Time series learning in a many-body Rydberg system with emergent collective amplification

Data for "Time series learning in a many-body Rydberg system with emergent collective amplification". # Time-Series Learning in a Many-Body Rydberg System with Emergent Collective Amplification — Data & Code Repository This repository contains the complete dataset and Python simulation/plotting scripts required to reproduce the numerical and experimental figures presented in the manuscript **"Time-series learning in a many-body Rydberg system with emergent collective amplification"**. --- ## Repository Structure & Overview The directory layout: ```textzenodo_repository/├── experimental data/ <-- Raw time-series datasets grouped by task│ ├── lor/ <-- Lorenz time-series raw data│ ├── weather/ <-- Weather Beijing csv and raw EIT│ ├── MG/ <-- Mackey-Glass delay differential scan│ ├── NARMA/ <-- NARMA memory capacity│ ├── iid_random/ <-- EIT sweeps and NARMA pre-calculated summary│ └── data_Fig2/ <-- Plotting trajectory data for Figure 2 panels├── theoretical calculations/ <-- Mean-field phase space mapping data and MSE code│ ├── results_MF_model_3_parameters_7.txt│ ├── rydberg_gas_mse_cut.txt│ └── generate MSE/├── figures/ <-- All generated PDF plots are saved here├── requirements.txt <-- Python package dependencies├── paperstyle.mplstyle <-- Matplotlib stylesheet for publication-grade formatting├── plot_fig_2.py <-- Figure 2 prediction plots & NRMSE├── plot_fig_3.py <-- Figure 3 mean-field diagram (with automatic caching)├── plot_fig_4.py <-- Figure 4 benchmark NRMSEs comparison├── generate_methods_figures.py <-- Methods figures (Fig 6: EIT, Fig 7: IPC, Fig 8: NARMA scans)└── reservoir.py / reservoircomputer.py <-- Reservoir computer helper modules``` *Note: Large raw experimental dataset files (`.csv` format) are packaged inside `csv_data.zip` archives within their respective subfolders. All scripts are equipped with automatic extraction logic; when run, they will locate and extract the required CSV files on-the-fly.* --- ## System Requirements & Installation ### 1. Prerequisites - **Python**: Version `3.9` or higher is recommended. ### 2. Installation Create a virtual environment (recommended) and install the required Python packages: ```bash# Create and activate a virtual environmentpython3 -m venv venvsource venv/bin/activate # On Windows: venv\Scripts\activate # Install dependenciespip install -r requirements.txt``` ## Step-by-Step Reproduction Instructions Run the following commands from the top-level repository directory (`zenodo_repository/`): ### 1. Reproduce Figure 2 (Time-Series Prediction & NRMSE Scans) ```bashpython plot_fig_2.py``` ### 2. Reproduce Figure 3 (Dynamical Mean-Field Reservoir Simulations) ```bashpython plot_fig_3.py``` ### 3. Reproduce Figure 4 (Benchmark Tasks inside/outside Bistability) ```bashpython plot_fig_4.py``` ### 4. Reproduce Figures 6, 7, and 8 (Transfer Curves, IPC Profiles, NARMA Scans) ```bashpython generate_methods_figures.py``` *Note: All generated PDF plots will be automatically written to the `figures/` directory.* --- ## License & Citation If you use this code or dataset in your research, please cite the associated manuscript: > *Time-series learning in a many-body Rydberg system with emergent collective amplification.*