Clustering.yaml

id: clustering_benchmark
description: Clustering benchmark on Iris and Penguins Dataset.
version: 1.0
benchmarker: "OmniBenchmark Team at Robinsons Lab"
storage: https://play.min.io
benchmark_yaml_spec: 0.01
storage_api: S3
storage_bucket_name: clustering_benchmark
software_backend: envmodules
software_environments:
  R:
    description: "R 4.3.3 with gfbf-2023 toolchain"
    easyconfig: R-4.3.3-gfbf-2023b.eb
    envmodule: R/4.3.3-gfbf-2023b
    conda: R_4.3.3_try.yaml
    apptainer: http://registry.ch/R_4.3.3-gfbf-2023b.sif
  python:
    description: "Python3.6.3"
    easyconfig: Python-3.6.3-foss-2017b.eb
    envmodule: python/3.6.3-foss-2017b
    conda: python_v363_test.yaml
    apptainer: http://registry.ch/python_vX-gfbf-2023b.sif
stages:
  - id: data
    modules:
      - id: iris
        name: "Iris Dataset"
        software_environment: "python"
        repository:
          url: https://github.com/omnibenchmark-example/iris.git
          commit: 47c63f0
      - id: penguins
        name: "Penguins Dataset"
        software_environment: "python"
        repository:
          url: https://github.com/omnibenchmark-example/penguins.git
          commit: 9032478
    outputs:
      - id: data.features
        path: "{input}/{stage}/{module}/{params}/{dataset}.features.csv"
      - id: data.labels
        path: "{input}/{stage}/{module}/{params}/{dataset}.labels.csv"

  - id: distances
    modules:
      - id: D1
        software_environment: "python"
        parameters:
          - values: ["--measure", "cosine"]
          - values: ["--measure", "euclidean"]
          - values: ["--measure", "manhattan"]
          - values: ["--measure", "chebyshev"]
        repository:
          url: https://github.com/omnibenchmark-example/distance.git
          commit: dd99d4f
    inputs:
      - entries:
          - data.features
    outputs:
      - id: distances
        path: "{input}/{stage}/{module}/{params}/{dataset}.distances.csv"

  - id: methods
    modules:
      - id: kmeans
        software_environment: "python"
        repository:
          url: https://github.com/omnibenchmark-example/kmeans.git
          commit: 049c8b1
      - id: ward
        software_environment: "R"
        repository:
          url: https://github.com/omnibenchmark-example/ward.git
          commit: 976e3f3
    inputs:
      - entries:
          - distances
    outputs:
      - id: methods.clusters
        path: "{input}/{stage}/{module}/{params}/{dataset}.clusters.csv"

  - id: metrics
    modules:
      - id: ari
        software_environment: "R"
        repository:
          url: https://github.com/omnibenchmark-example/ari.git
          commit: 72708f0
      - id: accuracy
        software_environment: "R"
        repository:
          url: https://github.com/omnibenchmark-example/accuracy.git
          commit: e26b32f
    inputs:
      - entries:
          - methods.clusters
          - data.labels
    outputs:
      - id: metrics.mapping
        path: "{input}/{stage}/{module}/{params}/{dataset}.metrics.txt"