- 58
- 107 966
Danil Zherebtsov
Приєднався 2 січ 2023
Data Scientist on a journey to create a booming UA-cam channel. Stick around, I will post some amazing content.
You can't deploy ML model without these
How to properly prepare your model for serving and what are the deployment options available.
All videos in a series:
1️⃣ Business & Data understanding - ua-cam.com/video/c_CAK0tln_w/v-deo.html
2️⃣ Data Processing - ua-cam.com/video/Zkmyq-pDnmU/v-deo.html
3️⃣ Modeling Best Practices - ua-cam.com/video/EOWLqekVYp0/v-deo.html
4️⃣ Model Deployment - THIS VIDEO
Model deployment with Docker - ua-cam.com/video/vA0C0k72-b4/v-deo.html
Model deployment with UI & Streamlit - ua-cam.com/video/EEuoDuQiQYs/v-deo.html
----------------------------------------------------------------------------------
Attributes:
Deliberate Thought by Kevin MacLeod is licensed under a Creative Commons Attribution 4.0 license. creativecommons.org/licenses/by/4.0/
Source: incompetech.com/music/royalty-free/?keywords=deliberate+thought
Artist: incompetech.com/
All videos in a series:
1️⃣ Business & Data understanding - ua-cam.com/video/c_CAK0tln_w/v-deo.html
2️⃣ Data Processing - ua-cam.com/video/Zkmyq-pDnmU/v-deo.html
3️⃣ Modeling Best Practices - ua-cam.com/video/EOWLqekVYp0/v-deo.html
4️⃣ Model Deployment - THIS VIDEO
Model deployment with Docker - ua-cam.com/video/vA0C0k72-b4/v-deo.html
Model deployment with UI & Streamlit - ua-cam.com/video/EEuoDuQiQYs/v-deo.html
----------------------------------------------------------------------------------
Attributes:
Deliberate Thought by Kevin MacLeod is licensed under a Creative Commons Attribution 4.0 license. creativecommons.org/licenses/by/4.0/
Source: incompetech.com/music/royalty-free/?keywords=deliberate+thought
Artist: incompetech.com/
Переглядів: 2 553
Відео
How to train an effective model and prove everyone that it works.
Переглядів 3507 місяців тому
Fundamentals of correct ML model training. From selecting the optimization/evaluation metrics to the validation strategies. All videos in a series: 1️⃣ Business & Data understanding - ua-cam.com/video/c_CAK0tln_w/v-deo.html 2️⃣ Data Processing - ua-cam.com/video/Zkmyq-pDnmU/v-deo.html 3️⃣ Modeling Best Practices - THIS VIDEO 4️⃣ Model Deployment - ua-cam.com/video/oYlNP21g2Y0/v-deo.html Attribu...
Prepare data for Machine Learning like a Pro
Переглядів 2607 місяців тому
Here we'll discuss: - What are the different data types and how to work with all of them? - How to correctly transform everything into numeric format? - What goes into feature-engineering? - How make sure all the above won't break when new data starts coming in? All videos in a series: 1️⃣ Business & Data understanding - ua-cam.com/video/c_CAK0tln_w/v-deo.html 2️⃣ Data Processing - THIS VIDEO 3...
All the steps of any DS project Spelled Out by a Data Scientist
Переглядів 2697 місяців тому
Business understanding - Data assessment - Date processing - Modeling - Deployment: a comprehensive walkthrough about how to Data Science. All videos in a series: 1️⃣ Business & Data understanding - THIS VIDEO 2️⃣ Data Processing - ua-cam.com/video/Zkmyq-pDnmU/v-deo.html 3️⃣ Modeling Best Practices - ua-cam.com/video/EOWLqekVYp0/v-deo.html 4️⃣ Model Deployment - ua-cam.com/video/oYlNP21g2Y0/v-d...
New kind of Kaggle competitions just launched! Top 50 get rewarded.
Переглядів 3457 місяців тому
Spoiler: the whole competition is hosted on Telegram! Competition Telegram Bot: t.me/SynnaxCompetitionBot Competition Discussion Channel: t.me/SynnaxLab Competition Description & Files: tinyurl.com/eu2txd6y Attributes: Deliberate Thought by Kevin MacLeod is licensed under a Creative Commons Attribution 4.0 license. creativecommons.org/licenses/by/4.0/ Source: incompetech.com/music/royalty-free/...
Getting ahead of 99% of your peers is easy. Do this.
Переглядів 3568 місяців тому
Get a better job, recognition, financial freedom doing these simple things. Deliberate Thought by Kevin MacLeod is licensed under a Creative Commons Attribution 4.0 license. creativecommons.org/licenses/by/4.0/ Source: incompetech.com/music/royalty-free/?keywords=deliberate thought Artist: incompetech.com/ Inspired by Victor Cheng's newsletter!
Quick ML model cloud deployment with UI explained
Переглядів 1,1 тис.9 місяців тому
Quickly transform your local ML model into an online app/service with user interface using nothing but streamlit. Repository with code from the video: github.com/DanilZherebtsov/deploy-model-streamlit Deliberate Thought by Kevin MacLeod is licensed under a Creative Commons Attribution 4.0 license. creativecommons.org/licenses/by/4.0/ Source: incompetech.com/music/royalty-free/?keywords=delibera...
Learn Generative AI & Data Science from scratch in 2024: Complete guide.
Переглядів 2,9 тис.11 місяців тому
From Scratch: in 2024 start and finish with a job your journey in to Data Science with incline into Generative AI. Told by a Data Scientist. Courses in sequence: Step 1️⃣ PYTHON • Python masterclass - tinyurl.com/5n78h2py • Python OOP - tinyurl.com/58wkzhn7 Step 2️⃣ GIT • Git from zero to hero - tinyurl.com/mrxtyb57 Step 3️⃣ Data Science • Traditional Data Science & ML Master Class - tinyurl.co...
Learn these pandas tricks now
Переглядів 17811 місяців тому
Simple yet non obvious pandas capabilities that I use every day. 0:00 Intro 0:14 Import broken data 0.58 Get indexes of min/max values 1:26 Subset data by values 2:07 Remove records by value 3:01 Split data 3:28 Get % data distribution 3:59 Dataframe to markdown 4:14 Open data in Chrome/Safari Code from video here: gist.github.com/DanilZherebtsov/ca88245bfa4de56521a9107b73b55079
Process 100GB data like it is 20GB, told by a Data Scientist
Переглядів 386Рік тому
How to work with 100 GB datasets on your local machine. Code here: gist.github.com/DanilZherebtsov/4a2e0692f37d8db76b02d6130f10fe3f Automated option: $pip install verstack # from verstack import PandasOptimizer optimizer = PandasOptimizer() df = optimizer.optimize_memory_usage('data.csv') #
How to science the sh!t out of a problem.
Переглядів 377Рік тому
True story. Don't try this at home...
Night in life of a Data Scientist. True story...
Переглядів 2,6 тис.Рік тому
Night in life of a Data Scientist. True story...
Deploy ML model in 10 minutes. Explained
Переглядів 39 тис.Рік тому
Level up your Data Science to Machine Learning Engineering. Docker engine download: docs.docker.com/engine/install/ Repo with code from video: github.com/DanilZherebtsov/ml-docker-flask-api Study MACHINE LEARNING DEPLOYMENT INTO PRODUCTION ENVIRONMENT Course 1 (Intro in ML in prod): imp.i384100.net/MLProduction1 Course 2 (ML&Data Lifecycle in prod): imp.i384100.net/MLProduction2 Course 3 (ML Mo...
Advanced missing values imputation technique to supercharge your training data.
Переглядів 2,2 тис.Рік тому
Advanced missing values imputation technique to supercharge your training data.
LITTLE tings that make a BIG programmer
Переглядів 392Рік тому
LITTLE tings that make a BIG programmer
5 Breathtaking tech books that I will never forget
Переглядів 236Рік тому
5 Breathtaking tech books that I will never forget
M2 Max VS M2 Air - Machine Learning. Should you buy the Air for Data Science?
Переглядів 2,8 тис.Рік тому
M2 Max VS M2 Air - Machine Learning. Should you buy the Air for Data Science?
Correct Data Science setup for Arm Macs (M1/M2)
Переглядів 1,8 тис.Рік тому
Correct Data Science setup for Arm Macs (M1/M2)
M2 Mac python installation the right way
Переглядів 4 тис.Рік тому
M2 Mac python installation the right way
Killer Resume template that will get you a job
Переглядів 775Рік тому
Killer Resume template that will get you a job
Single skill to supercharge your Data Science career
Переглядів 1,5 тис.Рік тому
Single skill to supercharge your Data Science career
Spelled out: what is ChatGPT, how is it trained, is it conscious…
Переглядів 172Рік тому
Spelled out: what is ChatGPT, how is it trained, is it conscious…
Thanks a lot! Now I understand the core concepts. If I understand correctly, in Azure AI/ML, once a model is deployed, it provides a URL for inference along with a Docker image. I guess this is what the cloud provider is doing in the background - wrapping the model (e.g., the .joblib file) into a REST API (using something like Flask) and deploying it as a container in a Kubernetes (K8s) cluster. I guess something is happening with AWS Sage Maker
Thank you, I really enjoy the code, but is it possible to use it when we simultaneously have missing data in features and labels(multilabel)?
Thanks!
I am getting an Aborted ! Error whenever i am using the docker --run command. Anyone knows whats going on ?
you look like klaus mikaelson from vampire diaries lol
lol i am working on creating a sort of analysis automation tool for my college project and this is exactly what i was looking for. Initially i was thinking about going with the iterativeimputer or knnimputer. Is your nanimputer is better than them? if thats the case then you are a fucking genius
Informative
After the setup of my terminal apple logo is turned into question mark and the folder image how to fix it can you please replay
Your terminal apple logo turned into a question mark?? Can you provide some more details.
Can we deploy for the code that is written in jypter notebook
Jupyter notebook is a research environment, not development. For production applications use .py files.
Hi, I faced a problem importing tenzorflow after installing tensorflow-metal. Did anyone else face this as well?
Hi. Since the video release some changes happened. Try this: Step 1: Install TensorFlow dependencies from Apple Conda channel. conda install -c apple tensorflow-deps Step 2: Install base TensorFlow (Apple's fork of TensorFlow is called tensorflow-macos). python -m pip install tensorflow-macos Step 3: Install Apple's tensorflow-metal to leverage Apple Metal (Apple's GPU framework) for M1, M1 Pro, M1 Max GPU acceleration. python -m pip install tensorflow-metal
Exceptional video! Thank you Danil! I watched tons of other videos and have no idea on how to deploy my ML model. Stumble on this one and now I'm able to do it.
You're welcome! More cool videos coming soon
really well done. thank you!!
Glad you liked it!
very helpful thanks, But is it require to do hyperparameter tuning of lightgbm models?
For the purpose of missing values imputation - not necessary. Tuning can give a subtle accuracy improvement and it’s justified for an actual prediction model, but I wouldn’t do it for a data processing step.
thanks for letting people know, but can the camera on our phone be usable by the ****** tracking the data, to gather info=' like ' facial recognition && voice data location and finger prints data.
All of this data is processed on your phone. That's how your FaceId works and other authentication services you use every day. But how **** use this data - no-one knows. It is claimed that this data never leaves your phone. I surely hope so...
You look like a pornstar
Absolute mad lad
😎
very helpful !!!!!!!!!!!
Glad it helped!
Sharp
Subscribed immediately. What a straight to point video. Thanks, man.
Thanks! More cool videos coming soon
Thanks for the video and all the effort it took to produce it. It's easy to follow and nicely animated. 👍🏼 However, as of today I would rather use BentoML instead of FastAPI. It's even easier to set up an API and then publish a docker image.
Well, it's a matter of personal preference. I just got around Streamlit first...
This content is fire! I love the efficiency of information delivery, slow enough to understand every step and speedy enough not to waste time, with visual helpers to focus on the right visual information. I also love that you kept errors in the video, so that we learn the most common we might encounter and how to tackle them. Keep it up!
Thanks for the motivation!
Дружище, давай на русском. Больше ведь будет просмотров
Это как ты так посчитал?) Русскоязычное население 220 млн., а на английском говорят 1.4 млрд…
@@lifecrunch на международном английском говорят 1.4 млрд, но просмотры твои на международный уровень как-то не тянут. Не все математикой объясняется)
quick and clear... good job buddy
Glad it helped
cheers mate thanks!
Any time!
can i deploy my computer vision project using this method
Sure. You can deploy any model this way.
Thanks for sharing this video. This is very informative.
❤
Mark my words,one day you will going to be a super star 🎉❤
Thanks, Danil ! This is exactly what I was looking for. Clear and concise tutorial,🙏
Glad it was helpful!
How to use the libraries in VS CODE after installation ?
VSCode is just a text editor. You are installing the libraries in your Python environment. Import them at the top of your script and use them according to the documentation
Hi, First of all, your video provides very useful information, and I want to thank you for that. I have a question I would like to ask you. I am analyzing air pollution in a city in my country. For this purpose, I have created a dataset using air pollution data and meteorological data. I then organized these data into hourly intervals. However, I encountered a problem. My dataset contains null values. These null values appear consecutively in some parts of the dataset. For example, in the first 3000 rows, there are approximately 2500 null values for the NO2, NOX, and NO air pollutants, but in the remaining part of the dataset, there are very few null values. In addition, there are rows where data for all air pollutants are missing, but these rows cover a short period consecutively. I believe this might be due to workers turning off the devices after working hours on certain days. I have previously trained a few models to fill in these missing values, but I did not achieve good results. I would like to ask for your guidance. In these two cases, should I fill in the missing data or exclude them from the dataset? What would be the most accurate method to complete these missing values?
In the first place (a lot of consecutive missing values at the top) I would just drop them. As for those NaNs in the middle, since your data is a time series, I would use something like a rolling window or nearest neighbors values to fill in the blank spots.
Great video! Can you make a video seting up R with vitual environments as well?
I gave up R almost 10 years ago. Although beautiful framework, I had to switch to Python because it gives much more freedom in terms of software engineering and integration into the production environment.
Brilliant
Thank you !
You're welcome!
This is just wonderful and succinct. Thank you!
Thank you for watching!
Great videos
Thanks!
Thanks a lot Danil.. You saved 4hrs of time. Its working for me :)
Glad it helped!
Can I do llama 8b fine tuned with this sir ?
With what? Docker??
@@lifecrunch yeah, I fine tuned unsloth llama 8.1 how to deploy that with docker or cloud providers
@@thevicky1428 Just like any other model. Write the inference script to query the model with prompts or whatever you want to query it with, configure docker as explained in the video, save all the required llama artifacts into the corresponding directories and there you go. Basically repeat all the steps from the video only replacing the 'predict()' function with your llama inference code.
@@lifecrunch thanks sir
👑🙌🙌
Great tutorial ❤
Thank you! 😊
Thanks Sir
Welcome!
Thank you very much! I have the M2 Air and love it. I am starting to work on machine learning and was wondering, if the upgrade would make any sense. Your video answered that for me. Thank you for that!
Yeah, if you're just starting - this is more than enough. By the time you get to big volumes of data, you will figure out how to run it on google colab or kaggle. Anyway an M2 MacBook Air is a great machine for your task.
thanks, im the sole SWE intern working on an AI/ML team, these vids help a lot
Glad that it helped!
Dude, great video. I don’t know how this video doesn’t have more views!!! Algo I use the terminal confirmation from your preview video every day is great. thanks
Thanks man!
good video,keep going bro
Thanks man!
Thanks alot for these videos. Does you have a machine learning/data science course?
No I don't, but maybe sometime in future I will invest time to make it. Thanks!
loved how you touched on other aspects. well rounded
Thanks!
hello, great video. I quit the iTerm and reopened it. how do i get back to the Download directory with the blue border-box? Tried typed the source ~/.zschrc, but didn't work.
i bild a object detection model . that was 180mb in size . how can i deploy my model
That’s an open ended question. Deploy where? I have a few videos on the subject, check them out.
Thanks a lot for the great video. Somehow the links for course 3 and 4 are invalid. Could you please help update the links?
Updated. The problem was that courses 3 and 4 were merged together and had a new link. I've posted the updated link.