DistilBERT is ExecuTorch compatible (huggingface#34475)

* DistillBERT is ExecuTorch compatible

* [run_slow] distilbert

* [run_slow] distilbert

---------

Co-authored-by: Guang Yang <[email protected]>

Author: 2 people

tests/models/distilbert/test_modeling_distilbert.py

@@ -30,6 +30,7 @@
     import torch
 
     from transformers import (
+        AutoTokenizer,
         DistilBertForMaskedLM,
         DistilBertForMultipleChoice,
         DistilBertForQuestionAnswering,
@@ -38,6 +39,7 @@
         DistilBertModel,
     )
     from transformers.models.distilbert.modeling_distilbert import _create_sinusoidal_embeddings
+    from transformers.pytorch_utils import is_torch_greater_or_equal_than_2_4
 
 
 class DistilBertModelTester:
@@ -420,3 +422,45 @@ def test_inference_no_head_absolute_embedding(self):
         )
 
         self.assertTrue(torch.allclose(output[:, 1:4, 1:4], expected_slice, atol=1e-4))
+
+    @slow
+    def test_export(self):
+        if not is_torch_greater_or_equal_than_2_4:
+            self.skipTest(reason="This test requires torch >= 2.4 to run.")
+
+        distilbert_model = "distilbert-base-uncased"
+        device = "cpu"
+        attn_implementation = "sdpa"
+        max_length = 64
+
+        tokenizer = AutoTokenizer.from_pretrained(distilbert_model)
+        inputs = tokenizer(
+            f"Paris is the {tokenizer.mask_token} of France.",
+            return_tensors="pt",
+            padding="max_length",
+            max_length=max_length,
+        )
+
+        model = DistilBertForMaskedLM.from_pretrained(
+            distilbert_model,
+            device_map=device,
+            attn_implementation=attn_implementation,
+        )
+
+        logits = model(**inputs).logits
+        eager_predicted_mask = tokenizer.decode(logits[0, 4].topk(5).indices)
+        self.assertEqual(
+            eager_predicted_mask.split(),
+            ["capital", "birthplace", "northernmost", "centre", "southernmost"],
+        )
+
+        exported_program = torch.export.export(
+            model,
+            args=(inputs["input_ids"],),
+            kwargs={"attention_mask": inputs["attention_mask"]},
+            strict=True,
+        )
+
+        result = exported_program.module().forward(inputs["input_ids"], inputs["attention_mask"])
+        exported_predicted_mask = tokenizer.decode(result.logits[0, 4].topk(5).indices)
+        self.assertEqual(eager_predicted_mask, exported_predicted_mask)