scglue.models.glue.GLUETrainer.fit

GLUETrainer.fit(data, graph, val_split=None, data_batch_size=None, graph_batch_size=None, align_burnin=None, safe_burnin=True, max_epochs=None, patience=None, reduce_lr_patience=None, wait_n_lrs=None, random_seed=None, directory=None, plugins=None)

训练模型

参数:

• data (scglue.models.data.ArrayDataset) – Data dataset

• graph (scglue.models.data.GraphDataset) – Graph dataset

• val_split (typing.Optional[float]) – Validation split

• data_batch_size (typing.Optional[int]) – Number of samples in each data minibatch

• graph_batch_size (typing.Optional[int]) – Number of edges in each graph minibatch

• align_burnin (typing.Optional[int]) – Number of epochs to wait before starting alignment

• safe_burnin (bool) – Whether to postpone learning rate scheduling and earlystopping until after the burnin stage

• max_epochs (typing.Optional[int]) – Maximal number of epochs

• patience (typing.Optional[int]) – Patience of early stopping

• reduce_lr_patience (typing.Optional[int]) – Patience to reduce learning rate

• wait_n_lrs (typing.Optional[int]) – Wait n learning rate scheduling events before starting early stopping

• random_seed (typing.Optional[int]) – 随机种子

• directory (typing.Optional[os.PathLike]) – Directory to store checkpoints and tensorboard logs

• plugins (typing.Optional[typing.List[scglue.models.base.TrainingPlugin]]) – Optional list of training plugins

返回类型:

None