Support Vector Machine - SVM - Classification Implementation for Beginners (using python) - Detailed

Steps followed are:
----------------------------------------
1. Introduction to SVM
Used SVM to build and train a model using human cell records, and classify cells to whether the samples are benign (mild state) or malignant (evil state).
SVM works by mapping data to a high-dimensional feature space so that data points can be categorized, even when the data are not otherwise linearly separable (This gets done by kernel function of SVM classifier). A separator between the categories is found, then the data is transformed in such a way that the separator could be drawn as a hyperplane.
----------------------------------------
2. Necessary imports
----------------------------------------
3. About the Cancer data
----------------------------------------
Original Author - UCI Machine Learning Repository (Asuncion and Newman, 2007)[mlearn.ics.uci....]
Public Source - s3-api.us-geo....
----------------------------------------
4. Load Data From CSV File
The characteristics of the cell samples from each patient are contained in fields Clump to Mit. The values are graded from 1 to 10, with 1 being the closest to benign.
The Class field contains the diagnosis, as confirmed by separate medical procedures, as to whether the samples are benign (value = 2) or malignant (value = 4).
----------------------------------------
5. Distribution of the classes
----------------------------------------
6. Selection of unwanted columns
----------------------------------------
7. Remove unwanted columns
----------------------------------------
8. Divide the data as Train/Test dataset
----------------------------------------
9. Modeling (SVM with Scikit-learn)
----------------------------------------
10. Evaluation (Results)
----------------------------------------

[@ajaykushwaha-je6mw]

best ever video on youtube on SVM.

[@romariofilipovic7133]

It's been almost two years now and you're still helping people with this video. Thank you!!

[@mrbug100yearsago5]

*four years

[@iamindian6913]

Thankyou for your Explanation, I went through a lot of videos in youtube about python, but no one told about Help function.

[@ayushjain7055]

Great explanation. I was shocked to know that this is the only video he put up in this channel... i really liked this video and patiently listened to it. I have subscribed this channel in the hope that one day you will continue.
Thank you for creating this video

[@asmaaahmedabdelfattahelsaw3305]

The better video ever clarifying SVM!!😃

[@connorl1885]

Thank you - extremely helpful.

[@MrMonkey820]

Very thorough explanation, Thank You!

[@jonelleallen7142]

Best video to explain SVM to a beginner!

[@crystalphd4830]

Best video for intro to SVM!

[@sparshmehta2967]

Its such a great video on SVM that I had understood it from depth we want more such videos please...

[@fasiulhaq2366]

Best ever video on TH-cam on SVM❣

[@priyadeokar5829]

really very informative video on svm

[@dr.maheshnaik2919]

That's a very precise and best explanation I have come across.Excellent

[@silpas8740]

Thank u so much sir i have no words to explain my gratitude for this video

[@sheikhobada8305]

Thank you for making the best lecture about SVM❤

[@BhushanPatil-yz9ty]

keep going ,the session was very good!

[@anuradhaghotekar6618]

Thank you for great explanation Sir, it helped a lot learning practical implementation of SVM

[@himawanramadhan]

This video helping me to finish my essay and got bachelor title. Thx.

[@VibrantVibes123]

Very comprehensive, detailed and well-elaborated video on SVM, The top best videos on SVM on TH-cam. Thanks for your effort and teaching...

[@sainiwinarth]

Thank you so much for sharing your knowledge, It was really helpful, keep doing the good work

[@vamsinvizag4936]

Beautifully explained. Thank you!

[@ayeshah9880]

this guy is amazing i swear

[@roopsingh5056]

Best video on svm.good explained

[@sulemandawood9140]

ThankYou so much. You are saviour.

[@oswisac]

thank you so much, it was very useful

[@MansaKundrapu]

Very informative and clear lecture 👍🏻👍🏻

[@otobonoah7955]

beautiful and a wonderful tutorial

[@brandonteller5360]

Awesome explanations!

[@vamsikrishna2394]

Awesome explanation..

[@dataanalysis2527]

great work sir . we also need a video for chatbot also

[@toniiicarbonelll287]

you are the absolute best!

[@mustaphataiwo6073]

Great job

[@mahamatdjibrine95]

Thank you Very much ,keep it up!!!

[@prakashdas8518]

Awsome explanation..... Thank you sir.......will you make something on random forest, dtree, ann, naive bays, kmeans

[@sanjanakale594]

Awesome explanation. Cleared all my doubts. Could you please share the code/ jupyter notebook in the comment section. It will be of a great help

[@boooringlearning]

well-explained, thanks!

[@suchitramohanty2091]

Excellent vdo

[@kelixoderamirez]

permission to learn sir. thank you

[@deepshah9353]

Thankyou so much sir!!!

[@mkay7800]

Awesome explanation in depth ;)
btw do you have a github repository or blog where i can find your code ?

[@arvindersingh9588]

Great explanation!!!
can we have more videos like this by you?

[@sudhanshu_kulshrestha]

Not now. Maybe in the future. I have a conflict of interest.

[@afshandaaiman2927]

Nice explanation sir can you make more videos about Naive bayes, KNN, DT. 👌👍

[@mimifarhana6855]

Explanation on SVM so perfect. how about if dataset is unstructed and non-numeric data? Is it can follow as the step in this video?

[@pargaienclasses]

very nice

[@arunjohn492]

Thank you SIR !!!!

[@Andreas32300]

This is so good and helping, but only lacks decision boundary. Any idea?

[@cheesebalz100]

exellent video

[@jeanyeager4252]

Hi sir, thank you for your video and the very clear explanation, really appreciated. Can i ask for the codings that are used in the video?

[@gauravojha1554]

mila kya code?

[@peehul]

Please teach to plot hyperplane also

[@nnamdinwoye7698]

Thank you so much

[@MohammedHAzeez]

THANX

[@monicagunjal2977]

Can I get code??

[@haseebrehman1805]

Bravo!!

[@mdraihanulislamtomal6064]

please make a video on SVM on Word2Vector... how to train and test data and prediction result using SVM on word2vec. Thanks

[@shukri3285]

THANK YOU!!!

[@saharnazes1649]

Hello. Thanks for your video. it was really great. I have one question though. does this line mean that only first 200 rows that their class value is 2 will be plotted? I mean we will see only 200 points in the plot?
benign_df = cell_df[cell_df['class'] == 2] [0:200]

[@csminor7456]

thank you for this code sir

[@Md_Fuzail_Ahmad]

Can we use it for trading?

[@13eau33]

awesome Thanks!

[@dineshk658]

Thanks 👍

[@cynthiamoricordova5099]

Good explanation but please, could you give the name of the book which you've been using during this video?

[@aimanjavid8441]

superb Sir.
Well Explained. I have faced issue at last. when I code classifier.fit(x_train , y_train) .
given error has occurred. ValueError: could not convert string to float: '?' . could you kindly help me out

[@livematters3206]

can u plz provide the notebook ... with source code

[@Earn26924]

Thank you

[@palawveeheerah9670]

Y we didn't normalise the dataset array before applying the model kernel?

[@gauthamsk4040]

Thanks a lot

[@mohammedbot1339]

How can we compute the training accuracy only ? Not the testing accuracy.

[@antaryami2203]

pls don't use mechanical keyboard the sound is so irritating . The video is overall good

[@indrakumari1854]

Sir, please make more videos related to ML

[@existingstars3665]

Can you please explain about radial base function in spam detection in jupyter please

[@ShivamGupta-xw2sy]

what does happen when we have more than 2 classes i.e. multiclass??

[@arpitpaithankar1607]

PLEASE help me sir I'm getting this error on different dataset....ValueError: bad input shape (166, 61)

[@solartechnologyservices6695]

Thanks

[@PoojaYadav-tq3yp]

Sir I have faced problem to split train and test data set one error is occured like value error about train set is empty

[@ashokjayam3231]

who will demonstrate of how graphs are plotted

[@dhavalrathod9827]

THANKS

[@priyadeokar5829]

Please upload some more videos related to ML

[@cliccme]

Sir, from where i downloaded this code ?

[@lettihera8325]

Sorry, can you help me. How to visualize the result with support vector, hyperplane and max margin?

[@sujaldhungana3216]

whats next after this? how do i use this trained model?

[@seungsooim2183]

i have one question. what if i want to make a model with svm that contains strings in my attributes?

[@jyotinanwal2672]

if at step 6 some of the columns are of integer type and some are floating type then. what to do in this case..

[@saksharora6560]

Share the Jupiter notebook link of this session pls

[@monashaaban2337]

please how can be calculation time training model?

[@anjalichoudhary2244]

Iam getting a bad input shape in step 9 can you explain?

[@ankeetbose]

17:00 how do you decide x and y?

[@ritikanigam5234]

Can you share that notebook?

[@nimawangchuk5497]

cant u make a graph showing the last code?

[@tanmaythaker2905]

There was a significant class imbalance.What about that?

[@FluffyVooghtje]

I have a dataset of (only) accidents with both numeric and categorical variables. How can I know with one-class SVM which variables are influencing?

[@abhinaygupta8243]

i think u have to first draw a pair plot then u have to see for the best gausian and according to that u have to apply svm for each good gausian attributes, so u will find out what is best influencing

[@pallawirajendra]

Hi, Thank you for the explanation. May I know if you can share the notebook.

[@sudhanshu_kulshrestha]

I have intentionally not provided the notebook so that viewers have to write down themselves. It's just 20 lines of code to write.

[@harishmanwani39]

Why have you stopped making new videos

[@AnjaliKumari-in7xc]

Please make some more videos

[@ShaheenMirja]

My dataset has no numeric value.its a news archive dataset and i want to detect the noveltyfrom this news archive.i want to use SVM.I need a help.Can anyone help me please?

[@dharmindave3033]

So I think this will be a dataset in for of statements. So you can try to learn them through SVM.

[@happybud2446]

refer to point 6 of the video.u can convert datatype of each and every column.else u can manually convert those non numeric values in the xlsx file.
say if u have attribute color{red,blue,green} then u can change them to color{1,2,3}

[@shamsulalam2982]

In min 19.08 u just forget to change the label of malignant......

[@sonibind9430]

I have one doubt can you help me please

[@anmolvarshney4519]

Sir, I need to know about the hyperplane. SVM is the separation so please plot that graph too or please tell me.

[@sudhanshu_kulshrestha]

Hi @Anmol, a hyperplane in a 2D plot, would be a simple line (or curve) that can separate the different available classes in the data sets. I will try posting a new video, but meanwhile you can refer this nice blog - chrisalbon.com/machine_learning/support_vector_machines/plot_support_vector_classifier_hyperplane/

[@anmolvarshney4519]

@@sudhanshu_kulshrestha Thank you sir

[@harishmanwani39]

47:10 recap

[@ganeshkashyap8573]

github link please

[@subinoymaity9318]

Sir want your help pls help me🥺