Skip to content

python_setup.md

Latest commit

 

History

History
77 lines (38 loc) · 2.37 KB

python_setup.md

File metadata and controls

77 lines (38 loc) · 2.37 KB

Python Setup

Use this setup if you are working on the terminal of a local machine (Laptop, MacBook, etc.) and you already have the CSV files generated from a previous collection run.

Required Data Files

To use all analysis modes, you need two specific CSV files in your workspace:

  • File Sizes (file_size.csv): A CSV containing file sizes in GB. The header must include Algo, AOD_type, and columns formatted as {Events}events_level{N} (e.g., 7000events_level1).

  • Performance Metrics (All_Compression_Algo_metrics.csv): A CSV containing execution metrics. Please refer to the version provided in the workspaces/ folder to ensure your headers match the expected format.

Using Pre-generated Data

The repository includes existing data generated during our analysis. You can explore the tool immediately without running new ATLAS Athena collection jobs or providing your own CSVs:

  • Workspace Data: The workspaces/ folder contains pre-existing All_Compression_Algo_metrics.csv and file_size.csv.
  • Quick Start: Simply clone the repository and run any of the Analysis Modes. The tool will automatically use the provided CSVs in the workspace to generate reports.

Important

Note: You cannot run the Collection Mode in this setup as it requires the ATLAS Athena framework. To collect new data, please refer to the Athena Setup Guide.

Install Dependencies

Navigate to the root of the project and install the environment using Poetry or Pip:

poetry install

or 


pip install --user -r requirements.txt

Analysis modes

Run these commands from the project root. If you are using Poetry, add prefix the commands with poetry run.

python3 -m daod_analysis.main --workspace ./workspaces  --mode fluctuation




python3 -m daod_analysis.main --workspace ./workspaces --mode plot --metrics


python3 -m daod_analysis.main --workspace ./workspaces --mode plot --csv_path path/to/your/file_size.csv


python3 -m daod_analysis.main --workspace ./workspaces --mode plot --gain_loss


python3 -m daod_analysis.main --workspace ./workspaces --mode plot --impact

Note: We recommnend you to read View Athena Setup Guide to understand the code above above and how the data collection and project structure work.