Skip to content

Took Deeplearning.ai course and experimented these code snippets to focus on summarizing

License

Notifications You must be signed in to change notification settings

muthuka/llm-summarize-demo

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

18 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

llm-summarize-demo

Took Deeplearning.ai course and experimented these code snippets to focus on summarizing. These samples are for summarizing a dialog between 2 people.

Install dependencies with

pip install -r requirements.txt

Run it with

python3 <filename>

Playground files

File Purpose
sample-dataset.py Loading a dataset into memory
encode-decode.py Tokenize & encode/decode
trainingdata.py Show the data used for experiment 50 & 500
zeroshot-summarize.py Zero shot inference using "Summarize the following conversation: "
zeroshot-dialogue,py Zero shot inference using "Dialogue:"
oneshot.py One shot inference using "Dialogue:"
fewshot.py Few shots inference using "Dialogue:"
fewshot-config.py Configurable few shot inference
full-fine-tuning.py Full fine-tuning of FLAN-T5
peft-fine-tuning.py PEFT fine-tuning of FLAN-T5
rlhf.py Reinforced learning from human feedback

Articles

To run RLHF, please run the following:

pip install git+https://github.com/lvwerra/trl.git@25fa1bd

About

Took Deeplearning.ai course and experimented these code snippets to focus on summarizing

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Sponsor this project

Packages

No packages published

Languages