very few Machine learning lectures are available online... or maybe I came through a few of them.... and I will say that this one is best... ma'am and sir both are just so good... made by learning easy Thank you so much ma'am and sir.
Step1 of the Question 1 is UnSupervised Learning as we are giving a set of only inputs/images and the algorithm has to learn /cluster to form groups of similar items.But in step2 also we are just giving a set of inputs/images and not (input, label) pairs. Then it should also be unsupervised learning ?? ..any help ?
We are using the label to classify so that's also an input for us - this is why it's supervised classification learning. The model is learning that given a certain image from this vehicle cluster - it is a car or a bus or a train etc.
watching your tutorial single video is more usefull than watching all the remaining videos..................you may start a new course in a new channel brother all the best.....
Features are basically the columns in a dataset. For example, The features of a tumor dataset might be the tumor size, tumor shape, thickness and stuff like that. Features are part of the dataset. Parameters are the weights and biases of a model. Parameters are what the model learns over time with the help of a learning algorithm.
in the video, I see age mentioned in Categorical and continuous both sections. Similarly weight can also be Categorical as heavy weight, light weight, etc...
features are the various characterstics of the input e.g pixel values in the image dataset, while parameters are internal to the model and are learned during the training of the model using these features only..
Anirban Santara Hey Anirban. My name is Yokhai , and I'm learning your course. You explane subjects very well, thank you. Can you please tell me how can I see the assignments? Though I dont take the online course with certificate, I would like to solve the assignments for better ubderatanding. Thank you, Yokhai
Hello, First of all thanks for sharing these videos. At 16:00, how second problem is an example of supervised learning? Based on earlier explanation it was mentioned that labeled data is used in supervised learning. Are we expecting labeled data from the outcome of clustering?
features are properties that describe data like features of a horse are its black color, brown eyes. There can be different features for different horses like brown color, black eyes. So these are features can be different for different horses. Parameters are properties that describe model or learning algorithm like the parameters of color of horse fur, color of horse eyes etc. So parameters are same for all horses as they are model specific unlike features that are data specific
Parameters are variables that you tune to model the function that you are trying to learn. Features are functions of the input variables that describe meaningful attributes of the input data
Machine learning made easy with Mr. A Santara.
the best explanation, thanks
good explanation
Good explanation, but somehow feel the confidence level is low. Why does he shake his head so much
very few Machine learning lectures are available online... or maybe I came through a few of them.... and I will say that this one is best... ma'am and sir both are just so good... made by learning easy
Thank you so much ma'am and sir.
This is the best explanation of ML concepts !!
Really u have beautiful handwriting....
Hope that is not sarcastic :P
Anyways, thank you!
nice
Step1 of the Question 1 is UnSupervised Learning as we are giving a set of only inputs/images and the algorithm has to learn /cluster to form groups of similar items.But in step2 also we are just giving a set of inputs/images and not (input, label) pairs. Then it should also be unsupervised learning ?? ..any help ?
We are using the label to classify so that's also an input for us - this is why it's supervised classification learning. The model is learning that given a certain image from this vehicle cluster - it is a car or a bus or a train etc.
Nice explanation made it clear. Post assignments here also so that we can also solve them :)
I can't make up with explanation without at least an alive example. So disappointed :(
The best explanation. Thank you sir
i was able to answer first question s unsupervised learning but i missed the second step answer
Thanks for the explanation, you made it interesting, easy to understand... covered all important topics with examples.
Great Explanation....really love the way you teach.
Thank you +Ritesh!
hi ritesh can i know how this course help in your carrer what your doing now??
hope you reply soon!!!
watching your tutorial single video is more usefull than watching all the remaining videos..................you may start a new course in a new channel brother all the best.....
Lucid explanation of all the topics...Thanks !!!
Thank you, Sir, too much, I really enjoyed your nice teaching it was the best.......
he is looking so enthusiastic.
15:29 I enjoyed learning. Thank you for this wonderful tutorial.
Amazing explanation Sir 👍 soo well explained
Nice way of presentation Appreciate your time and patience for great explanation of concepts. Thanks you very much!
you have a fantastic way of explaining things, superb
why cant we just check generalization on test data without introducing validation set
very specifically how parameters and fetaures that you explained during bias vs variance tradeoff is different??
if you know the answer now please explain it, iam looking for the answer
Features are basically the columns in a dataset. For example, The features of a tumor dataset might be the tumor size, tumor shape, thickness and stuff like that. Features are part of the dataset.
Parameters are the weights and biases of a model. Parameters are what the model learns over time with the help of a learning algorithm.
Can anybody provide me assignmentS ....as i haven't enrolled for the course
How to get the assignment problems for practice?
mam's knowledge>sir knowledge but....... sir's teaching ability>>>mam's ability
Brothers,u r explanation is damm good
Where can I get the assignment questions? Please give a link if there is any.
Sir unable to find the assignments in swayam portal.
Nice explanation. Loved It 😊
Good Explanation. Thanks @Santara.
Best Explanation ...thanks alot sir.
thank you very much sir
why do we need validation set.We can test our generalization on test data also.
in the video, I see age mentioned in Categorical and continuous both sections. Similarly weight can also be Categorical as heavy weight, light weight, etc...
Age-group is categorical whereas age alone is continuous. same with weight-group and simply weight
You are really a great teacher. Thank you sir.
i wish that you could teach the whole course.
Really liked the Explanation Mate @anirban santara
Great Lecture.
What is the difference between number of features and number of parameters. Aren't they same?
features are the various characterstics of the input e.g pixel values in the image dataset, while parameters are internal to the model and are learned during the training of the model using these features only..
thanks alot
🔥 nice video
best explanation sir I appreciate it.
Thank you Saif!
Anirban Santara
Hey Anirban.
My name is Yokhai , and I'm learning your course. You explane subjects very well, thank you.
Can you please tell me how can I see the assignments?
Though I dont take the online course with certificate, I would like to solve the assignments for better ubderatanding.
Thank you,
Yokhai
yokhai yagdanov please let me know if you get 'em.
And just keep changing the unit number..
Awesome lecture
nice!
Hello, First of all thanks for sharing these videos. At 16:00, how second problem is an example of supervised learning? Based on earlier explanation it was mentioned that labeled data is used in supervised learning. Are we expecting labeled data from the outcome of clustering?
After clustering images, we have to identify the image category, this will be done by giving labeled data (data with targets) in training part
Classic case of semi supervised learning
Thank you so much
Thank u so much Mr Santara
Nice explaination 🙏.
But according to me the speed of explaination is little bit high.
Great 👍🏻
what is the difference between features and parameters?
features are properties that describe data like features of a horse are its black color, brown eyes. There can be different features for different horses like brown color, black eyes. So these are features can be different for different horses.
Parameters are properties that describe model or learning algorithm like the parameters of color of horse fur, color of horse eyes etc. So parameters are same for all horses as they are model specific unlike features that are data specific
@@muskanjindal7912 thanks a lot, but you deserves WOW 🤩 for the simplest explanation 👏
@@harisankar6104 No problem! Happy to help.
@@muskanjindal7912😊😊
features vs parameters? what's the difference?
Parameters are variables that you tune to model the function that you are trying to learn. Features are functions of the input variables that describe meaningful attributes of the input data
@@AnirbanSantara Is Parameter same as weights associated with the different features in a target function ?
Bias and variance not clear sir please explain and also can you please tell the difference between features and parameters
u r behaving like a old professor that not suits to him and ur adjusting concepts not cleared urself clear first
Loved the effort to make everything interesting and you did a great job.