Data

Tools

Indexes

Applications

Experiments

Open Source Science

https://github.com/besser-pearl/caselens-agents

Agents for the CaseLens project with Autorité de la Concurrence

https://github.com/besser-pearl/caselens-agents

Science Score: 26.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ○
    CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ○
    Academic publication links
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (11.2%) to scientific vocabulary
Last synced: 10 months ago · JSON representation

Repository

Agents for the CaseLens project with Autorité de la Concurrence

Basic Info
  • Host: GitHub
  • Owner: BESSER-PEARL
  • License: mit
  • Language: Python
  • Default Branch: main
  • Size: 177 KB
Statistics
  • Stars: 0
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created over 1 year ago · Last pushed 11 months ago
Metadata Files
Readme License

README.md

CaseLens Agents

Agents for the CaseLens project. Built with BESSER Agentic Framework

Run the app

Requirements

  • Python >= 3.10
  • Recommended: Create a virtual environment (e.g. venv, conda)

shell pip install -r requirements.txt

shell python run.py

You can access the application in http://localhost:8501

Deploy with Docker

1. Build Docker image

This command uses the file Dockerfile

shell docker build -t besser-agents .

2. Create and run container

This command uses the file docker-compose.yml

shell docker-compose up -d

You can access the application in http://localhost:8501

Volumes

In Docker, a volume is a persistent storage mechanism used to store and share data between containers and the host system, independent of the container's lifecycle.

This app has 2 volumes, as specified in docker-compose.yml:

The data volume contains the following:

  • config.ini: properties for the agents. We can define the following properties here:
    • nlp.ollama.host = localhost Host address of the Ollama LLM
    • nlp.ollama.port = 11434 Port of the Ollama LLM
    • nlp.ollama.max_tokens = 8000 Maximum number of input tokens for the LLM
    • nlp.ollama.model = gemma3:12b Name of the Ollama LLM (full list here)
    • nlp.hf.tokenizer = google/gemma-2-2b-it Name of the tokenizer to use (should be the same family of the LLM. (full list here))
    • nlp.hf.api_key = YOUR-API-KEY HuggingFace API Key. Some tokenizers may need authentication and therefore it is necessary to provide this key.
    • elasticsearch.host = localhost Host address of the elasticsearch database
    • elasticsearch.port = 19200 Port of the elasticsearch database
    • elasticsearch.index = castor-test-enron Name of the elastiscearch index
  • datalabelingagent folder: Contains the file request_history.json, which stores the requests done with this agent.
  • chatfilesagent folder: Contains the file chat_notebook.json, which stores the requests done with this agent. Also contains the chats folder. All imported chats are processed and exported in JSON format into this folder. The agent actually uses these files to analyze the chat files.

Owner

  • Name: BESSER-PEARL
  • Login: BESSER-PEARL
  • Kind: organization
  • Email: jordi.cabot@list.lu
  • Location: Luxembourg

GitHub Events

Total
  • Push event: 2
  • Public event: 1
Last Year
  • Push event: 2
  • Public event: 1

Dependencies

Dockerfile docker
  • python 3.12-slim build
docker-compose.yml docker
  • besser-agents latest
requirements.txt pypi
  • besser-agentic-framework ==3.0.1
  • elasticsearch ==8.11.0
  • ollama ==0.5.1
  • streamlit-browser-session-storage ==0.0.3
  • streamlit-local-storage ==0.0.25
  • streamlit-screen-stats ==0.0.82