[AvroTensorDataset] Add more py test to cover various scenarios#1795
Open
lijuanzhang78 wants to merge 6 commits intotensorflow:masterfrom
Open
[AvroTensorDataset] Add more py test to cover various scenarios#1795lijuanzhang78 wants to merge 6 commits intotensorflow:masterfrom
lijuanzhang78 wants to merge 6 commits intotensorflow:masterfrom
Conversation
fix linter
Contributor
Author
|
@kvignesh1420 Hi! Could you help review this PR adding more py tests? Thanks! |
Member
|
Thanks @lijuanzhang78. Will take a look soon! |
Comment on lines
+27
to
+54
| BATCH_SIZES = [8, 16, 32, 64, 128, 256, 512, 1024] | ||
| PARALLELISM = [1, 2, 3, 4, 5, 6, tf.data.AUTOTUNE] | ||
| PARAMS = [ | ||
| (batch_size, 1024, "deflate", parallelism) | ||
| for batch_size in BATCH_SIZES | ||
| for parallelism in PARALLELISM | ||
| ] | ||
|
|
||
|
|
||
| @pytest.mark.benchmark( | ||
| group="autotuning", | ||
| ) | ||
| @pytest.mark.parametrize( | ||
| ["batch_size", "shuffle_buffer_size", "codec", "parallelism"], PARAMS | ||
| ) | ||
| def test_autotuning(batch_size, shuffle_buffer_size, codec, parallelism, benchmark): | ||
| data_source = DataSource( | ||
| scenario=MIXED_TYPES_SCENARIO, num_records=LARGE_NUM_RECORDS | ||
| ) | ||
| run_atds_benchmark_from_data_source( | ||
| data_source, | ||
| batch_size, | ||
| benchmark, | ||
| parallelism=parallelism, | ||
| codec=codec, | ||
| shuffle_buffer_size=shuffle_buffer_size, | ||
| rounds=10, | ||
| ) |
Member
There was a problem hiding this comment.
The file name says that you are testing the autotune functionality. However, I see similar tests in tests/test_atds_avro/benchmark/test_atds_parallelism_benchmark.py as well. Maybe the tests can be combined ?
Comment on lines
+17
to
+37
| import pytest | ||
|
|
||
| from tests.test_atds_avro.utils.data_source import DataSource | ||
| from tests.test_atds_avro.utils.data_source_registry import SMALL_NUM_RECORDS | ||
| from tests.test_atds_avro.utils.atds_benchmark_utils import ( | ||
| run_atds_benchmark_from_data_source, | ||
| ) | ||
| from tests.test_atds_avro.utils.benchmark_utils import MIXED_TYPES_SCENARIO | ||
|
|
||
|
|
||
| @pytest.mark.benchmark( | ||
| group="codec", | ||
| ) | ||
| @pytest.mark.parametrize( | ||
| ["batch_size", "codec"], [(128, "null"), (128, "deflate"), (128, "snappy")] | ||
| ) | ||
| def test_codec(batch_size, codec, benchmark): | ||
| data_source = DataSource( | ||
| scenario=MIXED_TYPES_SCENARIO, num_records=SMALL_NUM_RECORDS | ||
| ) | ||
| run_atds_benchmark_from_data_source(data_source, batch_size, benchmark, codec=codec) |
Member
There was a problem hiding this comment.
same issue as before, I see similar tests in tests/test_atds_avro/benchmark/test_atds_parallelism_benchmark.py as well. Maybe the tests can be combined?
Comment on lines
+30
to
+48
| @pytest.mark.skipif( | ||
| os.getenv("ATDS_MEM_LEAK_CHECK") != "1", | ||
| reason="This benchmark test is only used in memory leak check.", | ||
| ) | ||
| @pytest.mark.benchmark( | ||
| group="all_types_of_tensors", | ||
| ) | ||
| @pytest.mark.parametrize("batch_size", [(16)]) | ||
| def test_all_types_of_tensors_for_memory_leak_check(batch_size, benchmark): | ||
| data_source = get_data_source_from_registry(ALL_TYPES_DATA_SOURCE_NAME) | ||
| shuffle_buffer_size = batch_size * 8 | ||
| run_atds_benchmark_from_data_source( | ||
| data_source, | ||
| batch_size, | ||
| benchmark, | ||
| codec="deflate", | ||
| shuffle_buffer_size=shuffle_buffer_size, | ||
| rounds=1, | ||
| ) |
Member
There was a problem hiding this comment.
I might be missing something but how exactly is the leak detected (if there is one)?
Member
|
@lijuanzhang78 gentle ping as the review comments might have been missed 🙂 . |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This pr add more tests for shuffle, autotune, parallelism for various data types.