Update pick_best_tokenizer.py

pick_best_tokenizer.py

@@ -12,7 +12,7 @@ class PickBestTokenizer(PreTrainedTokenizer):
         self.tokenizers_offsets = []
         self.vocab = {}
         self._vocab_size = sum(len(tokenizer) for tokenizer in self.tokenizers)
-        self.pad_token = '[0]'+self.tokenizers[0].pad_token
+        self.pad_token = '[0]'+(self.tokenizers[0].pad_token if self.tokenizers[0].pad_token else self.tokenizers[0].eos_token)
         
         offset = 0
         for i, tokenizer in enumerate(self.tokenizers):