Update README.md

README.md

@@ -120,6 +120,15 @@ K:D
 "G" Bc d2"A7" c2 |"D" d4 :|
 ```
 
+### Use your own dataset
+Follow these steps:
+
+1. First, unzip the `data_curation.zip` file.
+2. Next, place your own symbol music data (in .xml, .mxl, .musicxml format) into the `xmls` folder.
+3. Run the `batch_converter.py` file to convert your data to ABC notation into the `abcs` folder.
+4. Run the `add_control_codes.py` file. This will add control codes to your ABC notation and save the data as `dataset.json`.
+5. Finally, run the `train.py` file to train your own model. The weights will be saved as `weights.pth`
+
 ### Usage
 ```
 usage: generate.py [-h] [-num_tunes NUM_TUNES] [-max_patch MAX_PATCH]
@@ -146,20 +155,11 @@ optional arguments:
 
 ```bibtex
 @misc{https://doi.org/10.48550/arxiv.2301.02884,
-  doi = {10.48550/ARXIV.2301.02884},
-  
-  url = {https://arxiv.org/abs/2301.02884},
-  
-  author = {Wu, Shangda and Sun, Xiaobing Li, Feng Yu, Maosong},
-  
-  keywords = {Sound (cs.SD), Audio and Speech Processing (eess.AS), FOS: Computer and information sciences, FOS: Computer and information sciences, FOS: Electrical engineering, electronic engineering, information engineering, FOS: Electrical engineering, electronic engineering, information engineering},
-  
   title = {TunesFormer: Forming Irish Tunes with Control Codes by Bar Patching},
-  
-  publisher = {arXiv},
-  
-  year = {2023},
-  
-  copyright = {Creative Commons Attribution 4.0 International}
+      author={Shangda Wu and Feng Yu and Xiaobing Li and Maosong Sun},
+      year={2023},
+      eprint={2301.02884},
+      archivePrefix={arXiv},
+      primaryClass={cs.SD}
 }
 ```