MLproject

name: ape-classifier

entry_points:
  main:
    parameters:
      remote_server_uri: {type: str, default: projet-ape-mlflow.user.lab.sspcloud.fr}
      experiment_name: {type: str, default: test}
      run_name: {type: str, default: default}
      Y: {type: str, default: APE_NIV5}
      dim: {type: str, default: 180}
      ws: {type: str, default: 5}
      lr: {type: str, default: 0.2}
      epoch: {type: str, default: 50}
      wordNgrams: {type: str, default: 3}
      minn: {type: str, default: 3}
      maxn: {type: str, default: 4}
      minCount: {type: str, default: 3}
      bucket: {type: str, default: 2000000}
      loss: {type: str, default: ova}
      label_prefix: {type: str, default: __label__}
      text_feature: {type: str, default: LMNP}
      textual_features_1: {type: str, default: "activ_nat_lib_et"}
      textual_features_2: {type: str, default: "activ_sec_agri_et"}
      categorical_features_1: {type: str, default: AUTO}
      categorical_features_2: {type: str, default: NAT_SICORE}
      categorical_features_3: {type: str, default: SURF}
      categorical_features_4: {type: str, default: EVT_SICORE}
      categorical_features_5: {type: str, default: cj}
      categorical_features_6: {type: str, default: activ_perm_et}
      embedding_dim_1: {type: int, default: 1}
      embedding_dim_2: {type: int, default: 3}
      embedding_dim_3: {type: int, default: 3}
      embedding_dim_4: {type: int, default: 1}
      embedding_dim_5: {type: int, default: 3}
      embedding_dim_6: {type: int, default: 1}
      model_class: {type: str, default: fasttext}
      pre_training_weights: {type: str, default: camembert/camembert-base}
      start_month: {type: int, default: 1}
      start_year: {type: int, default: 2018}

    command: |
          python src/train.py \
            --remote_server_uri {remote_server_uri} \
            --experiment_name {experiment_name} \
            --run_name {run_name} \
            --Y {Y} \
            --dim {dim} \
            --ws {ws} \
            --lr {lr} \
            --epoch {epoch} \
            --wordNgrams {wordNgrams} \
            --minn {minn} \
            --maxn {maxn} \
            --minCount {minCount} \
            --bucket {bucket} \
            --loss {loss} \
            --label_prefix {label_prefix} \
            --text_feature {text_feature} \
            --textual_features_1 {textual_features_1} \
            --textual_features_2 {textual_features_2} \
            --categorical_features_1 {categorical_features_1} \
            --categorical_features_2 {categorical_features_2} \
            --categorical_features_3 {categorical_features_3} \
            --categorical_features_4 {categorical_features_4} \
            --categorical_features_5 {categorical_features_5} \
            --categorical_features_6 {categorical_features_6} \
            --embedding_dim_1 {embedding_dim_1} \
            --embedding_dim_2 {embedding_dim_2} \
            --embedding_dim_3 {embedding_dim_3} \
            --embedding_dim_4 {embedding_dim_4} \
            --embedding_dim_5 {embedding_dim_5} \
            --embedding_dim_6 {embedding_dim_6} \
            --model_class {model_class} \
            --pre_training_weights {pre_training_weights} \
            --start_month {start_month} \
            --start_year {start_year}