General usage

A conversion pipeline is executed via the llm-convert tool, with a pipeline typically consisting of a reader and writer, with optional filters in-between. However, only the reader is required.

The following loads the Alpaca JSON file and stores it in CSV format (P/R):

llm-convert \
  from-alpaca \
    --input ./alpaca_data_cleaned.json \
  to-csv-pr \
    --output alpaca_data_cleaned.csv

If you want some logging output, e.g., on progress and what files are being processed/generated:

llm-convert \
  -l INFO \
  from-alpaca \
    --input ./alpaca_data_cleaned.json \
    -l INFO \
  to-csv-pr \
    --output alpaca_data_cleaned.csv
    -l INFO