campa.tl.Predictor

class Predictor(exp, batch_size=None)

Predict results from trained model.

Parameters

• exp (Experiment) – Trained Experiment.

• batch_size (Optional[int]) – Batch size to use for prediction. If None, training batch size is used.

Methods

evaluate_model()	Predict val/test split and images.
get_representation(mpp_data[, rep])	Return representation from given mpp_data inputs.
predict(mpp_data[, save_dir, reps, mpp_params])	Predict representations from mpp_data.
predict_split(split[, img_ids, reps])	Predict data from train/val/test split of dataset that the model was trained with.