arekit-ss 0.23.1

Low Resource Context Relation Sampler for contexts with relations for fact-checking and fine-tuning your LLM models, powered by AREkit

arekit-ss 0.23.1

arekit-ss [AREkit double "s"] -- is an extension for instant object-pair context sampling from AREkit collection of datasources.

For custom text sampling, please follow the ARElight project.

Mind the case (issue #18): switching to another language may changed amount of extracted data due to terms_per_context parameter that crops context with fixed amount of words.

Installation

Install dependencies:

pip install git+https://github.com/nicolay-r/arekit-ss.git@0.23.1

Download AREkit related data, from which sources are required:

python -m arekit.download_data

Usage

Example of composing prompts:

python -m arekit_ss.sample --writer csv --source rusentrel --sampler prompt \
  --prompt "For text: '{text}', the attitude between '{s_val}' and '{t_val}' is: '{label_val}'" \
  --dest_lang en --docs_limit 1 --text_parser lm --output_dir 'out_rusentrel_prompt'

samplers

• nn -- CNN/LSTM architecture related, including frames annotation from RuAttitudes.
• bert -- BERT-based, single-input sequence.
• prompt -- prompt-based sampler for ChatGPT and the related conversational systems [prompt engeneering guide]

Writers

• csv -- for AREnets framework;
• jsonl -- for OpenNRE framework.

Powered by

• AREkit framework