You can deploy this model and start earning money today!

Views

No views yet

	DAGGER Training Dataset
















	


	Dataset Description

Training data for DAGGER (Distractor-Aware Graph Generation for Executable Reasoning) models. This dataset contains Bangla mathematical word problems paired with computational graph solutions, formatted for both SFT and GRPO training pipelines.

Highlights

3,000 training examples with verified computational graphs Two training configs: SFT (with validation) and GRPO formats GPT-4.1… See the full description on the dataset page: https://huggingface.co/datasets/dipta007/dagger.