mpnice_embedding_ml_driver.py Command Help

Command: $SCHRODINGER/run mpnice_embedding_ml_gui_dir/mpnice_embedding_ml_driver.py

usage: $SCHRODINGER/run mpnice_embedding_ml_gui_dir/mpnice_embedding_ml_driver.py
       [-h] -mode {train,predict}
       [-models {ElasticNet,GCN,GIN,GlobalAttention,GraphSAGE,MultiTaskClassifier,MultiTaskRegressor,RandomForrestClassifier,RandomForrestRegressor,SAGPool,SVMRegressorLinear,SVMRegressorRBF,Set2Set,SortPool,TopK,TorchGraphConv,XGBoost} [{ElasticNet,GCN,GIN,GlobalAttention,GraphSAGE,MultiTaskClassifier,MultiTaskRegressor,RandomForrestClassifier,RandomForrestRegressor,SAGPool,SVMRegressorLinear,SVMRegressorRBF,Set2Set,SortPool,TopK,TorchGraphConv,XGBoost} ...]]
       [-featurizers [{LearnedGeoMPNICEFeaturizer_Hybrid,LearnedGeoMPNICEFeaturizer_Hybrid_c,LearnedGeoMPNICEFeaturizer_Inorganic,LearnedGeoMPNICEFeaturizer_Inorganic_c,LearnedGeoMPNICEFeaturizer_Organic,LearnedGeoMPNICEFeaturizer_Organic_c,MPNICEEmbeddingFeaturizer_Hybrid,MPNICEEmbeddingFeaturizer_Inorganic,MPNICEEmbeddingFeaturizer_Organic} ...]]
       [-target TARGET [TARGET ...]]
       [-additional_features ADDITIONAL_FEATURES [ADDITIONAL_FEATURES ...]]
       [-hyperparameter HYPERPARAMETER] [-time TIME] [-test_size TEST_SIZE]
       [-model [MODEL ...]] [-downsample DOWNSAMPLE]
       [-cross_validation CROSS_VALIDATION] [-split_seed SPLIT_SEED]
       [-HOST <hostname>] [-D] [-VIEWNAME <viewname>] [-JOBNAME JOBNAME]
       input_file

Driver for training ML Models to predict structure property relation based on
MPNICE Embedding. Copyright Schrodinger, LLC. All rights reserved.

positional arguments:
  input_file            Input MAE/MAEGZ file with structures and properties

options:
  -h, -help             Show this help message and exit.
  -mode {train,predict}
                        Use train to train or predict to predict (default:
                        None)
  -models {ElasticNet,GCN,GIN,GlobalAttention,GraphSAGE,MultiTaskClassifier,MultiTaskRegressor,RandomForrestClassifier,RandomForrestRegressor,SAGPool,SVMRegressorLinear,SVMRegressorRBF,Set2Set,SortPool,TopK,TorchGraphConv,XGBoost} [{ElasticNet,GCN,GIN,GlobalAttention,GraphSAGE,MultiTaskClassifier,MultiTaskRegressor,RandomForrestClassifier,RandomForrestRegressor,SAGPool,SVMRegressorLinear,SVMRegressorRBF,Set2Set,SortPool,TopK,TorchGraphConv,XGBoost} ...]
                        List of Machine learning models for training (default:
                        None)
  -featurizers [{LearnedGeoMPNICEFeaturizer_Hybrid,LearnedGeoMPNICEFeaturizer_Hybrid_c,LearnedGeoMPNICEFeaturizer_Inorganic,LearnedGeoMPNICEFeaturizer_Inorganic_c,LearnedGeoMPNICEFeaturizer_Organic,LearnedGeoMPNICEFeaturizer_Organic_c,MPNICEEmbeddingFeaturizer_Hybrid,MPNICEEmbeddingFeaturizer_Inorganic,MPNICEEmbeddingFeaturizer_Organic} ...]
                        MPNICE featurizers (default: None)
  -target TARGET [TARGET ...]
                        List of target property for training/prediction
                        (default: None)
  -additional_features ADDITIONAL_FEATURES [ADDITIONAL_FEATURES ...]
                        List of additional property names to include in
                        training (default: None)
  -hyperparameter HYPERPARAMETER
                        Hyperparameter tuning iterations. Either
                        -hyperparameter or -time must be provided. Both cannot
                        be provided. (default: None)
  -time TIME            Time limit in hours for training. Either
                        -hyperparameter or -time must be provided. Both cannot
                        be provided. (default: None)
  -test_size TEST_SIZE  Proportion of dataset for test split (0.0-1.0)
                        (default: 0.1)
  -model [MODEL ...]    Trained model (.model) for prediction (default: None)
  -downsample DOWNSAMPLE
                        Downsample limit for hyperparameter tuning (default:
                        10000)
  -cross_validation CROSS_VALIDATION
                        Number of cross validation splits (default: 5)
  -split_seed SPLIT_SEED
                        Random seed for train/test splitting (default: 1234)

Job Control Options:
  -HOST <hostname>      Run job remotely on the indicated host entry.
                        (default: localhost)
  -D, -DEBUG            Show details of Job Control operation. (default:
                        False)
  -VIEWNAME <viewname>  Specifies viewname used in job filtering in maestro.
                        (default: False)
  -JOBNAME JOBNAME      Provide an explicit name for the job. (default: None)