Composio equip's your AI agents & LLMs with 100+ high-quality integrations via function calling

Foundation model benchmarking tool. Run any model on any AWS platform and benchmark for performance across instance type and serving stack options.

Guide: Finetune GPT2-XL (1.5 Billion Parameters) and finetune GPT-NEO (2.7 B) on a single GPU with Huggingface Transformers using DeepSpeed

A flexible package for multimodal-deep-learning to combine tabular data with text and images using Wide and Deep models in Pytorch

Implementation of TabTransformer, attention network for tabular data, in Pytorch