After setting the parameters to train_batcher.sh, please run:
bash train_batcher.sh
Then you will get result files in result directory.
You can make raw result files to visualized results (multiple results at one time) by running:
bash organize_data.sh $natt $nval $cvoc $clen id1 id2 ...
Then you will get the visualized results as a markdown file in result_md directory.
When the input size is big, it may be hard to calculate topsim in GPU environment. If you want to run in the CPU environment, run:
bash topsim.sh $natt $nval $cvoc $clen id1 id2 ...
After making a markdown file, you can calculate averaged results by running:
bash average.sh id1 id2 id3 ...
bash state.sh id1 id2 ...
- feat: 新しい機能
- fix: バグの修正
- docs: ドキュメントのみの変更
- style: 空白、フォーマット、セミコロン追加など
- refactor: 仕様に影響がないコード改善(リファクタ)
- perf: パフォーマンス向上関連
- test: テスト関連
- chore: ビルド、補助ツール、ライブラリ関連