gobbli.experiment package¶
Submodules¶
Module contents¶
-
class
gobbli.experiment.
ClassificationExperiment
(*args, **kwargs)[source]¶ Bases:
gobbli.experiment.base.BaseExperiment
Run a classification experiment. This entails training a model to make predictions given some input.
The experiment will, for each combination of model hyperparameters, train the model on a training set and evaluate it on a validation set. The best combination of hyperparameters will be retrained on the combined training/validation sets and evaluated on the test set. After completion, the experiment will return
ClassificationExperimentResults
, which will allow the user to examine the results in various ways.-
data_dir
()¶ - Return type
Path
- Returns
The main data directory unique to this experiment.
-
property
metadata_path
¶ - Return type
Path
- Returns
The path to the experiment’s metadata file containing information about the experiment parameters.
-
run
(dataset_split=None, seed=1, train_batch_size=32, valid_batch_size=32, test_batch_size=32, num_train_epochs=5)[source]¶ Run the experiment.
- Parameters
dataset_split¶ (
Union
[Tuple
[float
,float
],Tuple
[float
,float
,float
],None
]) – A tuple describing the proportion of the dataset to be added to the train/validation/test splits. If the experiment uses an explicit test set (passesBaseExperiment.params.test_dataset
), this should be a 2-tuple describing the train/validation split. Otherwise, it should be a 3-tuple describing the train/validation/test split. The tuple must sum to 1.seed¶ (
int
) – Random seed to be used for dataset splitting for reproducibility.train_batch_size¶ (
int
) – Number of observations per batch on the training dataset.valid_batch_size¶ (
int
) – Number of observations per batch on the validation dataset.test_batch_size¶ (
int
) – Number of observations per batch on the test dataset.num_train_epochs¶ (
int
) – Number of epochs to use for training.
- Return type
- Returns
The results of the experiment.
-