add dataset card

README.md

@@ -1,3 +1,71 @@
 ---
 license: apache-2.0
 ---
+
+# Dataset Card
+
+
+## Dataset Details
+
+This dataset is primarily created for the work [Fast muon tracking with machine learning implemented in FPGA](http://dx.doi.org/10.1016/j.nima.2022.167546) ([Arxiv link](https://arxiv.org/abs/2202.04976)) that contains ~3M simulated muon events with Geant4.
+Hits in the muon chamber and ground truth of track angle are saved.
+
+### Dataset Description
+
+Please refer to the Section Simulation samples in the referenced work for details.
+
+The file contains 3 keys: 'X', 'Y', and 'corr'.
+
+'X' is a boolean array of size (3072000, 7, 50) used as the input information.
+
+'Y' is a float vector of size (3072000) that contains the ground truth angle to be predicted.
+
+'corr' contains three keys each of size (100, 100) that contains the Pearson correlation factor between the named stations that can be derived from X. Not useful for general purpose.
+
+## Uses
+
+```python
+import h5py as h5
+with open('dataset.h5','r') as f:
+  X = np.array(f['X'])
+  Y = np.array(f['Y'])
+```
+
+## Dataset Structure
+
+```
+<ROOT>
+├── X: bool[3072000, 7, 50]
+│
+├── Y: float64[3072000]
+│
+└── corr
+    ├── 12: float64[100, 100]
+    │
+    ├── 23: float64[100, 100]
+    │
+    └── 13: float64[100, 100]
+```
+
+
+## Citation [optional]
+
+You can cite the original work that introduces this dataset.
+
+**BibTeX:**
+
+```
+@article{Sun_2023,
+   title={Fast muon tracking with machine learning implemented in FPGA},
+   volume={1045},
+   ISSN={0168-9002},
+   url={http://dx.doi.org/10.1016/j.nima.2022.167546},
+   DOI={10.1016/j.nima.2022.167546},
+   journal={Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment},
+   publisher={Elsevier BV},
+   author={Sun, Chang and Nakajima, Takumi and Mitsumori, Yuki and Horii, Yasuyuki and Tomoto, Makoto},
+   year={2023},
+   month=jan, pages={167546}
+}
+```
+