Skip to content

Version 0.1.2

Latest
Latest

Choose a tag to compare

View all tags
@peremartra peremartra released this 31 Mar 21:10
· 1 commit to main since this release
v0.1.2
1438a3b

GitHub Release: WizardSData v0.1.2

Overview

Version 0.1.2 of WizardSData introduces new templates for financial retail banking scenarios, expanded profile datasets, and bug fixes to improve dataset generation.

What's New

Added

  • New Financial Retail Banking Template: Enhanced conversation generation with specialized templates for diverse banking scenarios, available in /templates/financial_retail_banking/prompts/
  • Expanded Profile Dataset: Added 160 comprehensive profiles for retail banking covering multiple topics like mortgages, investments, savings, and insurance in /templates/financial_retail_banking/profiles/
  • Complete Generated Dataset: Included a full dataset with 1700 rows spanning 320 unique conversations in /templates/financial_retail_banking/outputs/

Fixed

  • Topic Storage: Topic information is now properly stored in generated datasets, making categorization and filtering easier
  • Sequence Numbering: Fixed sequence number progression in dataset generation for more accurate conversation tracking

Getting Started

To use the new templates, update to the latest version:

pip install --upgrade wizardsdata

Documentation

For detailed usage of the new templates, please refer to the documentation at: https://peremartra.github.io/WizardSData/

Feedback

If you encounter any issues or have suggestions for improvement, please file an issue on the GitHub repository.

Thank you for using WizardSData! This release continues our mission to provide high-quality synthetic conversation datasets for training and evaluating language models.

Assets 2
Loading