Upload README.md with huggingface_hub

README.md

@@ -1,30 +1,59 @@
----
-license: apache-2.0
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
-dataset_info:
-  features:
-  - name: Question
-    dtype: string
-  - name: Answer
-    dtype: string
-  - name: TargetMolecule
-    dtype: string
-  - name: SampleMethod
-    dtype: string
-  - name: SampleNum
-    dtype: int64
-  - name: SampleRep
-    dtype: string
-  - name: image
-    dtype: image
-  splits:
-  - name: train
-    num_bytes: 17901933
-    num_examples: 1510
-  download_size: 16747432
-  dataset_size: 17901933
----
+---
+license: apache-2.0
+task_categories:
+- question-answering
+- image-to-text
+language:
+- en
+tags:
+- chemistry
+- molecules
+- SMILES
+- molecular-biology
+size_categories:
+- 1K<n<10K
+---
+
+# BACE-V-SMILES Train Dataset
+
+## Dataset Description
+
+This dataset contains molecular data with visual representations for BACE related compounds.
+
+### Features
+
+- **Question**: Question related to the molecule
+- **Answer**: Corresponding answer
+- **TargetMolecule**: SMILES representation of the target molecule  
+- **SampleMethod**: Method used for sampling
+- **SampleNum**: Sample number
+- **SampleRep**: Sample repetition
+- **image**: Generated molecular structure image from SMILES
+
+### Dataset Statistics
+
+- **Total samples**: 1510
+- **Image format**: PIL Image (RGB)
+- **Image size**: 300x300 pixels
+
+### Usage
+
+```python
+from datasets import load_dataset
+
+dataset = load_dataset("molvision/BACE-V-Train")
+```
+
+### Data Fields
+
+- `Question` (string): Question text
+- `Answer` (string): Answer text  
+- `TargetMolecule` (string): SMILES representation
+- `SampleMethod` (string): Sampling method
+- `SampleNum` (int): Sample number
+- `SampleRep` (string): Sample repetition
+- `image` (PIL.Image): Molecular structure visualization
+
+### Citation
+
+Please cite this dataset if you use it in your research.