docs: fix dataset creation example

README.md

@@ -21,8 +21,8 @@ from flair.datasets import NER_GERMAN_BIOFID
 
 corpus = NER_GERMAN_BIOFID()
 
-with open("./germeval14/train.jsonl", "wt") as f_out:
-    for sentence in germeval_corpus.train:
+with open("./train.jsonl", "wt") as f_out:
+    for sentence in corpus.train:
         current_example = {
             "text": sentence.to_tokenized_string()
         }