01:18 Kickoff Introduction
02:25 Agenda for the day
03:58 What is GPT 2 ?
05:46 How to do this ?
06:40 Fetching the model form Hugging Face
11:10 How to scale it ?
13:00 How to go from model artifact to deployed model
13:45 Optimise the model using Onnx format
15:05 Productioning using tempo
18:35 Defining the wrapper
19:40 Run with triton in Docker
21:00 Custom Transformer logic
23:40 Run full pipeline in Docker
24:30 Run in K8s
26:00 Conclusion and QA
01:18 Kickoff Introduction 02:25 Agenda for the day 03:58 What is GPT 2 ? 05:46 How to do this ? 06:40 Fetching the model form Hugging Face 11:10 How to scale it ? 13:00 How to go from model artifact to deployed model 13:45 Optimise the model using Onnx format 15:05 Productioning using tempo 18:35 Defining the wrapper 19:40 Run with triton in Docker 21:00 Custom Transformer logic 23:40 Run full pipeline in Docker 24:30 Run in K8s 26:00 Conclusion and QA