Hi Keith, I have watched your tutorials for Pandas (older and newer ones). I really find your teaching very effective for me to understand. It would be really helpful if you could do some Python tutorials for cloud computing.
Hi Keith, wanted to say that your content is for me the best in the Python YT space, I just wish you can help us with using Open Source LLMs and a complete Hugging Face tutorial which doesn't use Open AI key, since they dont provide them anymore. If I may, some other topics would also love to learn from you would be Git/github tutorial, Docker, Python + SQL project since there aren't many Anyways thanks for all the hard work you do, may God Bless you 🙏
Hey Keith, I'm enjoying watching your videos and always love your techniques for solving problems. I am starting my data science journey, but I am really confused about which technologies I should learn. Could you please create a video on the roadmap for data science so that more people like me can get a clear view on the path to becoming a data science developer like you? Until you upload the video, could you please give me some starting topics so I can start my journey in the right direction?
Thank you for the kind words! I'm not sure I'll make a roadmap video, but here are some thoughts to start. Begin by getting down the logic & syntax of either Python or SQL (or both). Platforms like Leetcode, Hackerrank, Datacamp, etc. are solid for this. Once you feel confident in the basics, my focus would be on projects. Give yourself a challenge that you're personally interested in. A recent example I shared is that I'm interested in the Olympics so I gave myself the task to scrape data from the site olympedia.org and put it into a spreadsheet that I then cleaned & analyzed. You may know none of the skills needed for this to begin with, but by having a goal that you'd like to accomplish, it will force you to start picking up Python libraries and external APIs that can help you accomplish your goals. If starting a project completely on your own seems daunting, try some guided projects that you can find on UA-cam until you're more comfortable. The goal though should be accomplishing projects on your own that increase more and more in difficulty. That's how data scientists operate in the real world. You don't always know how to accomplish tasks you are assigned, but you know the fundamentals and have the tools to figure it out as you go (this may include a lot of googling or ChatGPT usage). As you are accomplishing projects, I recommend building a portfolio of what you've done and leveraging this as you start applying to jobs. Hope this is helpful!
@KeithGalli Thank you for your time. I got most of the ideas. I am challenging myself to develop at least two projects along with learning this month. I am leaving this comment as a reminder to myself. I promise I will be here commenting after one month, completing and sharing my project links for your review. See you at the month's end.
Hi Keith , I just want to know how does a ML model handle data once its deployed in production.? Like when we build a model we scale the data , remove nulls ,transform it and then use it , but how does all this happen in already deployed models? Because a normal day to day life will have all the uncleaned data. Pleas help , I m really confused. I can build the ml , dl , transformers etc but am confused how is data preprocessing tackled after model is deployed . Basically how is all preprocessing captured in the model to be used after deployment , is it through columntransformers and pipelines or are there any other steps or is it under mlops umbrella ?
Hi Keith, I have been watching your video for many years now and I've learned a lot from you. Just wanted to thank you again...just a quick question. Your videos have high quality. How did you record your screen and yourself? Which software did you use? I would be thankful if you could let me know. Best wishes and regards
Usualmente, yo trate de tener mas de zoom al código, pero fue difícil en este tutorial porque yo necessité compartir el applicación tambien. El código por cada parte es disponible en GitHub y tu puedes verlo mejor allí. Graciás para el comentario.
I agree that it can feel like there are a lot of web frameworks out there. Streamlit is another great library. Having multiple platforms promotes competition and ultimately improves the Python ecosystem for all of us. My recommendation is to start by building expertise in one framework (whether it be Streamlit, Shiny, or something else) and using it for the majority of your projects. Each framework has its own strengths and unique functionalities, so understanding one well can make it easier to learn and adapt to others as needed.
Make a video where you fetch data of the Olympic 2024 medalists using web scraping and display it on the frontend using Flask or Streamlit, with a feature for filtering as well. This project will give many ideas, and there isn't a video like this on UA-cam.
Hi Keith, I have watched your tutorials for Pandas (older and newer ones). I really find your teaching very effective for me to understand. It would be really helpful if you could do some Python tutorials for cloud computing.
Thanks!
You are very welcome! Thank you for the kind donation!!
Thanks
You are very welcome! Thank you for the donation 🙂
Saving all your videos in order to watch and practice later 😊
Love it!
great tutorial, I'm learning a lot. Thanks for your efforts.
Hi Keith, wanted to say that your content is for me the best in the Python YT space, I just wish you can help us with using Open Source LLMs and a complete Hugging Face tutorial which doesn't use Open AI key, since they dont provide them anymore.
If I may, some other topics would also love to learn from you would be Git/github tutorial, Docker, Python + SQL project since there aren't many
Anyways thanks for all the hard work you do, may God Bless you 🙏
Hey Keith, I'm enjoying watching your videos and always love your techniques for solving problems. I am starting my data science journey, but I am really confused about which technologies I should learn. Could you please create a video on the roadmap for data science so that more people like me can get a clear view on the path to becoming a data science developer like you? Until you upload the video, could you please give me some starting topics so I can start my journey in the right direction?
Thank you for the kind words! I'm not sure I'll make a roadmap video, but here are some thoughts to start. Begin by getting down the logic & syntax of either Python or SQL (or both). Platforms like Leetcode, Hackerrank, Datacamp, etc. are solid for this. Once you feel confident in the basics, my focus would be on projects. Give yourself a challenge that you're personally interested in. A recent example I shared is that I'm interested in the Olympics so I gave myself the task to scrape data from the site olympedia.org and put it into a spreadsheet that I then cleaned & analyzed. You may know none of the skills needed for this to begin with, but by having a goal that you'd like to accomplish, it will force you to start picking up Python libraries and external APIs that can help you accomplish your goals. If starting a project completely on your own seems daunting, try some guided projects that you can find on UA-cam until you're more comfortable. The goal though should be accomplishing projects on your own that increase more and more in difficulty. That's how data scientists operate in the real world. You don't always know how to accomplish tasks you are assigned, but you know the fundamentals and have the tools to figure it out as you go (this may include a lot of googling or ChatGPT usage). As you are accomplishing projects, I recommend building a portfolio of what you've done and leveraging this as you start applying to jobs. Hope this is helpful!
@KeithGalli Thank you for your time. I got most of the ideas. I am challenging myself to develop at least two projects along with learning this month. I am leaving this comment as a reminder to myself. I promise I will be here commenting after one month, completing and sharing my project links for your review. See you at the month's end.
today is a great day 🎉🤞🔥
Excellent .
Hi Keith ,
I just want to know how does a ML model handle data once its deployed in production.? Like when we build a model we scale the data , remove nulls ,transform it and then use it , but how does all this happen in already deployed models? Because a normal day to day life will have all the uncleaned data. Pleas help , I m really confused. I can build the ml , dl , transformers etc but am confused how is data preprocessing tackled after model is deployed .
Basically how is all preprocessing captured in the model to be used after deployment , is it through columntransformers and pipelines or are there any other steps or is it under mlops umbrella ?
Hi Keith, I have been watching your video for many years now and I've learned a lot from you. Just wanted to thank you again...just a quick question. Your videos have high quality. How did you record your screen and yourself? Which software did you use? I would be thankful if you could let me know. Best wishes and regards
nice, btw do you have tutorial to run shiny on server using docker ?
How can we add authentication management in shiny python apps?
Hot to make it fullscreen? It is only in the middle
great vid
Thank you!
not quite my tempo but its great!
¡Excelente!, ¿podrías darle un poco mas de zoom al código?
Usualmente, yo trate de tener mas de zoom al código, pero fue difícil en este tutorial porque yo necessité compartir el applicación tambien. El código por cada parte es disponible en GitHub y tu puedes verlo mejor allí. Graciás para el comentario.
@@KeithGalli ok don´t worry, excelent work, sorry for my bad english
¡No problema! Me gusta practicar Español
@@KeithGalli ¡Bien por ti, gracias!
1fst!
ok
😎kick ass!!! thanks man!
me too..... the second😂😂😂.
I appreciate every and all early comments!! 🤠
I hate shiny name. She is our neighbour's dog who barks whole day.😤
haha oh no 😬
Well I hope that this tutorial can help you like the name a bit better!
@@KeithGalli 😀😂
Why so many web frameworks like there is already streamlit which is popular for ml ai dashboard and easy for deployment
I agree that it can feel like there are a lot of web frameworks out there. Streamlit is another great library. Having multiple platforms promotes competition and ultimately improves the Python ecosystem for all of us. My recommendation is to start by building expertise in one framework (whether it be Streamlit, Shiny, or something else) and using it for the majority of your projects. Each framework has its own strengths and unique functionalities, so understanding one well can make it easier to learn and adapt to others as needed.
Make a video where you fetch data of the Olympic 2024 medalists using web scraping and display it on the frontend using Flask or Streamlit, with a feature for filtering as well. This project will give many ideas, and there isn't a video like this on UA-cam.