Update README.md

README.md

@@ -1,38 +1,26 @@
 ---
-dataset_info:
-  features:
-  - name: lesion_id
-    dtype: string
-  - name: image
-    dtype: image
-  - name: dx
-    dtype: string
-  - name: dx_type
-    dtype: string
-  - name: age
-    dtype: float64
-  - name: sex
-    dtype: string
-  - name: localization
-    dtype: string
-  - name: dataset
-    dtype: string
-  - name: label
-    dtype: string
-  splits:
-  - name: train
-    num_bytes: 554067440.0
-    num_examples: 2000
-  - name: test
-    num_bytes: 416096785.405
-    num_examples: 1511
-  download_size: 974906400
-  dataset_size: 970164225.405
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
-  - split: test
-    path: data/test-*
+license: cc-by-nc-4.0
 ---
+
+# DermAI Workshop: Balanced Binary Skin Lesion Dataset
+
+## Dataset Description
+
+This dataset is a modified version of the **HAM10000 ("Human Against Machine with 10000 training images") dataset**. It has been prepared specifically for educational purposes for the "DermAI: Building an AI-Powered Skin Lesion Classifier" workshop at the UTD Biotech Club.
+
+The original dataset contains 7 classes of skin lesions. This version has been processed into binary classes: **'Benign'** and **'Malignant'**.
+
+This repository contains:
+* `train_balanced_2k`: A training set with 2K original images between the two classes.
+* `test_binary`: A test set with binary labels.
+
+## Original Source & Citation
+
+This dataset is a derivative of the HAM10000 dataset. Please cite the original authors if you use this data:
+
+* **Original Dataset Link:** [https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/DBW86T](https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/DBW86T)
+* **Publication:** Tschandl, P., Rosendahl, C. & Kittler, H. The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. Sci. Data 5, 180161 doi:10.1038/sdata.2018.161 (2018).
+
+## Licensing Information
+
+This derivative dataset is shared under the same license as the original: **Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)**. It is intended for non-commercial, educational use only.