DashAI Backend

This reference details all the backend components available in DashAI. For more information on how to add models, please refer to the user_guide.

Datasets

DashAI.back.dataloaders.classes.dashai_dataset.DashAIDataset(table)

DashAI dataset wrapper for Huggingface datasets with extra metadata.

Tasks

DashAI.back.tasks.BaseTask()	Base class for DashAI compatible tasks.
DashAI.back.tasks.ImageClassificationTask()	Base class for image classification tasks.
DashAI.back.tasks.RegressionTask()	Base class for regression tasks.
DashAI.back.tasks.TabularClassificationTask()	Base class for tabular classification tasks.
DashAI.back.tasks.TextClassificationTask()	Base class for Text Classification Task.
DashAI.back.tasks.TranslationTask()	Base class for translation task.

Models

DashAI.back.models.BaseModel()	Abstract class of all machine learning models.
DashAI.back.models.SVC(**kwargs)	Scikit-learn's Support Vector Machine (SVM) classifier wrapper for DashAI.
DashAI.back.models.BagOfWordsTextClassificationModel(...)	Text classification meta-model.
DashAI.back.models.DecisionTreeClassifier(...)	Scikit-learn's Decision Tree Classifier wrapper for DashAI.
DashAI.back.models.DistilBertTransformer([model])	Pre-trained transformer DistilBERT allowing English text classification.
DashAI.back.models.DummyClassifier(**kwargs)	Scikit-learn's DummyClassifier wrapper for DashAI.
DashAI.back.models.GradientBoostingR(**kwargs)	Scikit-learn's Ridge Regression wrapper for DashAI.
DashAI.back.models.HistGradientBoostingClassifier(...)	Scikit-learn's HistGradientBoostingRegressor wrapper for DashAI.
DashAI.back.models.KNeighborsClassifier(**kwargs)	Scikit-learn's K-Nearest Neighbors (KNN) classifier wrapper for DashAI.
DashAI.back.models.LinearRegression(**kwargs)	Scikit-learn's Linear Regression wrapper for DashAI.
DashAI.back.models.LinearSVR(**kwargs)	Scikit-learn's Linear Support Vector Regression (LinearSVR) wrapper for DashAI.
DashAI.back.models.LogisticRegression(**kwargs)	Scikit-learn's Logistic Regression wrapper for DashAI.
DashAI.back.models.MLPRegression(**kwargs)	Scikit-learn's MLP Regression wrapper for DashAI.
DashAI.back.models.OpusMtEnESTransformer([model])	Pre-trained transformer for english-spanish translation.
DashAI.back.models.RandomForestClassifier(...)	Scikit-learn's Random Forest classifier wrapper for DashAI.
DashAI.back.models.RandomForestRegression(...)	Scikit-learn's Ridge Regression wrapper for DashAI.
DashAI.back.models.RidgeRegression(**kwargs)	Scikit-learn's Ridge regression wrapper for DashAI.
DashAI.back.models.ViTTransformer([model])	Pre-trained Vision Transformer (ViT) for image classification.

Dataloaders

DashAI.back.dataloaders.CSVDataLoader()	Data loader for tabular data in CSV files.
DashAI.back.dataloaders.ExcelDataLoader()	Data loader for tabular data in Excel files.
DashAI.back.dataloaders.ImageDataLoader()	Data loader for data from image files.
DashAI.back.dataloaders.JSONDataLoader()	Data loader for tabular data in JSON files.

Metrics

DashAI.back.metrics.BaseMetric()	Abstract class of all metrics.
DashAI.back.metrics.F1()	F1 score to classification tasks.
DashAI.back.metrics.Accuracy()	Accuracy metric to classification tasks.
DashAI.back.metrics.Precision()	Precision metric to classification tasks.
DashAI.back.metrics.Recall()	Recall metric to classification tasks.
DashAI.back.metrics.Bleu()	A class for calculating BLEU scores between source and target sentences.
DashAI.back.metrics.MAE()	Mean Absolute Error metric for regression tasks.
DashAI.back.metrics.RMSE()	Root Mean Squared Error metric for regression tasks.

Optimizers

DashAI.back.optimizers.BaseOptimizer()	Abstract class of all hyperparameter's Optimizers.
DashAI.back.optimizers.OptunaOptimizer([...])
DashAI.back.optimizers.HyperOptOptimizer([...])

Jobs

DashAI.back.job.ExplainerJob(**kwargs)	ExplainerJob class to calculate explanations.
DashAI.back.job.ModelJob(**kwargs)	ModelJob class to run the model training.
DashAI.back.job.PredictJob(**kwargs)	PredictJob class to run the prediction.

Explainers

DashAI.back.explainability.KernelShap(model)	Kernel SHAP is a model-agnostic explainability method for approximating SHAP values to explain the output of machine learning model by attributing contributions of each feature to the model's prediction.
DashAI.back.explainability.PartialDependence(model)	PartialDependence is a model-agnostic explainability method that shows the average prediction of a machine learning model for each possible value of a feature.
DashAI.back.explainability.PermutationFeatureImportance(model)	Permutation Feature Importance is a explanation method to asses the importance of each feature in a model by evaluating how much the model's performance decreases when the values of a specific feature are randomly shuffled.