2/6/2024 0 Comments Nn tween model setsThis is identical to passing an empty dictionary.Ĭan_compile ( compiler_configs : dict | None = None ) → bool # Hyperparameters – Hyperparameters that will be used by the model (can be search spaces instead of fixed values). You can also pass your own evaluation function here as long as it follows formatting of the functions defined in folder. Options for quantile regression:įor more information on these options, see trics: ‘precision_weighted’, ‘recall’, ‘recall_macro’, ‘recall_micro’, ‘recall_weighted’, ‘log_loss’, ‘pac_score’] Options for regression: ‘roc_auc’, ‘roc_auc_ovo_macro’, ‘average_precision’, ‘precision’, ‘precision_macro’, ‘precision_micro’, If eval_metric = None, it is automatically chosen based on problem_type.ĭefaults to ‘accuracy’ for binary and multiclass classification and ‘root_mean_squared_error’ for regression. This only impacts model.score(), as eval_metric is not used during training. Metric by which predictions will be ultimately evaluated on test data. If None, will attempt to infer the problem type based on training data labels during training.Įval_metric ( or str, default = None) – is this a binary/multiclass classification or regression problem (options: ‘binary’, ‘multiclass’, ‘regression’). Problem_type ( str, default = None) – Type of prediction problem, i.e. If None, defaults to the model’s class name: self._class_._name_ The final model directory will be os.path.join(path, name) Name ( str, default = None) – Name of the subdirectory inside path where model will be saved. If None, a new unique time-stamped directory is chosen. Path ( str, default = None) – Directory location to store all outputs. AbstractModel ( path : str | None = None, name : str | None = None, problem_type : str | None = None, eval_metric : str | Scorer | None = None, hyperparameters : dict | None = None ) #Ībstract model implementation from which all AutoGluon models inherit. Additionally has special null image handling to improve performance in the presence of null images (aka image path of '') Note: null handling has not been compared to the built-in null handling of MultimodalPredictor yet.ĪbstractModel # class. Currently only supports 1 image column, with 1 image per sample. MultimodalPredictor that only uses image features. MultimodalPredictor that doesn't use image features PyTorch neural network models for classification/regression with tabular data.Ĭlass for fastai v1 neural network models that operate on tabular data. Models #Ībstract model implementation from which all AutoGluon models inherit. So this suffix was added to avoid overwriting the pre-existing model.Īn example would be “LightGBM_2”. “_x”: Indicates that the name without this added suffix already existed in a different model, Validation scores of distilled models should only be compared against other distilled models. Via a call to TabularPredictor’s distill method. “_DSTL”: Indicates the model was created through model distillation Often, this model can outperform the original model because of using more data during training,īut is usually weaker if the original was a bagged ensemble (“_BAG”), but with much faster inference speed. Usually, there will be another model with the same name as this model minus the “_FULL” suffix. This model will have no validation score because all of the data (train and validation) was used as training data. “_FULL”: Indicates the model has been refit via TabularPredictor’s refit_full method. Refer to “_FULL” for instructions on how to improve inference speed. This typically achieves a stronger result than any of the individual models alone,īut slows down inference speed significantly. “_BAG”: Indicates that the model is a bagged ensemble.Ī bagged ensemble contains multiple instances of the model (children) trained with different subsets of the data.ĭuring inference, these child models each predict on the data and their predictions are averaged in the final result. “/Tx”: Indicates that the model was trained via hyperparameter search (HPO). If a model lacks this suffix, then it is a base model and is at level 1 (“_L1”). “_Lx”: Indicates the stack level (x) the model is trained in, such as “_L1”, “_L2”, etc.Ī model with “_L1” suffix is a base model, meaning it does not depend on any other models. Models trained by TabularPredictor can have suffixes in their names that have special meanings. Here is the mapping of model types to their default names when trained: Model Name Suffixes # Hyperparameters takes in a dictionary of models, where each key is a model name, and the values are a list of dictionaries of model hyperparameters. To fit a model with TabularPredictor, you must specify it in the TabularPredictor.fit hyperparameters argument. This documentation is for advanced users, and is not comprehensive.įor a stable public API, refer to TabularPredictor. TabularPredictor.calibrate_decision_threshold.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |