Hi Rob, I want to thank you for sharing your videos. I' ve found them very useful to start my way on data science. Cheers from Argentina!.

Thanks again Rob... Always very high quality content and lessons!

@robmulla

2 жыл бұрын

Thanks Filippo!

Using this to help with a final project for an intro to ML class in university. Really helpful to have such a well organized set of explanations. Even with small errors mentioned in the comments this is a really helpful resource. Thanks for your work Rob!

@robmulla

Жыл бұрын

Love that you found the video helpful. You should share it with the rest of your class! And have them share…. 😁

Great video! Looking forward to watching more videos like this one. Wish you all the best ❤️

@robmulla

2 жыл бұрын

Thanks so much Kha! I'm glad you found the video helpful.

Excellent introductory video, well done. I'm highly motivated to go off and explore now.

@robmulla

Жыл бұрын

“Curiosity driven data science” is the mission statement of my channel. This comment makes me so happy because this video inspired you to be curious!

The representation of topic was very clear, precise and nice. I wish to see such tutorial on color image processing with wavelet transform and quaternions color image processing. Please make such video if possible. Thank you for such nice video

compiler company should sponsor people like this, it will legit be the easiest way to get their compiler attention. Also thank you for making these videos and sharing your knowledge like this. Used it to unstuck myself for a uni project.

Awesome video, thanks man

Great video as always!!

@robmulla

2 жыл бұрын

I apprecaite that a lot Akshat!

Hi, just a correction, in the blurring section, to generate a more blurred image you need to change the size of the kernel too, not just divide for a large number, that's why you get a darker image instead of a blurred one. Thanks for the video.

@robmulla

2 жыл бұрын

Excellent point. Thanks for pointing that out.

@bofa722

4 ай бұрын

Change the size to what? How does it have to match the number you divide by?

thank you for your tutorial. What if I want to create for example animations based on the numpy array of an image. What next steps you suggest? Thanks!

Mr Mulla, you are not a genius, you are a god ! Thanks for the inspiration !

@02:44: one second of OCD. Thank you!!!

Thanks for your video 😊

@robmulla

Жыл бұрын

Thanks for watching!

Nice intro video. If you wanted future video suggestions, can you dive a bit deeper into how to manipulate those numpy arrays that contain image data? So things like np.transpose, np.where, using zip to extract pairs of coordinates, and so on. I know some of this isn't opencv perse, but that knowledge seems key to doing more complicated image processing, object detection, etc.

@robmulla

Жыл бұрын

Thanks for the suggestions. Diving deeper into numpy expressions would be important when working with images in numpy format. Maybe in the future I could cover it.

@meitoli

Жыл бұрын

i was thinking in using the image numpy array to create animations programmatically manipulating the values. Sounds coherent?

Hi Rob thanks for the tutorial! Can you make one on analyzing trends with a large volume of images, common colors, materials, objects etc ?

Really good presentation and content flow. I think THIS video would have been better without the kernel bits since it wasn't clear what the numerical values (of the kernels) were doing. A video focusing on kernels would actually be excellent all on it's own (with a deeper dive into them and what they're doing). Overall, still an excellent video though! Thanks!

@robmulla

2 жыл бұрын

Thanks Dan, that's great feedback. I'm glad you liked the overall flow of the video. Now that you mention it the kernel stuff at the end was a bit forced. I keep trying to remind myself that short and sweet videos are better than overloading with too much information. Thanks for the feedback and I'll keep that in mind for the next video!

# Dear Rob, thank you for sharing, you knowledge, it would be very helpful if we could apply given tutorials, on some kaggle competitions

@robmulla

2 жыл бұрын

Thanks Aka. I'm still working up towards more advanced machine learning videos but hope to get to them in the future. Hopefully this video is still helpful for anyone brand new to image data working on an image kaggle competition.

Hi Rob, just a quick novice's question. What are the advantages/disadvantages using Matplotlib vs OpenCV? Or in other words, what situation which library is better? I'm happy to find your tutorial, you aren't speaking like you have to win a "fastest talking competition" and you really share the principal informations leaving us to look into further details if necessary. (I'm not an Egnlish native)

@robmulla

Жыл бұрын

Thanks for the feedback. The two libraries are very different and only really overlap in their ability to read and display images. OpenCV is a powerhouse for all things computer vision. Matplotlib can do plotting of data and some image processing.

Thanks for your videos! I recently started watching your videos, and kaggle keeps failing to save my draft and I lose my notes, what can I do about this?

thanks, great video ! just a correction though - when you make the image smaller by x in each axis you actually make the whole image x^2 times smaller

@robmulla

Жыл бұрын

Oh, great point! Thanks for pointing that out.

Plz use geotiff images for GIS use as well in next lecture.

Is it work for svg format

Kindly make a video on a cardiomyocyte Data Analysis using Python.

hi please i need help with a topic speaker identification in python. how convert dataset to array two dimension

Hello teacher. I have a question. Why the numpy array of PNG image have range 0~1, JPG have range from 0~255?

please do tutorial of this with tar file. am trying to open a custom dataset with tar file. having trouble opening

hello sir, can you make a video for image processing with opencv using python

Would be great if you could have explained how to flip an image with the usage of pixels

@robmulla

Жыл бұрын

Thanks for watching. It should be really easy to do manually using numpy.flip or you could flip using the built in cv2. Here is the numpy documentation you would just need to select the correct axis: numpy.org/doc/stable/reference/generated/numpy.flip.html

For a university project, I must read a car registration with Python, get its characters. I don't know if you are interested in doing something with this. A camera that reads the car registration, collects the characters and can use those characters to compare it in a database.

@robmulla

Жыл бұрын

Sounds like a cool project. I’ve been asked this type of question a lot so maybe I’ll try to make a video about it. Good luck!

@diegofernandogiraldoportil6572

Жыл бұрын

@@robmulla thanks for answering. You are very good teaching I am from Colombia and the automatic translator works very well. I can read everything you say in the videos and really teach too well. Congratulations.

Dear data 👨‍🔬 Let me know about initial image data for face and accidental face data to find the difference in uaing open cv and matplotlib🎉please sir

Do one with detecting different colors within a pic

Which interface are you using? Is it google colab pro

@robmulla

Жыл бұрын

This is a kaggle notebook. Check out my tutorial on jupyter notebooks and I explain my setup.

hi thanks for this video,this video has 5 likes.😃

@robmulla

2 жыл бұрын

Thank you too. Hope you liked it!

I'm from Pakistan your best video for welcome

How can I link this model with interface I need your help please reply to me In the fastest time

Why use OpenCV instead of Scikit-Image?

Wonderful tutorial! But upscaling width and height 10 times simultaneously doesn't make the image 10 times bigger, but 100 times bigger

13:22 Gray scale but also negative

what is that platform u using?

@robmulla

11 ай бұрын

Linux.

I think the chanel images are inversed like a negative

@robmulla

Жыл бұрын

Oh yea? Do you mean with RGB?

@octavstanculescu1100

Жыл бұрын

I mean like the negative on a film photography

sneaky suplots to subplots @6:49

@robmulla

Жыл бұрын

Nice catch! Must have fixed it and edited that part out 😅

@andrejtursky1551

Жыл бұрын

@@robmullayeah xd, but anyway thank you for your videos and Happy new year from the future! pd.Timestamp.now(tz='Etc/UTC+1')

It looks like you are working in Linux. I have job interview. The source data come from TV. Can you teach me?

@robmulla

Жыл бұрын

Yes I use Linux. My Chanel is my teaching. Hope you find it helpful.

How do masks work?

@robmulla

Жыл бұрын

Depends what type of mask we are talking about...

python is for fast prototyping, not for long lasting code.... new releases of whatever in python kills previous code.... what a shame!!!!

@hanss1754

3 ай бұрын

Gotta love heavy abstractions