58 docs tagged with "fine-tuning"

Definition of AI and concepts of machine learning, supervised learning, unsupervised learning, and reinforcement learning

6

5

11

Test different fine-tuning models to compare the effects of hyperparameters on model performance

Comparing responses between a Casual Tone Bot and a General AI Model

9

10

Comparing the differences between a general AI model and an AI model fine-tuned with the speaking style of a Tudor Queen

Creating a JSONL dataset for fine-tuning in a practice environment

4

1

Key data formats used to train AI models, such as CSV, JSON, and XML

Definition of hyperparameters and key hyperparameters

Structure and Supported Data Types of JSON

5

The role of loss functions and key examples

Understanding Accuracy, Precision, Recall, and F1 Score for AI Model Evaluation

Observe response differences according to various MBTI types

A comparison between a general AI model and a fine-tuned model specializing in tabular data, highlighting differences in response results.

Introduction to resources and communities for advanced learning

Execute fine-tuning after selecting training data and setting hyperparameters

7

Definition of JSONL and why it's used for fine-tuning

9

Definition and role of gradient with metaphorical explanation

Comparing Key Differences Between Machine Learning and Deep Learning

A comparison of the differences and characteristics of fine-tuning and prompt engineering

The Process of Fine-Tuning and Its Impact on AI Models

Common methods for deploying a fine-tuned model.

Dataset splitting ratios, random sampling, and stratified sampling methods

The JSONL format used for fine-tuning AI models on the OpenAI platform

The most important hyperparameters for fine-tuning learning rate, batch size, number of epochs

The importance of learning rate and strategies to optimize it

The impact of the number of epochs on model training and how to determine the optimal number of epochs

The necessity of preprocessing and an example using JSONL data

Types of data in datasets and their common structures

8

9

2

3

Summary of Key Learning Points from the First Fine-Tuning Course

4

The impact of batch size on model training and how to set the optimal batch size

Understanding AI and the concept of fine-tuning with practical examples

8

The roles of training, validation, and test datasets

The concept of overfitting and its prevention strategies

Concept of Underfitting and Prevention Strategies

7

Key advantages gained from fine-tuning

Various data formats used by text processing and image processing AI models.

Understanding the AI learning process through data collection, preprocessing, pattern analysis, storage of learned information, and model utilization.

Comparison between a general AI model and an AI model trained to speak in the manner of a Colonial America scholar

Key concepts and examples of deep learning, including neurons, layers, and learning

Causes and solutions for training instability

6

Introduction to the fine-tuning process data preparation, model initialization and configuration, model training and evaluation, and applying and deploying results.