is diabetic even I import the non diabetic input datas,
How to correct this?
I am very thanks towards the you and I will now be going to Internships of your projects because. MIT is also on the search for me👀👀👀👀👀👀
finished practicing coding .Feeling confident
hare krishna prabhu ji ....
i like the way u make videos ..
thanku for this video...
So thanks, dear my teacher, it was so useful for me ❤
bro you worked hard ...you deserve this
The file diabetes.csv is not in the description ... pls add it
Its amazing thanks alot ❤
Thank you sir for this lecture. This playlist is helping me a lot.
Hi @Siddhardhan thanks for assisting budding ML engineers with your videos. I had a query regarding standard scaler. Why didn't we use it during Sonar Rock vs Mine project and used it here. How can we identify in which cases we should use standard scaler.
Thanks again
For the Standard Scaler, we used it here because the values being fed into the model totally differed in range, we had values 0.08 all the way to 846, which will not be helpful to the model so we used the standard scaler to scale all the values to values between 0 to 1 to make it easier on the model, but in the submarine one, all the values were in a similar range so we didn't have to do that.
Hi Siddhardhan, after analysing the dataset, I found that few features need imputation.:
My Analysis:
- 11 entries found with BMI 0, which is not practical
- similarly 35 women Blood Pressure entry was 0
- 5 women Glucose entry was 0
- 374 women were having 0 Insulin entry and surprisingly, out of them 236 were non diabetic and only 138 were diabetic.
How to impute insulin value, which is crucial in predicting a person is diabetic or not..
Thank you so much sir I've seen so many videos but your video is perfect
Great video thank you for that my question is when the dataset is a mess and you do some data manipulation works like changing " yes/no" as" 1/0" after everyting when you predicting the value as your code : input_data_as_numpy_array=np.asarray(input_data) part
are you going to use data set with" yes" in it or the converted version
thank you
Explanation is very simple and understandable brother
very underrated thanks a lot very very useful
Well done bro, even experienced professionals can't explain like you .. keep going like this only
Thank you so much 😀
It was a very helpful video. But I think Standardization should be done after splitting the data into training and test sets.
Yes, that's right. I made the change in the other videos
@@Siddhardhan okay, thanks
thank you so much for the help, couldn't do this by myself. thanks!!!
I like what you are doing would love to know colab usage maybe a little simple tutorial of sort
Thanks
Already posted a video on Colab basics. Do check out
Tysm Proff. for crystal clear explanation
you're most welcome😇
can u pls tell me how its gonna take input in real wold like how it will take the details of a actual patient and predict it we cant just give away our code and say put inputs according to the coloumns name it doesn't make sense pls explain
This isn't actually a prediction problem rather it is a classification problem. Good work though :)
Great Bro. Wonderful Channel on Machine Learning.
very clear explanation thank you,please add a video about hybrid machine learning algorithms (like random forest with xgboost or adaboost etc) for predicting diabetes
It is usually recommended to do train test split before feature scaling. Why you did feature scaling first
Thanks sir for this project. It was much useful☺️
Thanks a lot for your great effort you explain it very clearly .
Thanks a lot for this tutorial..Just one suggestion, there are way too MANY ads in all of your playlist, it is really annoying, can we please reduce them? Thanks.
Very well explained, thank you so much for making this video 🙏
My pleasure 😊
The Way You Explain Is Amazing...
am confused in data standardization..When I Run the same code it Gives Name error..How To fix IT?
Hi Siddhardhan, thank you for a comprehensive project tutorial.
At 24:30, the StandardScaler() has not limited the values between [0,1]. Clearly some values exceeded 1 and some negative vales are also there. Could you please explain.
I googled and got to know that the resized distribution should have the mean=0 and standard deviation=1.
Hi Siddhardhan, Thank you very much for the tutorial. In the Rock vs Mine project, you mentioned that the data should be closer together for better predictions. However, for this video, there are 500 records for non diabetes and 268 for diabetes. The data is not close right?
Exactly how many modules are used in this project ? Can i get names?
Why can't we use linear regression model for predicting the diabetes
could you please let me know why are we using SVM for diabetes prediction and Logistic regression for sonar rock vs mine prediction as both are binary classification problem
Hello sir, after the predictive system, how do we sort the result of the prediction system into write and wrong, since the model is not predicting 100% accuracy?
Thank you wish you all the best
loved it man!!! keep up the work
Hi sir, Thanks for the great effort.Can u plz explain which model (linear or logisticregression or svm or random forest )has to be chosen at what criteria for the data,as i m confused to chose which model.
Just now i subscribed u r channel
One main thing u didn't explain about confusion matrix rather than that all is good
Hi, I tried your code and it gives same accuracy as like your result but input results are always coming Zero. Why?
@siddhardhan Thanks for such good videos..
Hey bro but feature scaling should always be done after doing the train test split.
good video and also try to make some projects as per industry standards, end to end
thanks. sure😇
Thank you sir... Very well explained..
You are welcome
thank you so much for your greatest work
Sir can you explain why pregnancies=6 , 7 and so on in data set , I think data should be like during pregnancy can you please explain me this sir please
Can i know what way u learnt the machine learning? How you started with machine learning .Any suggestions for beginners.
Hi danda umasai! It all started with curiosity. Lot of things that could be achieved by Artificial Intelligence surprised me. And I started learning little by little... Then I started loving it... I couldn't think of a particular resource for learning... I used several books, online lectures, etc. Later I wanted to share what I have learnt with lots of people who are interested in the field of AI. As a beginner, you just need that curiosity to learn and you can find your path... There are several RUclips channels like edureka which you can follow...
Excuse me sir, in this case the data is imbalanced?
or what conditions do you need to have to know that the data is imbalanced?
Great content!
Do you get right predicted value,? I got the wrong predictions but it doesn't show any errors,when I try to use different input datas it always show the person is diabetic even I import the non diabetic input datas,
How to correct this?
Hello! Thank you for this awesome and helpful example. Is there a video with the improvement of accuracy for our model?
Great videos, I am new in ML and I have followed the steps but when i input the data I always get 0 and never get 1. I dont know where I can check to make it correct pls help
Hii sir ur videos helped me a lot.....could u plß tell which data type u r using
Which dataset u r using and why
I have one question, sir, you are doing feature scaling before the train test split. This will lead to data leakage or overfitting type of problems. So I think it will be more accurate to do feature scaling after the train test split. can you tell, am I right or wrong?
Hi Gaurav! You are absolutely right! Thanks for pointing it out. It is always a good practice to do standardization after the train test split. In this case, the dataset we are using is a clean dataset i.e. there are not much outliers... So it won't make much difference... If we have several outliers, it will be a problem... Which leads to data spilling as you pointed out..
I already recorded next project video. In that also I followed the same procedure... And this is the way I generally use unless the model is overtrained...
But from the next week's video onwards I'll make sure that I change the procedure... Thanks for pointing it out...
@@Siddhardhan thank you for your response and you are doing great work 👍
Thanks for your appreciation 😇
Bro, Why did you used SVM only , what about other algorithms ..How we can find use Decision tree, How do you choose algorithm please explain me
Why does some of the values from the dataset when inserted gives incorrect answer? Can you please tell. for example when i inserted 8,125,96,0,0,0,0.232,54 into the model it says the person is not diabetic but according to dataset it is 1, which means diabetic.
Well, the accuracy didn't say 100% right, meaning it is prone to error. Not all the predictions will be correct.
Hello sir, Thank you for this superb video. Sir I am facing a problem in executing the fit(), it doesn't give the appropriate output even I have also checked every line of code no error is there.
Nice explanation thanku
I have one query if I do a comptition in kaggle after write all the code how we upload the file in kaggle how we do it pls explains in one video..
Clear explanation 👌
Hey - you seem to flip Y_true & Y_predicted parameters when using the 'accuracy_score' function. Great series o/w, thanks
thanks! I don't get you.. can you elaborate on that accuracy score comment?
accuracy_score(y_true, y_pred, normalize=True, sample_weight=None)
the first argument to the function should be 'y_true" and second argument should be 'y_pred', does that clarify?
Yeah. Got it.
Hi sir one request to you... please make one video on end to end project deployment...using flask or django on one of the cloud platform
hi! deployment videos will be made after deep learning course.
very good video , well explain
thanks a lot 😇
Very well explained
Could you explain why we used svm instead logistic regression ? thank you
I just wanted to demonstrate how to implement svm. no particular reason. But generally speaking, Logistic regression is better for Binary classification problems.
I'd like to ask the difference on the use of parentheses on Outcome: "diabetes_dataset['Outcome'].value_counts()" versus here "diabetes_dataset.groupby('Outcome').mean()" Why did you use square brackets for the first one and curly brackets for the other? What's the diff
the value_counts() method will work only on a Series and not on a dataframe. So in the step diabetes_dataset['Outcome'], we are taking a single column and thus creating a series.
The mean() method will work on both series and dataframes so diabetes_dataset.groupby('Outcome').mean() is fine.
The bracket is round (and not curly), because groupby won't take a list as a parameter.
anna u are doing a great job !! just cant explain how happy i am by watching ur videos. You are explaining things in a straightforward and in a clear cut manner. long way to go bro!!!.... i just have a small doubt . why are we using SVM here . cant we use binary classification ..i am little confused.
Thank you so much 😇 we can use other classification models as well. I just want to introduce different models in different projects. that's the reason.
While spling data x and y I got error like singleton array array(standards aler(), dtype object) cannot be considered a valid collection
Comprehensive explanation. However, why did you use linear kernel. Are the data linearly separable or did you just jump into it as a default.
Just as a default option. It would have been too much to process for a viewer as this was one of the initial videos.
Ok. Thank you for the reply
Anyone explain me why we are especially using SVM algorithm here ? could we use KNN insted of svm?
Very well explained, can you please extend these videos with optimizing techniques.
We are more interested in optimizing techniques to rise the accuracy score.
Is it possible to achieve an accuracy score of 1?
Hi! Optimization topics will be covered after explaining ML models