このリポジトリには、論文「 Generative AI for Business Process Management - Suitability of Modalities
のソース コードが含まれています。目標は、生成 AI を使用してマルチモーダル ドキュメントからプロセス モデルを作成する実現可能性を評価することです。リポジトリは、SAP SAM リポジトリのコードとデータを使用します。
このリポジトリのサンプル コードには、次のようにライセンスが付与されています。データセット自体には別のライセンスが適用されることに注意してください。
Copyright (c) 2024 by SAP.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
次のライセンスは、データ フォルダー内のデータセットに適用されます。
Copyright (c) 2024 by SAP.
SAP grants to Recipient a non-exclusive copyright license to the Model Collection to use the Model Collection for Non-Commercial Research purposes of evaluating Recipient’s algorithms or other academic research artefacts against the Model Collection. Any rights not explicitly granted herein are reserved to SAP. For the avoidance of doubt, no rights to make derivative works of the Model Collection is granted and the license granted hereunder is for Non-Commercial Research purposes only.
"Model Collection" shall mean all files in the archive (which are JSON, XML, or other representation of business process models or other models).
"Recipient" means any natural person receiving the Model Collection.
"Non-Commercial Research" means research solely for the advancement of knowledge whether by a university or other learning institution and does not include any commercial or other sales objectives.
サードパーティのコンポーネントとそのライセンス/著作権情報を含む詳細情報は、REUSE ツールから入手できます。
新しい環境を作成し、必要な依存関係をインストールするために使用できる 2 つの condaenvironment.yml ファイルが提供されています。
environment.yml
: 抽象的な依存関係 (pandas、numpy など) が含まれています。environment-lock.yml
: 再現可能な結果を保証するために、すべての依存関係と推移的な依存関係のバージョンが含まれています。次の conda コマンドを使用して環境を作成できます。
conda env create -f environment.yml
または
conda env create -f environment-lock.yml
複数の Jupyter Notebook を提供します。
data_set_preparation Jupyter Notebook では、データセットがどのように作成されたのかを説明します。
explorer_the_dataset Jupyter Notebook は、作成されたデータセットの特性に関する洞察を提供します。
bpmn_generation Jupyter Notebook は、GPT-4V とゼロショット、ワンショット、およびフューショット プロンプトを使用して、マルチモーダル ドキュメントからプロセス モデルを作成します。
評価用 Jupyter Notebook では、生成されたプロセス モデルとグランド トゥルース モデルの類似性スコアを計算するための評価フレームワークが導入されています。さらにフレームワークを適用して結果を提示します。
├── data
│ ├── examples <- Some example models for illustrating main ideas.
| └── sapsam
│ ├── cleaned <- The created dataset.
| ├── enriched <- Original SAP-SAM data set enriched by some meta data
| ├── evaluations <- Evaluation results
| ├── generated <- Generated process models
| ├── raw <- Original SAP-SAM data set
| └── tmp <- Temporary data
├── notebooks <- Jupyter notebooks.
├── src
| ├── multimodalgenai <- Source code for use in this project.
│ └── sapsam <- Adapted clone of the [SAP SAM repo](https://github.com/signavio/sap-sam)
├── LICENSE <- License that applies to the example code in this repository.
├── README.md <- The top-level README for developers using this project.
├── Thesis_Gen-AI-for-BPM-Modalities_Marvin-Voelter_v04_2024-03-21_final_blurred.pdf <- Thesis with more detailed explanations and thoughts
├── environment-lock.yml <- Contains versions for all dependencies and the transitive dependencies to ensure reproducible results.
├── environment.yml <- Contains the abstract dependencies (pandas, numpy, ...).
└── setup.py <- Makes project pip installable (pip install -e .) such that src can be imported.
このプロジェクトは、GitHub の問題を通じて機能のリクエスト/提案、バグレポートなどを受け付けています。貢献とフィードバックは奨励されており、いつでも歓迎します。貢献方法、プロジェクトの構造、および追加の貢献情報の詳細については、貢献ガイドラインを参照してください。
私たちはメンバー、貢献者、リーダーとして、誰もがハラスメントのないコミュニティに参加できるようにすることを誓います。このプロジェクトに参加すると、常にその行動規範に従うことに同意したことになります。