trial_outcome.LogisticRegression
- class pytrial.tasks.trial_outcome.logistic_regression.LogisticRegression(C=1.0, dual=False, solver='lbfgs', max_iter=100)[source]
Implement Logistic Regression model for clinical trial outcome prediction.
- Parameters
C (float) – Regularization strength for l2 norm; must be a positive float. Like in support vector machines, smaller values specify weaker regularization.
dual (bool) – Dual or primal formulation. Dual formulation is only implemented for l2 penalty with liblinear solver. Prefer dual=False when n_samples > n_features.
solver ({'newton-cg','lbfgs','liblinear','sag','saga'}) – Algorithm to use in the optimization problem. default=’lbfgs’.
max_iter (int (default=100)) – Maximum number of iterations taken for the solvers to converge.
- fit(train_data, valid_data)[source]
Train logistic regression model to predict clinical trial outcomes.
- Parameters
train_data (TrialOutcomeDatasetBase) – Training data, should be a TrialOutcomeDatasetBase object.
valid_data (TrialOutcomeDatasetBase) – Validation data, should be a TrialOutcomeDatasetBase object. Ignored for logistic regression model. Keep this parameter for compatibility with other models.
- load_model(checkpoint=None)[source]
Load the learned model from the disk.
- Parameters
checkpoint (str) – The checkpoint folder to load the learned model. The checkpoint under this folder should be model.ckpt.
- predict(test_data)[source]
Make clinical trial outcome predictions.
- Parameters
test_data (TrialOutcomeDatasetBase) – Testing data, should be a TrialOutcomeDatasetBase object.