Dockerizing and Deploying your First Generative AI model
A Talk by Vishal Balashankar and Byron Chu
About this Talk
Building on the excitement around Generative AI models, Cybera’s Applied Data Science Lab team will provide a hands-on demonstration on deploying your first text-2-img pre-trained language model application using a Hugging Face Diffusers library. The demonstration will be done on a remote server using VSCode, while building an inference API using FastAPI and a client-facing interface serving Streamlit using containerization practices. The presentation will highlight the need for incorporating ML deployment principles much earlier in the model development life cycle to ensure a model that is robust and flexible for a wide range of applications.