doc: add test generation instructions and update run.sh

Added documentation for test generation through Ollama, including new command-line arguments for `generate_tests.py` and updated `run.sh` script. Also added a new `gen` command to `run.sh` for generating tests via Ollama. This improves usability by providing clear instructions and automation for test generation.
This commit is contained in:
2026-01-17 02:40:38 +03:00
parent 5c17378ce4
commit f117c7b23c
11 changed files with 393 additions and 1 deletions

View File

@@ -16,13 +16,31 @@ pip install -r requirements.txt
./run.sh run --model llama3 --ollama-url http://localhost:11434
```
### Генерация тестов через Ollama
```bash
# Генерировать 1 тест для каждой категории через Ollama
./run.sh gen
# Или через Python напрямую
python scripts/generate_tests.py --count 1 --category all --model llama3 --ollama-url http://localhost:11434
```
### Через Python
```bash
python src/main.py --model llama3 --ollama-url http://localhost:11434
```
### Аргументы
### Аргументы для generate_tests.py
- `--count`: Количество тестов для генерации (по умолчанию: 1)
- `--category`: Категория тестов (translation, summarization, codegen, или all) (по умолчанию: all)
- `--model`: Название модели для генерации тестов (обязательный)
- `--ollama-url`: URL подключения к Ollama серверу (обязательный)
- `--validate`: Валидировать тесты в указанной директории
### Аргументы для main.py
- `--model`: Название модели для тестирования (обязательный)
- `--ollama-url`: URL подключения к Ollama серверу (обязательный)
@@ -37,6 +55,11 @@ python src/main.py --model llama3 --ollama-url http://localhost:11434
./run.sh run --model llama3 --ollama-url http://localhost:11434
```
Генерация тестов через Ollama:
```bash
./run.sh gen
```
Запуск только тестов переводов:
```bash
./run.sh run --model llama3 --ollama-url http://localhost:11434 --benchmarks translation