강의

멘토링

커뮤니티

NEW
AI Technology

/

Deep Learning & Machine Learning

Machine Learning Pipeline

You will develop the ability to define problems based on data and clearly explain the rationale and decision-making process behind them. Additionally, rather than focusing solely on the performance of a single model, you will acquire a pipeline-oriented mindset that evaluates the completeness and reliability of the entire machine learning workflow. Furthermore, you will strengthen your problem-solving skills by tracing back the causes of errors when they occur and deriving improvement directions, and through end-to-end project experience, you will acquire practical ML pipeline capabilities that can be immediately applied in real-world settings.

18 learners are taking this course

Level Basic

Course period Unlimited

  • aisw
AI 코딩
AI 코딩
백엔드이해하기
백엔드이해하기
실습 중심
실습 중심
Machine Learning(ML)
Machine Learning(ML)
AI
AI
Python
Python
Docker
Docker
Tensorflow
Tensorflow
AI 코딩
AI 코딩
백엔드이해하기
백엔드이해하기
실습 중심
실습 중심
Machine Learning(ML)
Machine Learning(ML)
AI
AI
Python
Python
Docker
Docker
Tensorflow
Tensorflow
Thumbnail

What you will gain after the course

  • End-to-end process design and management capability: You will acquire practical skills to not only build models, but also directly design and manage the entire workflow of the 'machine learning pipeline' from data collection to preprocessing, training, and deployment.

  • Automated Machine Learning Pipeline (End-to-End Pipeline): You can have your own automated pipeline system as a deliverable that automatically preprocesses new incoming data, trains and evaluates models, and proceeds all the way to deployment.

  • Practical Data Problem-Solving Skills: Able to independently perform the entire process of transforming raw data into an analyzable format, optimizing model performance, and applying it to real service environments.


Just doing modeling?
Real work is all about pipelines.

Go beyond simple modeling and build practical skills to design and manage the entire workflow from data collection to deployment.
Complete 'your own automated ML pipeline' as a deliverable that can be immediately applied to real service environments.


Are you familiar with training models in Jupyter Notebook, but feel overwhelmed when it comes to actual service deployment?

Do you want to prove your ability to design 'end-to-end workflows' that can be directly applied in real-world work, beyond simple algorithm implementation?

Do you want to move beyond the 'experimentation' phase and advance to building 'production services'?

Through this course, you will fully understand the entire machine learning pipeline that once seemed complex, and acquire problem-solving skills that work directly in practice and the ability to build automated systems.


Develop practical skills to design and manage the entire workflow of machine learning pipelines,
from data collection to preprocessing, model training, and deployment.


Go beyond simple modeling and seize the opportunity to grow into a
'practical ML pipeline expert'
by building complex ML systems yourself.

By the end of this course, you will


You will acquire the ability to design and manage the entire machine learning workflow.

  • You will develop practical skills to directly design and manage the entire machine learning pipeline, from data collection to preprocessing, model training, and deployment. You'll go beyond simply developing models and become capable of systematic planning and execution for the successful completion of projects.

Build your own automated ML pipeline system and create it as a deliverable.

  • You will build an automated pipeline system that automatically handles preprocessing, model training, evaluation, and deployment when new data is input. By reducing repetitive tasks and maximizing efficiency, you will obtain your own ML system as a deliverable that can be immediately applied to real service environments.

You will acquire practical data problem-solving skills required in the field.

  • You will be able to independently perform the entire process of transforming raw data into an analyzable format, optimizing model performance, and deploying it to actual service environments. By strengthening your problem-solving abilities to trace back and improve upon errors when they occur, you will grow into an expert who confidently tackles complex data problems.

Enhance your practical competitiveness through end-to-end machine learning project experience.

  • Beyond the Jupyter Notebook environment, you'll gain experience understanding and building the entire pipeline, from deploying completed models to actual services. Through this, you'll develop the capability to build 'real-world services' beyond the 'experimentation' stage, complete a strong portfolio, and secure competitive advantage in the job market.


✔️

Strengthen Your Data-Centric ML Practical Skills - This Course is the Solution

Building End-to-End
Machine Learning Pipelines

In this course, you will learn how to directly design and manage the entire workflow of a machine learning pipeline, from data collection to preprocessing, model training, and deployment. Beyond simply focusing on model performance, you can develop a pipeline-oriented mindset for building reliable ML systems.

Building Production-Ready ML Pipelines

You'll practice building automated machine learning pipelines working with real data in Python and Docker environments. You'll experience the entire process from processing raw data into analyzable formats, optimizing model performance, and deploying to actual service environments, developing capabilities that can be immediately applied in the field.

Practical Examples and Docker-Based Pipeline

The course covers model building and training processes using TensorFlow, and systematically practices ML pipeline deployment using Docker. Through hands-on end-to-end project experience with code examples, you can complete a strong portfolio.

We can solve the concerns of
people like this!

📌

Data Analysts and Beginners

Those who can do modeling to some extent but feel overwhelmed about how to actually deploy to a service
Those who want to understand the entire machine learning workflow

📌

Job seekers who want to enhance their employment competitiveness

Those who want to prove their ability to design end-to-end workflows that can be immediately applied in the field, beyond simple algorithm implementation
Those who want to create a strong portfolio

📌

Developers who want to strengthen their machine learning project experience

Those who want to develop the ability to design and manage the entire pipeline from data collection to preprocessing, training, and deployment, rather than focusing on single model performance
Those who want to build problem-solving skills to track and improve the causes of errors when they occur

Notes Before Enrollment


Practice Environment

  • Operating System: Windows, macOS, and Linux are all supported.

  • Required installation tools: Python 3.7 or higher, Docker is required.

  • Recommended specifications: 16GB RAM or more, 100GB or more of SSD storage space is recommended.

Prerequisites and Important Notes

  • You should be familiar with basic Python programming syntax.

  • It would be even better if you have experience training machine learning models.

  • Experience using Jupyter Notebook environment is helpful.

Learning Materials

  • Lecture slide PDF files will be provided.

  • Practice example code is provided through a GitHub repository.

  • Additional resource links are provided to help understand the concepts.


Recommended for
these people

Who is this course right for?

  • "I know how to do modeling but deployment is overwhelming" - Recommended for data analysts and beginners who have become somewhat comfortable with data analysis and model training in Jupyter Notebook environments, but don't know how to apply (deploy) their completed models to actual service environments. By understanding the entire pipeline process, you can develop the capability to build 'real-world services' beyond the 'experimentation' stage.

  • Recommended for job seekers and aspiring data scientists who "want to enhance their employment competitiveness" and wish to demonstrate the ability to design 'End-to-End workflows' that can be immediately applied in the field, beyond simple algorithm implementation skills. The 'automated pipeline system' output mentioned earlier will serve as a strong portfolio piece.

Need to know before starting?

  • You need a basic understanding of the process of loading data, handling missing values, and converting formats. If you have experience working with data, you can gain a deeper understanding of the preprocessing stage in the pipeline.

  • Since the process includes training and evaluating models, it's good to be familiar with basic machine learning terms and workflows such as training, evaluation, and validation, even if you don't dive deeply into the algorithms themselves.

Hello
This is

국립부경대 SW융합혁신원

Curriculum

All

14 lectures ∙ (2hr 36min)

Published: 
Last updated: 

Reviews

Not enough reviews.
Please write a valuable review that helps everyone!

Free

Similar courses

Explore other courses in the same field!