Patch/2.3.1

[chakravarthik27]

Langtest 2.3.1 Dump Version

• NER Task Evaluation Fixes:
  • Issue where NER evaluations passed incorrectly when expected results were empty, but actual results contained predictions. This should have failed. (Fixed)
  • Issue where NER predictions had differing lengths between expected and actual results. (Fixed)
• Training Dataset Limit Issue:
  • Exceeded the maximum limit set on the training dataset during test augmentation allocation. (Fixed)
• Augmentation Data Allocation:
  • Uneven allocation of augmentation data, resulting in some test cases not undergoing any transformations. (Fixed)
• DataAugmenter Class Issues:
  • Export types were not functioning as expected after data augmentation. (Fixed)
• Template Generation with OpenAI API:
  • Issues with OpenAI API while generating different templates from user-provided ones, leading to invalid outputs like paragraphs or incorrect JSON. Implemented structured outputs to resolve this. (Fixed)
• Pydantic Model Enhancements:
  • Added validation steps in the Pydantic model to ensure templates are generated as required.
• Azure OpenAI Service Integration:
  • implementated of template-based augmentation using Azure OpenAI service. (In progress)
• Text Classification Support:
  • Adding support for multi-label classification in text classification tasks.