HuggingFace Models and API PDF
Document Details
Uploaded by BrilliantEpigram
Tags
Summary
This document provides an overview of Hugging Face, an open-source platform for sharing machine learning models. It explains how to access and use the models through the API, which processes tasks like tokenization and vectorization. The document covers different types of inference that can be performed using models from the platform.
Full Transcript
HuggingFace Models and API What is HuggingFace? HuggingFace is an open-source community for sharing model artifacts (different version of ML models) It contains many pretrained and packaged models for popular ML tasks Most of the models are for LLM applications but it’s no...
HuggingFace Models and API What is HuggingFace? HuggingFace is an open-source community for sharing model artifacts (different version of ML models) It contains many pretrained and packaged models for popular ML tasks Most of the models are for LLM applications but it’s not the only thing… Hugging Face Interfaces To access models for HuggingFace, you must use the Interfaces API The API encompasses the process that LLM go through including tokenization and vectorization Interface Hugging Face Interfaces By default, all you have to do to use the API is to just specify the type of inference you want like sentiment_analysis, ner, etc. You can change the API by giving specific models and tokenizers for the model to use (than the default values)
