evaluating

UnitGen 是一个用于生成微调代码的数据框架 —— 直接从你的代码库中生成微调数据：代码补全、测试生成、文档生成等。UnitGen is a code fine-tuning data framework that generates data from your existing codebase.

Software for evaluating the quality of synthetic data compared with real data.

B-OCL is a tool designed to parse and evaluate the OCL constraints designed on BESSER models. It is a python based implementation that utilises a parser and listener generated via an ANTLR4 grammar.

Token-Oriented Object Notation and JSON data format comparison in terms of token size , context window, token cost money, llm readability and data readability along with retrieval of context by the LLMs.

A trade assessment scale based on Nancy Leveson's Software Assessment Scale

Calculator App for evaluation of arithmetic expression and works for basic arithmetic operations

A tool to evaluate multiple options under weighted criteria

Five chapters on testing model output without a research budget: choosing criteria, building an eval set with known good answers, picking script versus model versus human judges, and catching regressions.

Using GPT-turbo-16K api to train the model in order give response for the queries provided

This project aims to research and construct a method for evaluating the coverage of a systematic multi-agent system (MAS) test, and to establish a test evaluation index system suitable for different application scenarios.