Some people have mentioned that the code text was small. I tested it on laptop and mobile and seemed fine but i will take care of it in further videos. The code for this episode is available here: github.com/abhishekkrthakur/e01

Video tutorials coming from one of the best data scientist the market has, I suppose the learners just got lucky. Keep up Abhi :)

Abhishek, after your first video, I feel that I can graduate from Jupyter notebook to vscode for training the models. Thanks a lot. Awesome first episode.

Was waiting for a long time to organize my code and jump from Jupyter to Modular Coding . It was really great to learn stuffs from a Champion. Was really happy to see you go to the documentation for the StratifiedKFold , felt like 'tu bilkul mere jaisa hai' :P

Thanks Abhishek, That's a lot of stuff for someone who trains models on jupyter notebook. Built my first ML framework using VCS thanks to you! Keep giving to the community!! Loved it.

Bro, you killed it....love the way it turned out. I must say it's very well planned and nothing gets better than starts with ML framework.

This is an awesome tutorial, I always strongly believe in templates to save time and your video had thrown more insights of keeping it clean and configurable at its very best. Helped me in learning more from this episode, Thank you for putting this up. Looking forward to this learning journey!!

Great video Abhishek. A few suggestions 1. try whether you can reduce the noise of keyword typing. 2. please give some background of what you going to do for a few minutes and then execute the steps. ( eg: you started creating a.py files ) . 3. people who have a basic background on ml steps only can able to understand what it is when you writing the code ( eg: kfold part ), so need some intro -> code -> intro -> code. I believe you have planned to target your audience group. Other than you doing a great job and helping the data science community. Thank you.

Like this way it reflects genuine working conditions...no editing ..keep it coming Abhishek!!

Thanks a lot abishek... Really appreciate your time and effort... Just recently I've started my ML studies, your video gave me lot of stuffs. Keep on posting videos abishek. Cheers!

The way you were coding, it was something Magnificent, so Fast and accurate.

Damn lots of information, thanks for making this video. It's more than ML learned so many best practices. Thanks for taking the time to make this video Abhishek. Let us know if you are planning on opening a patreon will contribute for sure.

Haven't seen a video that's upto the point...Thanks for creating!

Great video on how to organize the code effectively. Good session, waiting for more videos from you.

Thanks a lot Abhishek . Looking forward to see more videos from you. Make some videos on 1) upcoming trends and technologies for the year 2020. 2) What kind of research papers do you read and how do you implement them in your learning 3) What is the best way to remember & learn the data science skills etc..,,

Abhishek Thakur - Bas naam hi kaafi hai !!

Thank you Abhishek, I am not even a new bee in ML I am a just-born bee yet to explore the hive. Anything on opencv library and designing own neural network would be cool. Thank you again for taking out time and help the community.

Thanks , thanks a lot for starting this series of videos for teaching Data science and ML to students like us.

Hello Abhishek, I have just completed the foundation courses in Kaggle. After that I watched this episode. It gives me goosebumps being started and I am eager to share my journey with you someday. Thanks a lot for being so helpful to the community sir.

Thanks for the wonderful tutorial. Also, your story at the start gave great confidence to me who is not so good at ML and trying to find his paths from the teachings and guidance from people like you. if you could let us know how to train the model using on the cloud it would be great. Thanks a lot Sir.

This is Amazing. Please keep up this great work! Trust me you are going to be a Celebrity on youtube. Wanted this for a long time.

Thanks a lot for sharing this framework .. It is very useful ... The speed at which you code and think is amazing ...

Great! I am posting the channel in page. Congratulations.

You're a life saver for beginners like me ... keep up the work please

Inspiration and Application hand in hand in this video. Thans Grandmaster

awesome tutorial mate,this is how people should teach AI :) please keep the good work up!

Hi Abhishek, this was very useful as it is simple and provides a step by step guide to build our first ML model for real life scenarios. Thank you, Tushar

Thank you so much for this!! Probably the best applied ml/dl available for free!! Thank you again please keep going 🙏🙏

Thanks Abhishek for the nice hands-on. One small suggestion if you can announce may be on linkedin that what you will be going to cover in next class and which algorithm you are planning to use, so that people can easily look into that and little prepare to graps the code easily. Waiting for more such videos. Keep rocking.

Abhishek, thanks for this! This video has been extremely useful. Keep up the great work.

Great content. Looking forward to more of your videos. I recently moved away from Jupyter Notebooks to VSCode for developing models and it's reassuring to see my setup is quite similar to yours. I have been using the Remote SSH extension and find it works great but cool to see that VSCode can be run in the browser too, had no idea that was an option but makes sense considering VSCode is built on Electron.

Really good video. Can you explain next time the use of pipelines. Why should we use them and how should we use it?

You are amazing! I think you can visualize and to EDA as soon as you open the CSV!!😎 I hope your subscriber will increase exponentially!

Good things takes time . I bet this guy will teach us much more than just model.fit() and model.evaluate() 🤪

i like how its advanced

Thank you very much for this series. Appreciate sharing this!

Hello Abhishek,Thanks for sharing your story.

Thank you for the video. Appreciate you doing this. Just wanted to highlight that when you are switching between different windows, the correct working page is not visible sometimes (At time: 33.37).

Thanks Abhishek for the series. Very helpful for me :)

That was informative. Thanks a lot.

Thanks a lot abhishek...its really helpful...

I loved it! Thank you very much for this.

Have been planning to start a channel myself for quite sometime now, just couldn’t manage the time to get up and actually start recording videos. Good work man, my best wishes. Also interested to know the specifications of your “@beast” 😊

Nice one! and a really good HandsOn Machine Learning.

why do you use a web base IDE, instead of downloading an IDE? is it just because of the environments and flexibility or is it because it gives you more RAM and hardware to process data? im just cruious because apparantly you need ubuntu for doing the steps youd did and i have windows 10. By the way, great video and thanks for sharing!

Hopefully, there are also tips on how to do this on a windows command line for us beginners. But overall thank you for helping us learn how to replicate your framework! :D

Awesome! Greetings from Azerbaijan!

Dope video! Please make a video on Continual learning and incremental learning.

Subscribed a week ago and so far, it's good

The video was great as expected. Consider Zooming in, the texts are too small to view the video on smaller displays.

Good start! Demonstrate how you would solve a deep learning problems using PyTorch or Tensorflow. Is it the same template and approach? /Tack

Hey! Great video! I was wondering if you can go over what your conda environment looks like?

Hi, How would we do EDA in such work? In Jupyter notebook we have the liberty of seeing the result for each cell. Would you agree that for looking at very specific parts of project code Jupyter notebooks might be a better idea to use? And once labour work is done, then structure that code in different modules at VScode. Thanks for the video.

Hi Abhishek ! Really liked the content of your channel. Please prefer python 3 in your code as python 2 is no more supported.

Great video..keep it up 👍

This seems like Fastai part 2.. So good :D

Great Content

The tutorial is awesome. But it get complex when you introduced many new things and further it get difficult to understand.

Thanks Abhishek for Video :)

Your CPU works so well.

Thank you so much Sir :)

Thanks a lot man!

Amazing Video Sir! By the way, can you please tell me why you have used label encoding for all the features, including the numerical features? Don't we just use it only for the categorical features?

Thank you Sir very much ! I am not going to search any other tutorial channel after this . Please give me a suggestion where I can connect with you for some query as I am very eager to learn this tech .

Have a question about the LabelEconder 30:25 - how we should handle the situation when train and valid datasets are missing some values from test dataset?

sir, at time stamp 30:41 you're fitting both train and validation data into the label encoder won't that create data leakage? i think you should only fit train data and then transform both train and valid with that..please correct me if i am wrong. Thank you :)

Great to Watch you

Do you recommend someone to learn the whole Kaggle micro courses first, before diving into such projects ? I am kinda intermediate level and feel am not ready for this .. Thanks for such live code videos. Helps learning easier. Subscribed.

good work thx

Why I cannot see master (written in yellow text) in my terminal? I get it it has to do with integrating github but how to get that? Thanks

great video ..i have only used jupyter notebook

Hi Abhishek, many thanks for sharing this video! A quick qs, intellisense is not working while typing. In VS Code it works. Is it limitation of code-server? Can this be circumvent?

Sir, can u kindly post videos on how to get started on Kaggle for beginners

What to do for windows.?

Please make some more Machine learning project using reuasable code ,

Why code-server? Like if we want to code on server we can use vs code remote development extension. And thanks for sharing your awesome experience and making tutorials :) .

Stratified Fold creates fold with similar ratio of classes in it?

sir ... we are expecting more theoretical, what that means is explaining concepts theoretically. at last, give a brief overview of code snippets.

Hello, might be a dumb question. Why do you use code-server instead of writing on your own machine?

Hi Abhishek - I am new to Linux and using Windows Subsystem for Linux. Will it serve the purpose? Also, I am not well versed with some commands. Tried to google a lot but could not find a command to start a new terminal what you have done at time 12:46. Can you please advise here....

This being the first video in the series, am I in the wrong place if I didn't know most of the stuff described here? Should I be learning basics from some other sources and then come back here?

vs code is good but for EDA we have to depend on Jupyter notebooks..shouldn't we stick to one tool.. jupyter notebook as we can do everything there ?

Awesome video Abhishek, thanks for sharing framework buildup tips and its very clean n systematic... how to run code-server for windows machine? what other options do we have on windows machines . I was using jupyter notebooks in Anaconda.

I don't know if you'll reply to these question of mine, but I would rather take a shot for this. I want to be a Data Scientist, but I have completed just B.Sc in Computer Science. Should I go for higher degrees or just keep learning the stuffs online? Also Everywhere I see people recommending to do online project to build portfolios, but how the hell am I supposed to do if I have no idea what's the project and how to do it? I am currently learning Andrew NG ML course from Coursera. Thank you.

Great video ...can you please make video on how to generate features from images? I am also confused between CNN and Computer vision which one we should apply when we have images as datasets

0:06 Intro 6:00 Part1 Setup IDE

As someone who's not using a linux-based OS; What does "python -m src.create_folds" do and what is the "windows" equivalent?

Thanks for the video again. Quick question shouldn't the parameters passed to kf,split be X=df.drop(['target'], axis='columns'), y=df['target']..? I mean should we not drop the target column before assigning it to X?

very nicely organised code! I was wondering why didn't you use scikit-learn's cross_val_score function to get validation score ? Thank you

I do not see auto-complete on the IDE. I would appreciate your suggestions.

Thank you Abhishek, what would be the frequency of uploading your videos on youtube? daily or maybe any free day

I missed it. What's the benefit of code server?

what will be the "touch" equivalent in window command prompt??

how can i fix the error at 33:54, i'm using spyder for windows and stuck there for few days, can anyone help me?

What kind of hardware are you using ? My training was considerably slower than yours.

Hi Abhishek, I am a beginner & started with this video. I am getting this error can you pls say what is this about? Line: FOLD = int(os.environ.get("FOLD")) error = TypeError: int() argument must be a string, a bytes-like object or a number, not 'NoneType' Why was this line used?

instead of VS code server, can i use jupyter notebook to run code?

I have started learning ML ...was planning to buy laptop ...which laptop I shoud buy .. Budget is 1 lakh ...

2:54 is when I subbed.

Hi Abhishek, Thank you so much for making this series. I'm encountering a problem that I want to share: When I try to do lbl.transform on the first column of train_df with train_df.iloc[:, c] = lbl.transform(train_df[c].values.tolist()) in train.py, I get "ValueError: y contains previously unseen labels: [nan, nan, nan, ...]". I tried printing lbl.classes_ after we call lbl.fit on the first columns of train and valid dataframes and observed that a new class is being added for every nan value we encounter. Also, I believe we get this error because we are trying to compare two nan values during lbl.transform. I tried filling missing values with df = df.fillna(0) just after reading the TRAINING_DATA but I'm getting "ValueError: Can only index by location with a [integer, integer slice (START point is INCLUDED, END point is EXCLUDED), listlike of integers, boolean array]" in "train_df.iloc[:, c] = lbl.transform(train_df[c].values.tolist())" of train.py How should I handle these missing values. I did not see you facing this problem in the video. Could you please help me out? I'm using Python 3.6.9 and my sklearn version is 0.22.1 and pandas version is 0.25.3 Thanks