[github CI] add a multi-gpu job for all example tests

.github/workflows/self-scheduled.yml

@@ -1,3 +1,8 @@
+# configuration notes:
+#
+# - `source .env/bin/activate` is currently needed to be run first thing first in each step. Otherwise
+# the step uses the system-wide python interpreter.
+
 name: Self-hosted runner (scheduled)
 
 on:
@@ -227,7 +232,7 @@ jobs:
           python -c "import torch; print('Cuda available:', torch.cuda.is_available())"
           python -c "import torch; print('Number of GPUs available:', torch.cuda.device_count())"
 
-      - name: Run all tests on GPU
+      - name: Run all tests on multi-GPU
         env:
           OMP_NUM_THREADS: 1
           RUN_SLOW: yes
@@ -238,8 +243,20 @@ jobs:
       - name: Failure short reports
         if: ${{ always() }}
         run: cat reports/tests_torch_multiple_gpu_failures_short.txt
-
-      - name: Run all pipeline tests on GPU
+
+      - name: Run ported examples tests on multi-GPU
+        env:
+          OMP_NUM_THREADS: 1
+          RUN_SLOW: yes
+        run: |
+          source .env/bin/activate
+          python -m pytest -n 1 --dist=loadfile -s --make-reports=examples_torch_multiple_gpu examples
+
+      - name: Failure short reports
+        if: ${{ always() }}
+        run: cat reports/examples_torch_multiple_gpu_failures_short.txt
+
+      - name: Run all pipeline tests on multi-GPU
         if: ${{ always() }}
         env:
           TF_FORCE_GPU_ALLOW_GROWTH: "true"
@@ -306,7 +323,7 @@ jobs:
           TF_CPP_MIN_LOG_LEVEL=3 python -c "import tensorflow as tf; print('TF GPUs available:', bool(tf.config.list_physical_devices('GPU')))"
           TF_CPP_MIN_LOG_LEVEL=3 python -c "import tensorflow as tf; print('Number of TF GPUs available:', len(tf.config.list_physical_devices('GPU')))"
 
-      - name: Run all tests on GPU
+      - name: Run all tests on multi-GPU
         env:
           OMP_NUM_THREADS: 1
           RUN_SLOW: yes
@@ -318,7 +335,7 @@ jobs:
         if: ${{ always() }}
         run: cat reports/tests_tf_multiple_gpu_failures_short.txt
 
-      - name: Run all pipeline tests on GPU
+      - name: Run all pipeline tests on multi-GPU
         if: ${{ always() }}
         env:
           TF_FORCE_GPU_ALLOW_GROWTH: "true"

examples/bert-loses-patience/test_run_glue_with_pabee.py

@@ -4,7 +4,7 @@
 from unittest.mock import patch
 
 import run_glue_with_pabee
-from transformers.testing_utils import TestCasePlus
+from transformers.testing_utils import TestCasePlus, require_torch_non_multigpu_but_fix_me
 
 
 logging.basicConfig(level=logging.DEBUG)
@@ -20,6 +20,7 @@ def get_setup_file():
 
 
 class PabeeTests(TestCasePlus):
+    @require_torch_non_multigpu_but_fix_me
     def test_run_glue(self):
         stream_handler = logging.StreamHandler(sys.stdout)
         logger.addHandler(stream_handler)

examples/deebert/test_glue_deebert.py

@@ -5,7 +5,7 @@
 from unittest.mock import patch
 
 import run_glue_deebert
-from transformers.testing_utils import slow
+from transformers.testing_utils import require_torch_non_multigpu_but_fix_me, slow
 
 
 logging.basicConfig(level=logging.DEBUG)
@@ -26,6 +26,7 @@ def setup(self) -> None:
         logger.addHandler(stream_handler)
 
     @slow
+    @require_torch_non_multigpu_but_fix_me
     def test_glue_deebert_train(self):
 
         train_args = """

examples/rag/test_distributed_retriever.py

@@ -16,6 +16,7 @@
 from transformers.configuration_rag import RagConfig
 from transformers.file_utils import is_datasets_available, is_faiss_available, is_psutil_available, is_torch_available
 from transformers.retrieval_rag import CustomHFIndex
+from transformers.testing_utils import require_torch_non_multigpu_but_fix_me
 from transformers.tokenization_bart import BartTokenizer
 from transformers.tokenization_bert import VOCAB_FILES_NAMES as DPR_VOCAB_FILES_NAMES
 from transformers.tokenization_dpr import DPRQuestionEncoderTokenizer
@@ -178,6 +179,7 @@ def get_dummy_custom_hf_index_retriever(self, init_retrieval: bool, from_disk: b
             retriever.init_retrieval(port)
         return retriever
 
+    @require_torch_non_multigpu_but_fix_me
     def test_pytorch_distributed_retriever_retrieve(self):
         n_docs = 1
         retriever = self.get_dummy_pytorch_distributed_retriever(init_retrieval=True)
@@ -193,6 +195,7 @@ def test_pytorch_distributed_retriever_retrieve(self):
         self.assertEqual(doc_dicts[1]["id"][0], "0")  # max inner product is reached with first doc
         self.assertListEqual(doc_ids.tolist(), [[1], [0]])
 
+    @require_torch_non_multigpu_but_fix_me
     def test_custom_hf_index_retriever_retrieve(self):
         n_docs = 1
         retriever = self.get_dummy_custom_hf_index_retriever(init_retrieval=True, from_disk=False)
@@ -208,6 +211,7 @@ def test_custom_hf_index_retriever_retrieve(self):
         self.assertEqual(doc_dicts[1]["id"][0], "0")  # max inner product is reached with first doc
         self.assertListEqual(doc_ids.tolist(), [[1], [0]])
 
+    @require_torch_non_multigpu_but_fix_me
     def test_custom_pytorch_distributed_retriever_retrieve_from_disk(self):
         n_docs = 1
         retriever = self.get_dummy_custom_hf_index_retriever(init_retrieval=True, from_disk=True)

examples/seq2seq/test_bash_script.py

@@ -13,7 +13,7 @@
 from finetune import SummarizationModule, main
 from transformers import MarianMTModel
 from transformers.file_utils import cached_path
-from transformers.testing_utils import TestCasePlus, require_torch_gpu, slow
+from transformers.testing_utils import TestCasePlus, require_torch_gpu, require_torch_non_multigpu_but_fix_me, slow
 from utils import load_json
 
 
@@ -32,13 +32,15 @@ def setUp(self):
 
     @slow
     @require_torch_gpu
+    @require_torch_non_multigpu_but_fix_me
     def test_model_download(self):
         """This warms up the cache so that we can time the next test without including download time, which varies between machines."""
         MarianMTModel.from_pretrained(MARIAN_MODEL)
 
     # @timeout_decorator.timeout(1200)
     @slow
     @require_torch_gpu
+    @require_torch_non_multigpu_but_fix_me
     def test_train_mbart_cc25_enro_script(self):
         env_vars_to_replace = {
             "$MAX_LEN": 64,
@@ -127,6 +129,7 @@ class TestDistilMarianNoTeacher(TestCasePlus):
     @timeout_decorator.timeout(600)
     @slow
     @require_torch_gpu
+    @require_torch_non_multigpu_but_fix_me
     def test_opus_mt_distill_script(self):
         data_dir = f"{self.test_file_dir_str}/test_data/wmt_en_ro"
         env_vars_to_replace = {

examples/seq2seq/test_datasets.py

@@ -11,7 +11,7 @@
 from test_seq2seq_examples import ARTICLES, BART_TINY, MARIAN_TINY, MBART_TINY, SUMMARIES, T5_TINY, make_test_data_dir
 from transformers import AutoTokenizer
 from transformers.modeling_bart import shift_tokens_right
-from transformers.testing_utils import TestCasePlus, slow
+from transformers.testing_utils import TestCasePlus, require_torch_non_multigpu_but_fix_me, slow
 from utils import FAIRSEQ_AVAILABLE, DistributedSortishSampler, LegacySeq2SeqDataset, Seq2SeqDataset
 
 
@@ -30,6 +30,7 @@ class TestAll(TestCasePlus):
         ],
     )
     @slow
+    @require_torch_non_multigpu_but_fix_me
     def test_seq2seq_dataset_truncation(self, tok_name):
         tokenizer = AutoTokenizer.from_pretrained(tok_name)
         tmp_dir = make_test_data_dir(tmp_dir=self.get_auto_remove_tmp_dir())
@@ -69,6 +70,7 @@ def test_seq2seq_dataset_truncation(self, tok_name):
             break  # No need to test every batch
 
     @parameterized.expand([BART_TINY, BERT_BASE_CASED])
+    @require_torch_non_multigpu_but_fix_me
     def test_legacy_dataset_truncation(self, tok):
         tokenizer = AutoTokenizer.from_pretrained(tok)
         tmp_dir = make_test_data_dir(tmp_dir=self.get_auto_remove_tmp_dir())
@@ -93,6 +95,7 @@ def test_legacy_dataset_truncation(self, tok):
             assert max_len_target > trunc_target  # Truncated
             break  # No need to test every batch
 
+    @require_torch_non_multigpu_but_fix_me
     def test_pack_dataset(self):
         tokenizer = AutoTokenizer.from_pretrained("facebook/mbart-large-cc25")
 
@@ -111,6 +114,7 @@ def test_pack_dataset(self):
         assert orig_paths == new_paths
 
     @pytest.mark.skipif(not FAIRSEQ_AVAILABLE, reason="This test requires fairseq")
+    @require_torch_non_multigpu_but_fix_me
     def test_dynamic_batch_size(self):
         if not FAIRSEQ_AVAILABLE:
             return
@@ -135,6 +139,7 @@ def test_dynamic_batch_size(self):
         if failures:
             raise AssertionError(f"too many tokens in {len(failures)} batches")
 
+    @require_torch_non_multigpu_but_fix_me
     def test_sortish_sampler_reduces_padding(self):
         ds, _, tokenizer = self._get_dataset(max_len=512)
         bs = 2
@@ -174,6 +179,7 @@ def _get_dataset(self, n_obs=1000, max_len=128):
         )
         return ds, max_tokens, tokenizer
 
+    @require_torch_non_multigpu_but_fix_me
     def test_distributed_sortish_sampler_splits_indices_between_procs(self):
         ds, max_tokens, tokenizer = self._get_dataset()
         ids1 = set(DistributedSortishSampler(ds, 256, num_replicas=2, rank=0, add_extra_examples=False))
@@ -189,6 +195,7 @@ def test_distributed_sortish_sampler_splits_indices_between_procs(self):
             PEGASUS_XSUM,
         ],
     )
+    @require_torch_non_multigpu_but_fix_me
     def test_dataset_kwargs(self, tok_name):
         tokenizer = AutoTokenizer.from_pretrained(tok_name)
         if tok_name == MBART_TINY:

examples/seq2seq/test_fsmt_bleu_score.py

@@ -19,7 +19,13 @@
 
 from parameterized import parameterized
 from transformers import FSMTForConditionalGeneration, FSMTTokenizer
-from transformers.testing_utils import get_tests_dir, require_torch, slow, torch_device
+from transformers.testing_utils import (
+    get_tests_dir,
+    require_torch,
+    require_torch_non_multigpu_but_fix_me,
+    slow,
+    torch_device,
+)
 from utils import calculate_bleu
 
 
@@ -48,6 +54,7 @@ def get_model(self, mname):
         ]
     )
     @slow
+    @require_torch_non_multigpu_but_fix_me
     def test_bleu_scores(self, pair, min_bleu_score):
         # note: this test is not testing the best performance since it only evals a small batch
         # but it should be enough to detect a regression in the output quality

examples/seq2seq/test_make_student.py

@@ -4,7 +4,7 @@
 from make_student import create_student_by_copying_alternating_layers
 from transformers import AutoConfig
 from transformers.file_utils import cached_property
-from transformers.testing_utils import require_torch
+from transformers.testing_utils import require_torch, require_torch_non_multigpu_but_fix_me
 
 
 TINY_BART = "sshleifer/bart-tiny-random"
@@ -17,23 +17,28 @@ class MakeStudentTester(unittest.TestCase):
     def teacher_config(self):
         return AutoConfig.from_pretrained(TINY_BART)
 
+    @require_torch_non_multigpu_but_fix_me
     def test_valid_t5(self):
         student, *_ = create_student_by_copying_alternating_layers(TINY_T5, tempfile.mkdtemp(), e=1, d=1)
         self.assertEqual(student.config.num_hidden_layers, 1)
 
+    @require_torch_non_multigpu_but_fix_me
     def test_asymmetric_t5(self):
         student, *_ = create_student_by_copying_alternating_layers(TINY_T5, tempfile.mkdtemp(), e=1, d=None)
 
+    @require_torch_non_multigpu_but_fix_me
     def test_same_decoder_small_encoder(self):
         student, *_ = create_student_by_copying_alternating_layers(TINY_BART, tempfile.mkdtemp(), e=1, d=None)
         self.assertEqual(student.config.encoder_layers, 1)
         self.assertEqual(student.config.decoder_layers, self.teacher_config.encoder_layers)
 
+    @require_torch_non_multigpu_but_fix_me
     def test_small_enc_small_dec(self):
         student, *_ = create_student_by_copying_alternating_layers(TINY_BART, tempfile.mkdtemp(), e=1, d=1)
         self.assertEqual(student.config.encoder_layers, 1)
         self.assertEqual(student.config.decoder_layers, 1)
 
+    @require_torch_non_multigpu_but_fix_me
     def test_raises_assert(self):
         with self.assertRaises(AssertionError):
             create_student_by_copying_alternating_layers(TINY_BART, tempfile.mkdtemp(), e=None, d=None)

examples/seq2seq/test_seq2seq_examples.py

@@ -19,7 +19,14 @@
 from run_eval_search import run_search
 from transformers import AutoConfig, AutoModelForSeq2SeqLM
 from transformers.hf_api import HfApi
-from transformers.testing_utils import CaptureStderr, CaptureStdout, TestCasePlus, require_torch_gpu, slow
+from transformers.testing_utils import (
+    CaptureStderr,
+    CaptureStdout,
+    TestCasePlus,
+    require_torch_gpu,
+    require_torch_non_multigpu_but_fix_me,
+    slow,
+)
 from utils import ROUGE_KEYS, label_smoothed_nll_loss, lmap, load_json
 
 
@@ -126,6 +133,7 @@ def setUpClass(cls):
 
     @slow
     @require_torch_gpu
+    @require_torch_non_multigpu_but_fix_me
     def test_hub_configs(self):
         """I put require_torch_gpu cause I only want this to run with self-scheduled."""
 
@@ -143,10 +151,12 @@ def test_hub_configs(self):
                 failures.append(m)
         assert not failures, f"The following models could not be loaded through AutoConfig: {failures}"
 
+    @require_torch_non_multigpu_but_fix_me
     def test_distill_no_teacher(self):
         updates = dict(student_encoder_layers=2, student_decoder_layers=1, no_teacher=True)
         self._test_distiller_cli(updates)
 
+    @require_torch_non_multigpu_but_fix_me
     def test_distill_checkpointing_with_teacher(self):
         updates = dict(
             student_encoder_layers=2,
@@ -171,6 +181,7 @@ def test_distill_checkpointing_with_teacher(self):
         convert_pl_to_hf(ckpts[0], transformer_ckpts[0].parent, out_path_new)
         assert os.path.exists(os.path.join(out_path_new, "pytorch_model.bin"))
 
+    @require_torch_non_multigpu_but_fix_me
     def test_loss_fn(self):
         model = AutoModelForSeq2SeqLM.from_pretrained(BART_TINY, return_dict=True)
         input_ids, mask = model.dummy_inputs["input_ids"], model.dummy_inputs["attention_mask"]
@@ -191,6 +202,7 @@ def test_loss_fn(self):
             # TODO: understand why this breaks
             self.assertEqual(nll_loss, model_computed_loss)
 
+    @require_torch_non_multigpu_but_fix_me
     def test_distill_mbart(self):
         updates = dict(
             student_encoder_layers=2,
@@ -215,6 +227,7 @@ def test_distill_mbart(self):
         assert len(all_files) > 2
         self.assertEqual(len(transformer_ckpts), 2)
 
+    @require_torch_non_multigpu_but_fix_me
     def test_distill_t5(self):
         updates = dict(
             student_encoder_layers=1,
@@ -296,18 +309,21 @@ def run_eval_tester(self, model):
 
     # test one model to quickly (no-@slow) catch simple problems and do an
     # extensive testing of functionality with multiple models as @slow separately
+    @require_torch_non_multigpu_but_fix_me
     def test_run_eval(self):
         self.run_eval_tester(T5_TINY)
 
     # any extra models should go into the list here - can be slow
     @parameterized.expand([BART_TINY, MBART_TINY])
     @slow
+    @require_torch_non_multigpu_but_fix_me
     def test_run_eval_slow(self, model):
         self.run_eval_tester(model)
 
     # testing with 2 models to validate: 1. translation (t5) 2. summarization (mbart)
     @parameterized.expand([T5_TINY, MBART_TINY])
     @slow
+    @require_torch_non_multigpu_but_fix_me
     def test_run_eval_search(self, model):
         input_file_name = Path(self.get_auto_remove_tmp_dir()) / "utest_input.source"
         output_file_name = input_file_name.parent / "utest_output.txt"
@@ -358,6 +374,7 @@ def test_run_eval_search(self, model):
     @parameterized.expand(
         [T5_TINY, BART_TINY, MBART_TINY, MARIAN_TINY, FSMT_TINY],
     )
+    @require_torch_non_multigpu_but_fix_me
     def test_finetune(self, model):
         args_d: dict = CHEAP_ARGS.copy()
         task = "translation" if model in [MBART_TINY, MARIAN_TINY, FSMT_TINY] else "summarization"
@@ -409,6 +426,7 @@ def test_finetune(self, model):
         assert isinstance(example_batch, dict)
         assert len(example_batch) >= 4
 
+    @require_torch_non_multigpu_but_fix_me
     def test_finetune_extra_model_args(self):
         args_d: dict = CHEAP_ARGS.copy()
 
@@ -459,6 +477,7 @@ def test_finetune_extra_model_args(self):
             model = main(args)
         assert str(excinfo.value) == f"model config doesn't have a `{unsupported_param}` attribute"
 
+    @require_torch_non_multigpu_but_fix_me
     def test_finetune_lr_schedulers(self):
         args_d: dict = CHEAP_ARGS.copy()