[Icicle-announce] ICICLE Release 2025-07

[Yuan, Zeyu]

ICICLE Release 2025-07
The ICICLE team aims to build the next generation Cyberinfrastructure (CI) to render Artificial Intelligence (AI) more accessible to everyone and to drive its democratization further in solving larger societal problems.
It is with great pleasure that we announce 2025-07 release of ICICLE CI components.
This release includes the following components:
New to ICICLE CI Catalog
Intelligent Cyberinfrastructure
AI-for-CI

  *
Angular Frontend for the Freight Analysis Framework v0.1.0<https://github.com/Data-to-Insight-Center/faf-frontend-ICICLE>
     *   An Angular frontend for visualizing freight transportation data from the US Freight Analysis Framework.
     *
Deployed and running on Tapis Pods.<https://faffrontend.pods.icicleai.tapis.io/>
  *
API Access to the Freight Analysis Framework Dataset v0.1.0<https://github.com/Data-to-Insight-Center/faf-api-ICICLE>
     *   REST APIs that provide access to freight transportation data from the US Freight Analysis Framework, using a MySQL database.
     *   Deployed and running on Tapis Pods.<https://fafserver.pods.icicleai.tapis.io/api/schema/swagger-ui/#/>

AI for CI-for-AI

  *   LLM Training Cost Estimator v0.0.1<https://github.com/ICICLE-ai/distributed_training_estimator_of_LLM/tree/main>
     *   Time cost estimator for distributed training of LLMs.

Use Inspired Science
Smart Foodsheds

  *
Organization SIC Classifier for SmartFoodsheds v0.0.1<https://github.com/ICICLE-ai/organization-sic-classifier-for-smart-foodsheds/tree/main>
     *   A standard Industrial Classification (SIC) code classifier for organizations.
  *
FoodWasteWhiz v0.1<https://github.com/ICICLE-ai/FoodWasteWhiz>
     *   A Streamlit-based chatbot for building and exploring food waste ontologies. It uses LLMs with Retrieval-Augmented Generation (RAG) to answer questions from uploaded PDFs and supports editable feedback stored in a local database.
  *
Agricultural Routing Synthetic Data Generation v0.1.0<https://github.com/ICICLE-ai/ag_routing_data_generator>
     *   A python script to generate synthetic location and vehicle data (csv files) for developers and researchers aiming to train ML Models.

ICICLE CI Components Changelog
Intelligent Cyberinfrastructure
AI Foundations

  *
Patra Model Card Toolkit v0.2.0<https://github.com/Data-to-Insight-Center/patra-toolkit/tree/v0.2.0>
     *   Models in Patra are accessed by querying their model cards, which provide metadata and references. The actual model files are stored and managed externally.
     *   Patra includes an interoperability layer that enables seamless integration with external platforms such as Hugging Face and GitHub, allowing users to upload, retrieve, or link models across these services.
     *   Upgraded schema validation and error messages for clarity.
     *   Robust schema validation messages with detailed logging of JSONSchema errors.
  *
Patra Model Card Knowledge Graph v0.2.0<https://github.com/Data-to-Insight-Center/patra-kg/tree/v0.2.0>
     *   Adds API Endpoints for retrieving GitHub and HuggingFace username and token
     *   Provides model card linkset relations in the HTTP Link header for improved discoverability and interoperability.
     *   Integration guides for OpenAI, Hugging Face, and GitHub.
  *
Science Agents v1.0<https://github.com/ICICLE-ai/ScienceAgentInterface>
     *   Supports graphical user interface.

AI for CI-for-AI

  *   Cyberinfrastructure Knowledge Network (CKN) v0.2.0<https://github.com/ICICLE-ai/ArrayMorph/tree/main%5D(https://github.com/Data-to-Insight-Center/cyberinfrastructure-knowledge-network)>
     *   New plugin ckn-mqtt-cameratraps: MQTT-based camera trap ingestion plugin for edge event and image streaming. Enables scalable, decoupled data collection from edge devices. (ICICLE)
     *   New plugin experiment-alerts: Automated experiment monitoring and alerting plugin. Detects low-accuracy experiments and publishes alerts to Kafka for downstream workflows. (ICICLE)
     *   New plugin ckn_inference_daemon: Edge inference research plugin implementing FastAPI-based serving and model management. Intended for prototyping and experimentation. (Research / Experimental)
     *   Added comprehensive documentation for new plugins and event topics (docs/topics.md), plugin usage, and deployment.
     *   Introduced additional Neo4j indexes to improve query and analytics performance.
     *   Improved experiment completion detection and ensured robust accuracy reporting in monitoring and alerting workflows.
     *   Minor bug fixes, typo corrections, and formatting improvements throughout documentation and example configs.

Software Architecture and Design

  *   ICICLE CI Component Catalog v0.2.0<https://componentcatalog.pods.icicleai.tapis.io/>
     *   Supports interactive graphical view for ICICLE's software release.
     *   Bug fixes.

Use Inspired Science
Animal Ecology

  *   Camera Traps Edge Simulator v0.5.0<https://github.com/tapis-project/camera-traps>
     *   Expands the camera-traps application with a new functional mode, referred to as demo mode. Unlike the original simulation mode, when operated in demo mode, the software relies on a real camera mounted and accessible as a Linux device. It uses motion detection to drive the camera and generate new images.
     *
When running in demo mode, the software has support for detecting a specific object and notifying a downstream process via an MQTT message. In particular, this approach supports integrations with drones which can fly to a particular camera that has spotted an object of interest.

Digital Agriculture

  *   HARVEST v0.10<https://icicleai.tapis.io/#/harvest>
     *   Moved from local system calls to Tapis based submission/authentication
     *   Fully static url for service
     *   Enhanced Labeler based on community feed back
     *   Added ability to upload json of image labels instead of requiring labeler usage
     *   Fixes various visual bugs
     *   Fixes preprocessing/visualization input file errors

The ICICLE team is committed to delivering the best software and CI components. We welcome your feedback and suggestions for future releases. A list of all ICICLE components can be found on our website under CI & Software<https://icicle.ai/cyberinfrastructure/software>
Please subscribe to icicle-discuss<https://lists.osu.edu/mailman/listinfo/icicle-discuss> and post to discuss all installation/build problems, performance issues, features and other miscellaneous questions related to the different software and CI components of the ICICLE project. You are welcome to post patches and enhancements to the released components.
Subscribe to our mailing list icicle-announce<https://lists.osu.edu/mailman/listinfo/icicle-announce> to stay up to date on the latest ICICLE news and releases.
Acknowledgements
This release is brought to you by the National Science Foundation (NSF) funded AI institute for Intelligent Cyberinfrastructure with Computational Learning in the Environment (ICICLE) (OAC 2112606)


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osu.edu/pipermail/icicle-announce/attachments/20250711/9135f4e1/attachment-0001.html>