Getting Started With Large Language Models

  • IntermediateLevel

  • 1 Hr 20 MinsDuration

hero fold image

About this Course

  • Here, you will learn how to train LLMs for Code from Scratch, covering Training Data Curation, Data Preparation, Model Architecture, Training, and Evaluation Frameworks.
  • Explore each step in-depth, delving into the algorithms and techniques used to create StarCoder, a 15B code generation model trained on 80+ programming languages.
  • Understand and learn the best practices to train your own StarCoder on the data.

Learning Outcomes

LLM Training

Master training Large Language Models for code from the ground up.

StarCoder Insight

Gain a deep dive into the StarCoder development process.

Techniques Mastery

Learn techniques used in StarCoder through our Hands-On.

Essential Frameworks

Explore code LLM architecture and evaluation frameworks.

Who Should Enroll

  • Developers interested in training custom code Lsrge Language Modelss from scratch and doing a Hands-On practice.
  • Machine learning engineers and practitioners looking to gain expertise in code generation using large language models.
  • Researchers exploring Large Language Models applications and their usage in software development.

Course Curriculum

Learn the complete process of building Large Language Models for code, covering data curation, model training, fine-tuning, evaluation, and deployment strategies.

tools

  1. 1. The Evolution of NLP: Symbolic NLP

  2. 2. The Evolution of NLP: Statistical NLP

  3. 3. The Evolution of NLP: Deep Learning

  4. 4. The Evolution of NLP: Deep Learning Era II

  5. 5. The Evolution of NLP: Transfomers and Evolution

  1. 1. Introduction to Large Language Model

  2. 2. Understanding Foundational Models

  3. 3. Different types of LLMs: Based on Response

  4. 4. Different types of LLMs: Based on Model Architecture

  1. 1. The Current State of the Art in LLMs

Meet the instructor

Our instructor and mentors carry years of experience in data industry

company logo
Kunal Jain

Founder & CEO, Analytics Vidhya

Kunal has 15+ years of experience in the field of Data Science and is the founder and CEO of Analytics Vidhya- the world's 2nd largest Data Science community.

Get this Course Now

With this course you’ll get

  • 1 Hour 20 Mins

    Duration

  • Kunal Jain

    Instructor

  • Intermediate

    Level

Certificate of completion

Earn a professional certificate upon course completion

  • Globally recognized certificate
  • Verifiable online credential
  • Enhances professional credibility
certificate

Frequently Asked Questions

Looking for answers to other questions?

Foundational machine learning and deep learning knowledge is essential. Proficiency in Python and software development practices are also crucial.

t's an intermediate-level course, so beginners should establish foundational knowledge beforehand. It requires prior experience with ML and coding.

Yes, you'll learn practical insights and best practices for training similar models. The training pipeline, from data curation through model evaluation, will be thoroughly covered.

The training process encompasses data curation, preprocessing, model architecture, and evaluation. Specific techniques and frameworks used in StarCoder's development are explored in detail.

The core instructional content spans 38 minutes, yet additional time for practice and exploration is recommended. Learning is self-paced, allowing you to delve into topics as needed.

Yes, you will receive a certificate of completion after successfully finishing the course and assessments.

Related courses

Expand your knowledge with these related courses and expand way beyond

Popular free courses

Discover our most popular courses to boost your skills

Contact Us Today

Take the first step towards a future of innovation & excellence with Analytics Vidhya

Unlock Your AI & ML Potential

Get Expert Guidance

Need Support? We’ve Got Your Back Anytime!

We use cookies essential for this site to function well. Please click to help us improve its usefulness with additional cookies. Learn about our use of cookies in our Privacy Policy & Cookies Policy.

Show details