Add dataset card README.md

README.md

@@ -20,7 +20,6 @@ SimpleVQA is a simple vision-language question-answering dataset designed for te
 
 ## Dataset Description
 
-- **Homepage**: [https://github.com/your-org/inner_VeOmni](https://github.com/your-org/inner_VeOmni)
 - **Repository**: [JosephFace/simpleVQA](https://huggingface.co/datasets/JosephFace/simpleVQA)
 - **Paper**: N/A
 - **Point of Contact**: N/A
@@ -124,28 +123,6 @@ hf_dataset = load_dataset("JosephFace/simpleVQA", split="train")
 dataset = MappingDataset(data=hf_dataset, transform=your_transform_function)
 ```
 
-## Dataset Creation
-
-### Source Data
-
-The dataset can be created using the provided script:
-
-```bash
-python recipes/dataset/create_training_dataset.py \
-    --output_dir recipes/dataset/simplevqa \
-    --num_samples 128 \
-    --seed 42
-```
-
-### Conversion to HuggingFace Format
-
-Convert the JSONL format to HuggingFace Arrow format:
-
-```bash
-cd recipes/dataset/simplevqa
-python convert_to_hf_dataset.py --create_loading_script
-```
-
 ## Dataset Statistics
 
 - **Total samples**: 128
@@ -166,7 +143,7 @@ python convert_to_hf_dataset.py --create_loading_script
 @dataset{josephface_simplevqa,
   title={SimpleVQA: A Simple Vision-Language Question-Answering Dataset},
   author={JosephFace},
-  year={2024},
+  year={2025},
   url={https://huggingface.co/datasets/JosephFace/simpleVQA}
 }
 ```
@@ -175,7 +152,3 @@ python convert_to_hf_dataset.py --create_loading_script
 
 This dataset is released under the MIT License.
 
-## Acknowledgments
-
-This dataset is created for the VeOmni project to facilitate vision-language model training and testing.
-