shorten the test time

argonne-lcf · Dec 8, 2023 · 45e5730 · 45e5730
1 parent 00a253d
commit 45e5730
Show file tree

Hide file tree

Showing 2 changed files with 17 additions and 13 deletions.
diff --git a/.github/workflows/python-package-conda.yml b/.github/workflows/python-package-conda.yml
@@ -137,20 +137,19 @@ jobs:
     - name: test-tf-loader-tfrecord
       run: |
         source ${VENV}/bin/activate
-        mpirun -np 2 dlio_benchmark workload=resnet50 ++workload.dataset.num_files_train=64 ++workload.workflow.train=False ++workload.workflow.generate_data=True  ++workload.dataset.num_files_train=16 ++workload.dataset.num_samples_per_file=16
-        mpirun -np 2 dlio_benchmark workload=resnet50 ++workload.dataset.num_files_train=64 ++workload.workflow.train=True ++workload.workflow.generate_data=False  ++workload.dataset.num_files_train=16 ++workload.dataset.num_samples_per_file=16
+        mpirun -np 2 dlio_benchmark workload=resnet50 ++workload.dataset.num_files_train=64 ++workload.workflow.train=False ++workload.workflow.generate_data=True  ++workload.dataset.num_files_train=4 ++workload.dataset.num_samples_per_file=16
+        mpirun -np 2 dlio_benchmark workload=resnet50 ++workload.dataset.num_files_train=64 ++workload.workflow.train=True ++workload.workflow.generate_data=False  ++workload.dataset.num_files_train=4 ++workload.dataset.num_samples_per_file=16 ++workload.train.computation_time=0.01 ++workload.train.epochs=1
     - name: test-torch-loader-npz
       run: |
         source ${VENV}/bin/activate
-        mpirun -np 2 dlio_benchmark workload=unet3d ++workload.train.computation_time=0.05 ++workload.evaluation.eval_time=0.01 ++workload.train.epochs=2 ++workload.workflow.train=False ++workload.workflow.generate_data=True ++workload.dataset.num_files_train=16 ++workload.dataset.num_files_eval=16 ++workload.reader.read_threads=2 ++workload.dataset.record_length=4096 ++workload.dataset.record_length_stdev=0
-        mpirun -np 2 dlio_benchmark workload=unet3d ++workload.train.computation_time=0.05 ++workload.evaluation.eval_time=0.01 ++workload.train.epochs=2 ++workload.workflow.train=True ++workload.workflow.generate_data=False ++workload.dataset.num_files_train=16 ++workload.dataset.num_files_eval=16 ++workload.reader.read_threads=2  ++workload.dataset.record_length=4096 ++workload.dataset.record_length_stdev=0
+        mpirun -np 2 dlio_benchmark workload=unet3d ++workload.train.computation_time=0.05 ++workload.evaluation.eval_time=0.01 ++workload.workflow.train=False ++workload.workflow.generate_data=True ++workload.dataset.num_files_train=8 ++workload.dataset.num_files_eval=8 ++workload.reader.read_threads=2 ++workload.dataset.record_length=4096 ++workload.dataset.record_length_stdev=0
+        mpirun -np 2 dlio_benchmark workload=unet3d ++workload.train.computation_time=0.05 ++workload.evaluation.eval_time=0.01 ++workload.train.epochs=1 ++workload.workflow.train=True ++workload.workflow.generate_data=False ++workload.dataset.num_files_train=8 ++workload.dataset.num_files_eval=8 ++workload.reader.read_threads=2  ++workload.dataset.record_length=4096 ++workload.dataset.record_length_stdev=0
     - name: test-tf-loader-npz
       run: |
         source ${VENV}/bin/activate
-        mpirun -np 2 dlio_benchmark workload=unet3d ++workload.framework=tensorflow ++workload.data_reader.data_loader=tensorflow ++workload.train.computation_time=0.05 ++workload.evaluation.eval_time=0.01 ++workload.train.epochs=2 ++workload.workflow.train=False ++workload.workflow.generate_data=True ++workload.dataset.num_files_train=16 ++workload.dataset.num_files_eval=16 ++workload.reader.read_threads=2  ++workload.dataset.record_length=4096 ++workload.dataset.record_length_stdev=0
-        mpirun -np 2 dlio_benchmark workload=unet3d ++workload.framework=tensorflow ++workload.data_reader.data_loader=tensorflow ++workload.train.computation_time=0.05 ++workload.evaluation.eval_time=0.01 ++workload.train.epochs=2 ++workload.workflow.train=True ++workload.workflow.generate_data=False ++workload.dataset.num_files_train=16 ++workload.dataset.num_files_eval=16 ++workload.reader.read_threads=2  ++workload.dataset.record_length=4096 ++workload.dataset.record_length_stdev=0
+        mpirun -np 2 dlio_benchmark workload=unet3d ++workload.framework=tensorflow ++workload.data_reader.data_loader=tensorflow ++workload.train.computation_time=0.05 ++workload.evaluation.eval_time=0.01 ++workload.train.epochs=1 ++workload.workflow.train=False ++workload.workflow.generate_data=True ++workload.dataset.num_files_train=8 ++workload.dataset.num_files_eval=8 ++workload.reader.read_threads=2  ++workload.dataset.record_length=4096 ++workload.dataset.record_length_stdev=0
+        mpirun -np 2 dlio_benchmark workload=unet3d ++workload.framework=tensorflow ++workload.data_reader.data_loader=tensorflow ++workload.train.computation_time=0.05 ++workload.evaluation.eval_time=0.01 ++workload.train.epochs=1 ++workload.workflow.train=True ++workload.workflow.generate_data=False ++workload.dataset.num_files_train=8 ++workload.dataset.num_files_eval=8 ++workload.reader.read_threads=2  ++workload.dataset.record_length=4096 ++workload.dataset.record_length_stdev=0
     - name: test_subset
       run: |
         source ${VENV}/bin/activate
-        mpirun -np 2 dlio_benchmark ++workload.workflow.generate_data=True ++workload.workflow.train=False
-        mpirun -np 2 dlio_benchmark ++workload.workflow.generate_data=False ++workload.workflow.train=True ++workload.dataset.num_files_train=8
+        mpirun -np 2 pytest -k test_subset -v 
diff --git a/tests/dlio_benchmark_test.py b/tests/dlio_benchmark_test.py
@@ -88,7 +88,9 @@ def test_gen_data(fmt, framework) -> None:
                                                        f'++workload.reader.data_loader={framework}',
                                                        '++workload.workflow.train=False',
                                                        '++workload.workflow.generate_data=True',
-                                                       f"++workload.dataset.format={fmt}"])
+                                                       f"++workload.dataset.format={fmt}", 
+                                                       "++workload.dataset.num_files_train=8", 
+                                                       "++workload.dataset.num_files_eval=8"])
         benchmark = run_benchmark(cfg, verify=False)
         if benchmark.args.num_subfolders_train <= 1:
             train = pathlib.Path(f"{cfg.workload.dataset.data_folder}/train")
@@ -119,7 +121,9 @@ def test_subset() -> None:
                     '++workload.workflow.generate_data=True'])
         benchmark=run_benchmark(cfg, verify=False)
         cfg = compose(config_name='config', overrides=['++workload.workflow.train=True', \
-                        '++workload.workflow.generate_data=False', '++workload.dataset.num_files_train=8'])
+                        '++workload.workflow.generate_data=False', \
+                            '++workload.dataset.num_files_train=8', \
+                            '++workload.train.computation_time=0.01'])
         benchmark=run_benchmark(cfg, verify=True)
     clean()
 
@@ -142,7 +146,8 @@ def test_storage_root_gen_data(fmt, framework) -> None:
                                                        '++workload.workflow.train=False',
                                                        '++workload.workflow.generate_data=True',
                                                        f"++workload.storage.storage_root={storage_root}",
-                                                       f"++workload.dataset.format={fmt}"])
+                                                       f"++workload.dataset.format={fmt}", 
+                                                       "++workload.dataset.num_files_train=16"])
         benchmark = run_benchmark(cfg, verify=False)
         if benchmark.args.num_subfolders_train <= 1:
             assert (
@@ -302,8 +307,8 @@ def test_multi_threads(framework, nt) -> None:
                                                        'workload.train.computation_time=0.01',
                                                        'workload.evaluation.eval_time=0.005',
                                                        '++workload.train.epochs=1',
-                                                       '++workload.dataset.num_files_train=16',
-                                                       '++workload.dataset.num_files_eval=16'])
+                                                       '++workload.dataset.num_files_train=8',
+                                                       '++workload.dataset.num_files_eval=8'])
         benchmark = run_benchmark(cfg)
     clean()