Fixed a couple of conditonals

mc4/mc4.py

@@ -376,13 +376,13 @@ class Mc4(datasets.GeneratorBasedBuilder):
                 for lang in self.config.languages
                 for index in range(_N_SHARDS_PER_SPLIT[lang][split])
             ]
-        if "train" in self.data_files:
+        if self.data_files and "train" in self.data_files:
             train_downloaded_files = self.data_files["train"]
             if not isinstance(train_downloaded_files, (tuple, list)):
                 train_downloaded_files = [train_downloaded_files]
         else:
             train_downloaded_files = dl_manager.download(data_urls["train"])
-        if "validation" in self.data_files:
+        if self.data_files and "validation" in self.data_files:
             validation_downloaded_files = self.data_files["validation"]
             if not isinstance(validation_downloaded_files, (tuple, list)):
                 validation_downloaded_files = [validation_downloaded_files]

run_mlm_flax_stream.py

@@ -588,7 +588,7 @@ if __name__ == "__main__":
     # Setup train state
     state = train_state.TrainState.create(apply_fn=model.__call__, params=model.params, tx=adamw)
     saved_step = 0
-    if "checkpoint" in model_args.model_name_or_path:
+    if model_args.model_name_or_path and "checkpoint" in model_args.model_name_or_path:
         params, opt_state, saved_step, args, data_collator = restore_checkpoint(model_args.model_name_or_path, state)
         # Create learning rate schedule
         warmup_fn = optax.linear_schedule(