tweak

src/kwja/modules/seq2seq.py

@@ -102,7 +102,7 @@ def predict_step(self, batch: Any, batch_idx: int, dataloader_idx: int = 0) -> A
             if self.use_forced_decoding
             else None,
             **self.hparams.decoding,
-        )  # (b, max_tgt_len)
+        )
         return {
             "example_ids": batch["example_ids"] if "example_ids" in batch else [],
             "seq2seq_predictions": generations,

src/kwja/utils/seq2seq_format.py

@@ -148,6 +148,7 @@ def format_to_sent(self, text: str) -> Sentence:
                 )
                 canon = f'"代表表記:{canon}"' if canon != NO_CANON_TOKEN else "NIL"
 
+                # 例外処理
                 if surf == " " and reading == "\u3000" and lemma == "\u3000":
                     surf = "\u3000"
                 if surf == "°C":