Merge branch 'main' of github.com:EngineeringSoftware/codeditor into …

…main

python/deltr/coditT5/CodeT5.py

@@ -121,14 +121,16 @@ def tokenizer_collate_fn(
         )
 
     def tokenize_collate_fn_predict(self, batch_data: List[Tuple[str, str, int]]):
-
         source_batch = [self.tokenize_sequence(t[0]) for t in batch_data]
         target_batch = [self.tokenize_sequence(t[1]) for t in batch_data]
         index_batch = [t[2] for t in batch_data]
         max_length = MAX_LENGTH
         batch_size = len(source_batch)
 
-        batched_input_ids, batched_labels_ids, = (
+        (
+            batched_input_ids,
+            batched_labels_ids,
+        ) = (
             [],
             [],
         )
@@ -205,7 +207,6 @@ def predict_dataloader(self):
 
 
 class CodeT5Module(pl.LightningModule):
-
     # Instantiate the model
     def __init__(
         self,

python/deltr/coditT5/save_pretrained.py

@@ -27,7 +27,6 @@ def locate_ckpt(ckpt_dir: Path) -> Optional[Path]:
 
 
 def add_tokens_to_tokenizer():
-
     from transformers import RobertaTokenizer
     from deltr.collector.diff_utils import EDIT_TOKENS
 

python/deltr/coditT5/utils.py

@@ -282,11 +282,9 @@ def __init__(
         )
 
     def __len__(self):
-
         return len(self.source_code)
 
     def __getitem__(self, index: int):
-
         return {
             "code": self.source_code[index],
             "context": self.context[index],
@@ -325,7 +323,6 @@ def __init__(
         self.__split_data_to_chunks__(source_code, context, tokenized_labels)
 
     def __len__(self):
-
         return len(self.tokenized_code_input)
 
     def __split_data_to_chunks__(self, source_code, context, tokenized_labels):
@@ -372,11 +369,9 @@ def __split_data_to_chunks__(self, source_code, context, tokenized_labels):
                 self.data_index.append(index)
                 self.labels.append(tokenized_label)
 
-
         return
 
     def __getitem__(self, index: int):
-
         return {
             "code": self.tokenized_code_input[index],
             "context": self.tokenized_context_input[index],
@@ -388,7 +383,6 @@ def __getitem__(self, index: int):
 def tokenize_and_align_labels(
     source_code: List[str], labels: List[int], tokenizer: Any
 ) -> List[List[int]]:
-
     tokenized_labels = []
 
     for code, label in zip(source_code, labels):
@@ -445,7 +439,6 @@ def __len__(self) -> int:
         return self.n_data
 
     def __getitem__(self, index: int) -> Tuple:
-
         if index < 0:
             index = self.n_data + index
 
@@ -495,7 +488,6 @@ def __len__(self) -> int:
         return self.n_data
 
     def __getitem__(self, index: int) -> Tuple:
-
         if index < 0:
             index = self.n_data + index