PyTorch before and after 0.4

ckpt: Path of the checkpoint return: Checkpoint dict

pred: Prediction results with whatever size label: Label with whatever size return: [Accuracy] (A single value)

model: a FewShotREModel instance B: Batch size N: Num of classes for each batch K: Num of instances for each class in

logits: Logits with the size (..., class_num) label: Label with whatever size. return: [Loss] (A single value)

model: a FewShotREModel instance model_name: Name of the model B: Batch size N: Num of classes for each batch

sentence_encoder: Sentence encoder You need to set self.cost as your own loss function.

train_data_loader: DataLoader for training. val_data_loader: DataLoader for validating. test_data_loader: DataLoader for test

file_name: Json file storing the data in the following format { "P155": # relation id [

support: Inputs of the support set. query: Inputs of the query set. N: Num of classes K: Num of instances for each cl

support: Inputs of the support set. query: Inputs of the query set. N: Num of classes K: Num of instances for each cl

B: batch size. N: the number of relations for each batch K: the number of support instances for each relation Q: the