Fix flaky test vision encoder-decoder generate (#28923)

Author: zucchini-nlp

tests/models/vision_encoder_decoder/test_modeling_vision_encoder_decoder.py

@@ -23,7 +23,6 @@
 
 from transformers import DonutProcessor, NougatProcessor, TrOCRProcessor
 from transformers.testing_utils import (
-    is_flaky,
     require_levenshtein,
     require_nltk,
     require_sentencepiece,
@@ -286,6 +285,8 @@ def check_encoder_decoder_model_generate(self, config, decoder_config, pixel_val
             enc_dec_model.config.eos_token_id = None
         if hasattr(enc_dec_model.config, "decoder") and hasattr(enc_dec_model.config.decoder, "eos_token_id"):
             enc_dec_model.config.decoder.eos_token_id = None
+        if hasattr(enc_dec_model.generation_config, "eos_token_id"):
+            enc_dec_model.generation_config.eos_token_id = None
         enc_dec_model.to(torch_device)
 
         inputs = pixel_values
@@ -324,10 +325,6 @@ def test_encoder_decoder_model_output_attentions(self):
         input_ids_dict = self.prepare_config_and_inputs()
         self.check_encoder_decoder_model_output_attentions(**input_ids_dict)
 
-    # FIXME @gante: flaky test
-    @is_flaky(
-        description="Fails on distributed runs e.g.: https://app.circleci.com/pipelines/github/huggingface/transformers/83611/workflows/666b01c9-1be8-4daa-b85d-189e670fc168/jobs/1078635/tests#failed-test-0"
-    )
     def test_encoder_decoder_model_generate(self):
         input_ids_dict = self.prepare_config_and_inputs()
         self.check_encoder_decoder_model_generate(**input_ids_dict)