graphsc Config

This is the config for graphsc.

examples/tuning/cluster_graphsc/human_pbmc2_cell/pipeline_params_tuning_config.yaml

---
type: preprocessor
tune_mode: pipeline_params
pipeline_tuning_top_k: 3
parameter_tuning_freq_n: 10
pipeline:
  - type: filter.gene
    include:
      - FilterGenesPercentile
      - FilterGenesScanpyOrder
      - FilterGenesPlaceHolder
    default_params:
      FilterGenesScanpyOrder:
        order: [min_counts, min_cells, max_counts, max_cells]
        min_counts: 1270
        max_counts: 310000
        min_cells: 0.01
        max_cells: 0.99
  - type: filter.cell
    include: [FilterCellsScanpyOrder, FilterCellsPlaceHolder]
    default_params:
      FilterCellsScanpyOrder:
        order: [min_counts, min_genes, max_counts, max_genes]
        min_counts: 840
        max_counts: 8000
        min_genes: 0.01
        max_genes: 0.99
  - type: normalize
    include:
      - ColumnSumNormalize
      - ScTransform
      - Log1P
      - NormalizeTotal
      - NormalizePlaceHolder
      - NormalizeTotalLog1P
    default_params:
      ScTransform:
        processes_num: 8
  - type: filter.gene
    include:
      - HighlyVariableGenesLogarithmizedByMeanAndDisp
      - HighlyVariableGenesRawCount
      - HighlyVariableGenesLogarithmizedByTopGenes
      - FilterGenesTopK
      - FilterGenesRegression
      - FilterGenesPlaceHolder
    default_params:
      FilterGenesTopK:
        num_genes: 200
      FilterGenesRegression:
        num_genes: 200
      HighlyVariableGenesRawCount:
        n_top_genes: 200
      HighlyVariableGenesLogarithmizedByTopGenes:
        n_top_genes: 200
      # HighlyVariableGenesLogarithmizedByMeanAndDisp:
      #   max_mean: 20
      #   min_disp: 0.1
      #   min_mean: 0
  - type: feature.cell
    include:
      - WeightedFeaturePCA
      - WeightedFeatureSVD
      # - FeatureCellPlaceHolder
    params:
      out: feature.cell
      log_level: INFO
    default_params:
      WeightedFeaturePCA:
        split_name: train
      WeightedFeatureSVD:
        split_name: train
  - type: graph.cell
    target: CellFeatureGraph
    params:
      cell_feature_channel: feature.cell
  - type: misc
    target: SetConfig
    params:
      config_dict:
        feature_channel: CellFeatureGraph
        feature_channel_type: uns
        label_channel: Group
wandb:
  entity: xzy11632
  project: dance-dev
  method: grid  # try grid to provide a comprehensive search
  metric:
    name: acc  # val/acc
    goal: maximize