siyich/spacetools-rlfulltools

--- license: apache-2.0 task_categories: - visual-question-answering - image-to-text tags: - spatial-reasoning - robotics - tool-use - reinforcement-learning - grpo size_categories: - 1K<n<10K dataset_info: features: - name: data_source dtype: string - name: prompt list: - name: content dtype: string - name: role dtype: string - name: images sequence: string - name: ability dtype: string - name: reward_model struct: - name: ground_truth dtype: string - name: style dtype: string - name: extra_info struct: - name: answer dtype: string - name: image dtype: string - name: index dtype: int64 - name: qa_type dtype: string - name: question dtype: string - name: split dtype: string splits: - name: train num_examples: 5500 --- # SpaceTools RL Full-Tools Dataset RL training data for spatial reasoning with the full suite of vision tools. ## Dataset Summary - **Train**: 5,500 samples - **Format**: Parquet with relative image paths - **Images**: Stored in `images/` directory (5,421 files) > **Note**: The HuggingFace viewer may show fewer rows due to image path format. The full dataset contains **5,500 samples**. Use `load_dataset()` or download directly to access all rows. ## Usage ```python from datasets import load_dataset ds = load_dataset("siyich/spacetools-rlfulltools") print(f"Train size: {len(ds['train'])}") # 5500 ``` ## Tools (16 total) See `toolshed_config.yaml` for full tool schemas.