Remove metrics

[albertvillanova]

Remove all metrics, as part of the 3.0 release.

Note they are deprecated since 2.5.0 version.

[HuggingFaceDocBuilderDev]

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

[lhoestq]

Awesome ! You can also remove all the dependencies from https://github.com/huggingface/datasets/blob/main/additional-tests-requirements.txt (except torchdata)

[github-actions]

Show benchmarks

PyArrow==8.0.0

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)	0.005566 / 0.011353 (-0.005787)	0.003977 / 0.011008 (-0.007031)	0.063250 / 0.038508 (0.024742)	0.030907 / 0.023109 (0.007798)	0.244989 / 0.275898 (-0.030909)	0.272139 / 0.323480 (-0.051341)	0.004332 / 0.007986 (-0.003653)	0.002960 / 0.004328 (-0.001368)	0.050147 / 0.004250 (0.045896)	0.044740 / 0.037052 (0.007688)	0.256947 / 0.258489 (-0.001542)	0.290372 / 0.293841 (-0.003469)	0.030444 / 0.128546 (-0.098102)	0.012675 / 0.075646 (-0.062971)	0.203852 / 0.419271 (-0.215420)	0.036977 / 0.043533 (-0.006556)	0.244401 / 0.255139 (-0.010738)	0.270020 / 0.283200 (-0.013179)	0.018177 / 0.141683 (-0.123506)	1.122189 / 1.452155 (-0.329966)	1.176688 / 1.492716 (-0.316028)

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)	0.100721 / 0.018006 (0.082715)	0.311824 / 0.000490 (0.311335)	0.000222 / 0.000200 (0.000022)	0.000043 / 0.000054 (-0.000012)

Benchmark: benchmark_indices_mapping.json

metric	select	shard	shuffle	sort	train_test_split
new / old (diff)	0.020039 / 0.037411 (-0.017373)	0.062084 / 0.014526 (0.047558)	0.074317 / 0.176557 (-0.102240)	0.123935 / 0.737135 (-0.613200)	0.076186 / 0.296338 (-0.220153)

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)	0.284827 / 0.215209 (0.069618)	2.782727 / 2.077655 (0.705072)	1.417624 / 1.504120 (-0.086496)	1.294476 / 1.541195 (-0.246718)	1.332658 / 1.468490 (-0.135832)	0.724820 / 4.584777 (-3.859957)	2.384546 / 3.745712 (-1.361166)	2.866759 / 5.269862 (-2.403103)	1.930756 / 4.565676 (-2.634921)	0.083090 / 0.424275 (-0.341185)	0.005566 / 0.007607 (-0.002041)	0.340117 / 0.226044 (0.114072)	3.342417 / 2.268929 (1.073488)	1.807842 / 55.444624 (-53.636782)	1.511647 / 6.876477 (-5.364830)	1.653893 / 2.142072 (-0.488179)	0.803983 / 4.805227 (-4.001244)	0.136205 / 6.500664 (-6.364459)	0.042815 / 0.075469 (-0.032654)

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)	0.962346 / 1.841788 (-0.879442)	11.792239 / 8.074308 (3.717931)	9.236256 / 10.191392 (-0.955136)	0.143200 / 0.680424 (-0.537224)	0.015050 / 0.534201 (-0.519151)	0.304623 / 0.579283 (-0.274660)	0.266417 / 0.434364 (-0.167947)	0.341213 / 0.540337 (-0.199124)	0.454258 / 1.386936 (-0.932678)

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)	0.005917 / 0.011353 (-0.005436)	0.004005 / 0.011008 (-0.007003)	0.049781 / 0.038508 (0.011273)	0.033310 / 0.023109 (0.010200)	0.271881 / 0.275898 (-0.004017)	0.296855 / 0.323480 (-0.026625)	0.004479 / 0.007986 (-0.003507)	0.002818 / 0.004328 (-0.001510)	0.048213 / 0.004250 (0.043962)	0.043480 / 0.037052 (0.006428)	0.285963 / 0.258489 (0.027473)	0.317304 / 0.293841 (0.023463)	0.031619 / 0.128546 (-0.096928)	0.012312 / 0.075646 (-0.063335)	0.059904 / 0.419271 (-0.359368)	0.033152 / 0.043533 (-0.010381)	0.274198 / 0.255139 (0.019059)	0.290469 / 0.283200 (0.007269)	0.019424 / 0.141683 (-0.122258)	1.133669 / 1.452155 (-0.318485)	1.194427 / 1.492716 (-0.298290)

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)	0.101561 / 0.018006 (0.083555)	0.312617 / 0.000490 (0.312127)	0.000216 / 0.000200 (0.000016)	0.000045 / 0.000054 (-0.000009)

Benchmark: benchmark_indices_mapping.json

metric	select	shard	shuffle	sort	train_test_split
new / old (diff)	0.023705 / 0.037411 (-0.013706)	0.076781 / 0.014526 (0.062255)	0.089922 / 0.176557 (-0.086634)	0.129182 / 0.737135 (-0.607953)	0.092022 / 0.296338 (-0.204317)

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)	0.300977 / 0.215209 (0.085768)	2.909088 / 2.077655 (0.831433)	1.592821 / 1.504120 (0.088701)	1.466627 / 1.541195 (-0.074568)	1.497558 / 1.468490 (0.029068)	0.720986 / 4.584777 (-3.863791)	0.958039 / 3.745712 (-2.787673)	3.023413 / 5.269862 (-2.246448)	1.933245 / 4.565676 (-2.632432)	0.080500 / 0.424275 (-0.343775)	0.005243 / 0.007607 (-0.002364)	0.361259 / 0.226044 (0.135215)	3.447317 / 2.268929 (1.178389)	1.938234 / 55.444624 (-53.506390)	1.671563 / 6.876477 (-5.204913)	1.674647 / 2.142072 (-0.467425)	0.790606 / 4.805227 (-4.014621)	0.133312 / 6.500664 (-6.367352)	0.041241 / 0.075469 (-0.034228)

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)	0.996167 / 1.841788 (-0.845621)	12.460877 / 8.074308 (4.386569)	10.608415 / 10.191392 (0.417023)	0.134076 / 0.680424 (-0.546348)	0.016166 / 0.534201 (-0.518035)	0.301218 / 0.579283 (-0.278065)	0.128979 / 0.434364 (-0.305385)	0.336453 / 0.540337 (-0.203884)	0.435561 / 1.386936 (-0.951375)