Fix length of IterableDatasetShard and add test

src/transformers/trainer_pt_utils.py

@@ -775,9 +775,9 @@ def __iter__(self):
     def __len__(self):
         # Will raise an error if the underlying dataset is not sized.
         if self.drop_last:
-            return len(self.dataset) // self.num_processes
+            return (len(self.dataset) // (self.batch_size * self.num_processes)) * self.batch_size
         else:
-            return math.ceil(len(self.dataset) / self.num_processes)
+            return math.ceil(len(self.dataset) / (self.batch_size * self.num_processes)) * self.batch_size
 
 
 # In order to keep `trainer.py` compact and easy to understand, place any secondary PT Trainer

tests/test_trainer_utils.py

@@ -355,6 +355,34 @@ def test_iterable_dataset_shard(self):
         self.check_iterable_dataset_shard(dataset, 4, drop_last=True, num_processes=3, epoch=42)
         self.check_iterable_dataset_shard(dataset, 4, drop_last=False, num_processes=3, epoch=42)
 
+    def test_iterable_dataset_shard_with_length(self):
+        sampler_shards = [
+            IterableDatasetShard(list(range(100)), batch_size=4, drop_last=True, num_processes=2, process_index=i)
+            for i in range(2)
+        ]
+
+        # Build expected shards: each process will have batches of size 4 until there is not enough elements to
+        # form two full batches (so we stop at 96 = (100 // (4 * 2)) * 4)
+        expected_shards = [[], []]
+        current_shard = 0
+        for i in range(0, 96, 4):
+            expected_shards[current_shard].extend(list(range(i, i + 4)))
+            current_shard = 1 - current_shard
+
+        self.assertListEqual([list(shard) for shard in sampler_shards], expected_shards)
+        self.assertListEqual([len(shard) for shard in sampler_shards], [len(shard) for shard in expected_shards])
+
+        sampler_shards = [
+            IterableDatasetShard(list(range(100)), batch_size=4, drop_last=False, num_processes=2, process_index=i)
+            for i in range(2)
+        ]
+        # When drop_last=False, we get two last full batches by looping back to the beginning.
+        expected_shards[0].extend(list(range(96, 100)))
+        expected_shards[1].extend(list(range(0, 4)))
+
+        self.assertListEqual([list(shard) for shard in sampler_shards], expected_shards)
+        self.assertListEqual([len(shard) for shard in sampler_shards], [len(shard) for shard in expected_shards])
+
     def check_shard_sampler(self, dataset, batch_size, drop_last, num_processes=2):
         shards = [
             ShardSampler(

utils/tests_fetcher.py

@@ -281,6 +281,7 @@ def create_reverse_dependency_map():
         "test_trainer_distributed.py",
         "test_trainer_tpu.py",
     ],
+    "train_pt_utils.py": "test_trainer_utils.py",
     "utils/versions.py": "test_versions_utils.py",
 }