diff --git a/README.md b/README.md index 244b130..eded2ec 100644 --- a/README.md +++ b/README.md @@ -1,19 +1,44 @@ -# partiqlegan +# PartiqleDTR +Repository of the Partiqle Decay Tree Reconstruction (DTR) project. +The PartiqleDTR project attempts to tackle the problem of particle decay tree reconstruction using hybrid quantum-classical machine learning approaches. -decayfile -> decaytree -> decayevents - -masses -> decaytopologies (containing decaytrees) -> decayevents - - -## Installation +## :rocket: Setup and Run 1. Clone the repository -2. Create a venv -3. Install the requirements using -```bash -pip compile -``` +2. Install dependencies + - **Option A:** Poetry installed on your system + ``` + poetry install + ``` + - **Option B:** Create a venv, and install dependencies from ```pyproject.toml``` manually +3. Create a dataset and run training using the existing parameters: + ``` + kedro run + ``` + +## :wrench: Configuration + +*The command prefix ```poetry run``` can be omitted if you're using an already activated venv* + +- Pipeline Configuration: + - Open visualization: ```poetry run kedro viz``` + 1. Data Generation: + - Builds decay tree + - Generates decay events + 2. Data Processing + - Preprocessing events + - Generating datasets + 3. Data Science + - Model creation + - Training +- Parameters can be adjusted in the following locations: + - ```conf/base/parameters/data_generation.yml``` + - ```conf/base/parameters/data_processing.yml``` + - ```conf/base/parameters/data_science.yml``` +- Runs are being recorded using MLFlow + - Open dashboard: ```poetry run mlflow ui``` + ## Notes torch_scatter needs gcc-c++ and python3-devel packages to build successfully.