Merge pull request #1384 from huggingface/encoding-qol

Quality of life enhancements in encoding + patch MLM masking

examples/run_glue.py

@@ -271,7 +271,7 @@ def load_and_cache_examples(args, task, tokenizer, evaluate=False):
         list(filter(None, args.model_name_or_path.split('/'))).pop(),
         str(args.max_seq_length),
         str(task)))
-    if os.path.exists(cached_features_file):
+    if os.path.exists(cached_features_file) and not args.overwrite_cache:
         logger.info("Loading features from cached file %s", cached_features_file)
         features = torch.load(cached_features_file)
     else: