47 docs tagged with "AI4CI"

Scaling Data-Driven Discovery Tasks Toward Open Co-Scientists.

This component implements a time cost estimator for distributed training of large language models (LLMs). It is used to predict the time required to train one batch across multiple GPUs. The predictor module only requires at least a CPU. The computation sampling module needs one or more GPUs, while the communication sampling module requires multiple GPUs, depending on your computing platform.

AutoSDT is designed to maximize ecological validity of scientific programming tasks while minimizing manual curation cost.

What Version 1.2.0 Adds

- This dataset captures both tabular metadata and graph representations from deep learning training workflows, extracted via TensorFlow's XLA compiler.

What the playground does

Architecture

Architecture

Convention and Usage

Graph Neural Networks for Trade Flow Prediction

Architecture

Pipeline Architecture

System Architecture and Design Philosophy

The GNN Food Flow Portal is a Streamlit web app for exploring county-level U.S. food-flow predictions from the GNN FoodFlow Model. Version 1.2.0 includes the original parallel SCTG-specific model exploration workflow plus a new multi-task GNN what-if portal for one-to-one and one-to-many county scenario analysis.

Overview

A dataset designed to support AI-driven resource estimation like runtime prediction to support HPC scheduling optimization by leveraging compiler-level High-Level Optimizer (HLO) graph features and deep learning workload metadata.

WAYS to configure HARP to setup applications for profiling:

System Requirements

Installation

Run the Portal Locally

How to Predict Training Time Using Metadata

Get a Tapis access token

Authentication

Authentication

How to Implement a Hurdle Model for Trade Prediction

Backend Selection

How to Import Existing Annotations

Continuing Multi-Session Missions

An interactive marimo notebook that turns the ICICLE AI Tapis services into a hands-on RAG (retrieval-augmented generation) playground. Paste text or upload a document (PDF, DOCX, TXT, or MD, up to 2 MB), ingest it into the vector store, and chat against it — the notebook chains the embed, vector, and chat services behind a single Tapis access token.

FastAPI service that turns text into embedding vectors using Qwen3-Embedding-0.6B (GGUF quantized) via llama-cpp-python, designed for the ICICLE AI Tapis tenant. The service runs the model locally — no external API calls — so a single .gguf file plus a Tapis token is everything a deployment needs.

FastAPI + Qdrant vector storage and retrieval service for the ICICLE AI Tapis tenant. Clients provide their own pre-computed embeddings — the service handles storage, search, and reranking.

iSpLib is an accelerated sparse kernel library with PyTorch interface. This library has an auto-tuner which generates optimized custom sparse kernels based on the user environment. The goal of this library is to provide efficient sparse operations for Graph Neural Network implementations. Currently it has support for CPU-based efficient Sparse Dense Matrix Multiplication (spmm-sum only) with autograd.

A project leveraging Graph Neural Networks (GNNs) to predict food flows between counties and FAF zones for economic planning, infrastructure development, and policy-making. This model predicts food trade flows between U.S. counties and Freight Analysis Framework (FAF) zones using Graph Neural Networks (GNNs). It addresses the challenges of sparsity in trade data by applying a two-stage hurdle model that distinguishes between the presence and magnitude of trade.

LLM-based reasoning using Z3 theorem proving with multiple backend support (SMT2 and JSON).

A 7-step HPC-backed pipeline for few-shot object detection — from interactive image annotation through class support generation, proposal visualization, and Tapis-powered job execution — built to work across any research domain.

SpMM Example

Quickstart Tutorial

Enviroment

Explore Parallel Model Food Flows

Getting Started with the HLO Feature Dataset

Run the RAG playground end-to-end

Quickstart

Quickstart

Installation

Running Your First Detection Pipeline

Getting Started with WildWing

An open-source, autonomous and affordable UAS for animal behaviour video monitoring using Parrot Anafi drones to track group-living animals.