Skip to content

Commit

Permalink
more fixes
Browse files Browse the repository at this point in the history
  • Loading branch information
@lhoestq
lhoestq committed Jul 27, 2022
1 parent 47e1263 commit ecaa2b5
Show file tree
Hide file tree
Showing 2 changed files with 1 addition and 2 deletions.
2 changes: 1 addition & 1 deletion datasets/moroco/README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,7 @@ language_creators:
language:
- ro
language_bcp47:
- ro-md
- ro-MD
license:
- cc-by-4.0
multilinguality:
Expand Down
1 change: 0 additions & 1 deletion datasets/opus_gnome/README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -93,7 +93,6 @@ language:
- mt
- mus
- my
- n/o
- nb
- nds
- ne
Expand Down

1 comment on commit ecaa2b5

@github-actions
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Show benchmarks

PyArrow==6.0.0

Show updated benchmarks!

Benchmark: benchmark_array_xd.json

metric read_batch_formatted_as_numpy after write_array2d read_batch_formatted_as_numpy after write_flattened_sequence read_batch_formatted_as_numpy after write_nested_sequence read_batch_unformated after write_array2d read_batch_unformated after write_flattened_sequence read_batch_unformated after write_nested_sequence read_col_formatted_as_numpy after write_array2d read_col_formatted_as_numpy after write_flattened_sequence read_col_formatted_as_numpy after write_nested_sequence read_col_unformated after write_array2d read_col_unformated after write_flattened_sequence read_col_unformated after write_nested_sequence read_formatted_as_numpy after write_array2d read_formatted_as_numpy after write_flattened_sequence read_formatted_as_numpy after write_nested_sequence read_unformated after write_array2d read_unformated after write_flattened_sequence read_unformated after write_nested_sequence write_array2d write_flattened_sequence write_nested_sequence
new / old (diff) 0.009920 / 0.011353 (-0.001433) 0.004674 / 0.011008 (-0.006334) 0.036960 / 0.038508 (-0.001548) 0.043635 / 0.023109 (0.020526) 0.374408 / 0.275898 (0.098510) 0.408345 / 0.323480 (0.084865) 0.007406 / 0.007986 (-0.000579) 0.005680 / 0.004328 (0.001352) 0.008640 / 0.004250 (0.004389) 0.049083 / 0.037052 (0.012031) 0.385715 / 0.258489 (0.127226) 0.478752 / 0.293841 (0.184911) 0.038458 / 0.128546 (-0.090089) 0.011580 / 0.075646 (-0.064066) 0.321496 / 0.419271 (-0.097776) 0.063593 / 0.043533 (0.020060) 0.374490 / 0.255139 (0.119351) 0.398930 / 0.283200 (0.115731) 0.122599 / 0.141683 (-0.019084) 1.813242 / 1.452155 (0.361087) 1.840265 / 1.492716 (0.347548)

Benchmark: benchmark_getitem_100B.json

metric get_batch_of_1024_random_rows get_batch_of_1024_rows get_first_row get_last_row
new / old (diff) 0.228455 / 0.018006 (0.210449) 0.489961 / 0.000490 (0.489471) 0.006068 / 0.000200 (0.005868) 0.000108 / 0.000054 (0.000053)

Benchmark: benchmark_indices_mapping.json

metric select shard shuffle sort train_test_split
new / old (diff) 0.030381 / 0.037411 (-0.007030) 0.128457 / 0.014526 (0.113932) 0.144176 / 0.176557 (-0.032380) 0.197244 / 0.737135 (-0.539892) 0.149045 / 0.296338 (-0.147293)

Benchmark: benchmark_iterating.json

metric read 5000 read 50000 read_batch 50000 10 read_batch 50000 100 read_batch 50000 1000 read_formatted numpy 5000 read_formatted pandas 5000 read_formatted tensorflow 5000 read_formatted torch 5000 read_formatted_batch numpy 5000 10 read_formatted_batch numpy 5000 1000 shuffled read 5000 shuffled read 50000 shuffled read_batch 50000 10 shuffled read_batch 50000 100 shuffled read_batch 50000 1000 shuffled read_formatted numpy 5000 shuffled read_formatted_batch numpy 5000 10 shuffled read_formatted_batch numpy 5000 1000
new / old (diff) 0.508685 / 0.215209 (0.293476) 5.096716 / 2.077655 (3.019061) 2.343593 / 1.504120 (0.839473) 2.115220 / 1.541195 (0.574025) 2.147380 / 1.468490 (0.678890) 0.528692 / 4.584777 (-4.056085) 4.606646 / 3.745712 (0.860934) 4.362813 / 5.269862 (-0.907049) 2.248848 / 4.565676 (-2.316829) 0.065061 / 0.424275 (-0.359214) 0.013803 / 0.007607 (0.006196) 0.639679 / 0.226044 (0.413634) 6.430100 / 2.268929 (4.161171) 2.878419 / 55.444624 (-52.566205) 2.472140 / 6.876477 (-4.404337) 2.603598 / 2.142072 (0.461525) 0.675141 / 4.805227 (-4.130087) 0.146176 / 6.500664 (-6.354488) 0.075434 / 0.075469 (-0.000035)

Benchmark: benchmark_map_filter.json

metric filter map fast-tokenizer batched map identity map identity batched map no-op batched map no-op batched numpy map no-op batched pandas map no-op batched pytorch map no-op batched tensorflow
new / old (diff) 1.830053 / 1.841788 (-0.011734) 17.000893 / 8.074308 (8.926585) 30.919317 / 10.191392 (20.727925) 1.081012 / 0.680424 (0.400588) 0.692253 / 0.534201 (0.158053) 0.476755 / 0.579283 (-0.102528) 0.522265 / 0.434364 (0.087901) 0.329054 / 0.540337 (-0.211283) 0.335745 / 1.386936 (-1.051191)
PyArrow==latest
Show updated benchmarks!

Benchmark: benchmark_array_xd.json

metric read_batch_formatted_as_numpy after write_array2d read_batch_formatted_as_numpy after write_flattened_sequence read_batch_formatted_as_numpy after write_nested_sequence read_batch_unformated after write_array2d read_batch_unformated after write_flattened_sequence read_batch_unformated after write_nested_sequence read_col_formatted_as_numpy after write_array2d read_col_formatted_as_numpy after write_flattened_sequence read_col_formatted_as_numpy after write_nested_sequence read_col_unformated after write_array2d read_col_unformated after write_flattened_sequence read_col_unformated after write_nested_sequence read_formatted_as_numpy after write_array2d read_formatted_as_numpy after write_flattened_sequence read_formatted_as_numpy after write_nested_sequence read_unformated after write_array2d read_unformated after write_flattened_sequence read_unformated after write_nested_sequence write_array2d write_flattened_sequence write_nested_sequence
new / old (diff) 0.007068 / 0.011353 (-0.004285) 0.004565 / 0.011008 (-0.006444) 0.035405 / 0.038508 (-0.003103) 0.040300 / 0.023109 (0.017191) 0.404313 / 0.275898 (0.128415) 0.470429 / 0.323480 (0.146949) 0.004388 / 0.007986 (-0.003598) 0.003887 / 0.004328 (-0.000441) 0.005980 / 0.004250 (0.001730) 0.051631 / 0.037052 (0.014579) 0.414343 / 0.258489 (0.155854) 0.454389 / 0.293841 (0.160548) 0.035622 / 0.128546 (-0.092924) 0.011694 / 0.075646 (-0.063953) 0.331047 / 0.419271 (-0.088224) 0.065225 / 0.043533 (0.021692) 0.393898 / 0.255139 (0.138759) 0.423431 / 0.283200 (0.140231) 0.121758 / 0.141683 (-0.019924) 1.838998 / 1.452155 (0.386843) 1.854084 / 1.492716 (0.361368)

Benchmark: benchmark_getitem_100B.json

metric get_batch_of_1024_random_rows get_batch_of_1024_rows get_first_row get_last_row
new / old (diff) 0.250764 / 0.018006 (0.232758) 0.491610 / 0.000490 (0.491121) 0.001190 / 0.000200 (0.000990) 0.000097 / 0.000054 (0.000043)

Benchmark: benchmark_indices_mapping.json

metric select shard shuffle sort train_test_split
new / old (diff) 0.031029 / 0.037411 (-0.006383) 0.129263 / 0.014526 (0.114738) 0.144491 / 0.176557 (-0.032066) 0.206639 / 0.737135 (-0.530496) 0.149618 / 0.296338 (-0.146720)

Benchmark: benchmark_iterating.json

metric read 5000 read 50000 read_batch 50000 10 read_batch 50000 100 read_batch 50000 1000 read_formatted numpy 5000 read_formatted pandas 5000 read_formatted tensorflow 5000 read_formatted torch 5000 read_formatted_batch numpy 5000 10 read_formatted_batch numpy 5000 1000 shuffled read 5000 shuffled read 50000 shuffled read_batch 50000 10 shuffled read_batch 50000 100 shuffled read_batch 50000 1000 shuffled read_formatted numpy 5000 shuffled read_formatted_batch numpy 5000 10 shuffled read_formatted_batch numpy 5000 1000
new / old (diff) 0.487721 / 0.215209 (0.272512) 4.869615 / 2.077655 (2.791960) 2.207600 / 1.504120 (0.703481) 1.981923 / 1.541195 (0.440728) 2.046433 / 1.468490 (0.577942) 0.522395 / 4.584777 (-4.062382) 4.662158 / 3.745712 (0.916446) 4.522264 / 5.269862 (-0.747598) 2.247950 / 4.565676 (-2.317726) 0.063537 / 0.424275 (-0.360738) 0.013679 / 0.007607 (0.006072) 0.614269 / 0.226044 (0.388225) 6.139593 / 2.268929 (3.870664) 2.841116 / 55.444624 (-52.603508) 2.458791 / 6.876477 (-4.417686) 2.537541 / 2.142072 (0.395468) 0.659117 / 4.805227 (-4.146110) 0.150281 / 6.500664 (-6.350383) 0.077726 / 0.075469 (0.002257)

Benchmark: benchmark_map_filter.json

metric filter map fast-tokenizer batched map identity map identity batched map no-op batched map no-op batched numpy map no-op batched pandas map no-op batched pytorch map no-op batched tensorflow
new / old (diff) 1.803561 / 1.841788 (-0.038227) 16.731115 / 8.074308 (8.656807) 30.914286 / 10.191392 (20.722894) 1.052708 / 0.680424 (0.372284) 0.665522 / 0.534201 (0.131321) 0.482919 / 0.579283 (-0.096364) 0.522671 / 0.434364 (0.088307) 0.352109 / 0.540337 (-0.188229) 0.352853 / 1.386936 (-1.034083)

CML watermark

Please sign in to comment.