Create function to convert to parquet

[albertvillanova]

Analogously with delete_from_hub, this PR:

• creates the Python function convert_to_parquet
• makes the corresponding CLI command use that function.

This way, the functionality can be used both from a terminal and from a Python console.

This PR also implements a test for convert_to_parquet function.

[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.

[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)	0.005519 / 0.011353 (-0.005834)	0.003877 / 0.011008 (-0.007131)	0.063989 / 0.038508 (0.025480)	0.032348 / 0.023109 (0.009239)	0.238288 / 0.275898 (-0.037611)	0.265337 / 0.323480 (-0.058143)	0.004363 / 0.007986 (-0.003623)	0.002755 / 0.004328 (-0.001574)	0.049836 / 0.004250 (0.045585)	0.048456 / 0.037052 (0.011403)	0.246526 / 0.258489 (-0.011963)	0.280753 / 0.293841 (-0.013088)	0.027721 / 0.128546 (-0.100825)	0.011031 / 0.075646 (-0.064615)	0.204168 / 0.419271 (-0.215104)	0.036203 / 0.043533 (-0.007330)	0.238282 / 0.255139 (-0.016857)	0.259608 / 0.283200 (-0.023591)	0.017781 / 0.141683 (-0.123902)	1.147821 / 1.452155 (-0.304334)	1.194855 / 1.492716 (-0.297861)

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)	0.102837 / 0.018006 (0.084831)	0.312300 / 0.000490 (0.311811)	0.000224 / 0.000200 (0.000024)	0.000047 / 0.000054 (-0.000008)

Benchmark: benchmark_indices_mapping.json

metric	select	shard	shuffle	sort	train_test_split
new / old (diff)	0.019410 / 0.037411 (-0.018001)	0.065114 / 0.014526 (0.050588)	0.076828 / 0.176557 (-0.099728)	0.121741 / 0.737135 (-0.615394)	0.079864 / 0.296338 (-0.216474)

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)	0.287773 / 0.215209 (0.072564)	2.848936 / 2.077655 (0.771281)	1.543819 / 1.504120 (0.039700)	1.412708 / 1.541195 (-0.128487)	1.454685 / 1.468490 (-0.013805)	0.580155 / 4.584777 (-4.004622)	2.372783 / 3.745712 (-1.372929)	2.910514 / 5.269862 (-2.359347)	1.813542 / 4.565676 (-2.752134)	0.064569 / 0.424275 (-0.359706)	0.005434 / 0.007607 (-0.002173)	0.339309 / 0.226044 (0.113265)	3.329972 / 2.268929 (1.061043)	1.827597 / 55.444624 (-53.617028)	1.592324 / 6.876477 (-5.284152)	1.619743 / 2.142072 (-0.522329)	0.659358 / 4.805227 (-4.145869)	0.119887 / 6.500664 (-6.380777)	0.043649 / 0.075469 (-0.031821)

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)	0.984563 / 1.841788 (-0.857225)	12.395302 / 8.074308 (4.320994)	9.904944 / 10.191392 (-0.286448)	0.136141 / 0.680424 (-0.544282)	0.014779 / 0.534201 (-0.519422)	0.286146 / 0.579283 (-0.293137)	0.265392 / 0.434364 (-0.168972)	0.329484 / 0.540337 (-0.210854)	0.425530 / 1.386936 (-0.961406)

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)	0.005920 / 0.011353 (-0.005433)	0.004068 / 0.011008 (-0.006940)	0.052281 / 0.038508 (0.013773)	0.034907 / 0.023109 (0.011798)	0.269551 / 0.275898 (-0.006347)	0.292390 / 0.323480 (-0.031090)	0.004340 / 0.007986 (-0.003646)	0.002864 / 0.004328 (-0.001464)	0.051466 / 0.004250 (0.047216)	0.046410 / 0.037052 (0.009358)	0.280103 / 0.258489 (0.021614)	0.310616 / 0.293841 (0.016775)	0.031044 / 0.128546 (-0.097502)	0.011004 / 0.075646 (-0.064643)	0.059955 / 0.419271 (-0.359316)	0.034156 / 0.043533 (-0.009377)	0.268113 / 0.255139 (0.012974)	0.283569 / 0.283200 (0.000369)	0.019758 / 0.141683 (-0.121925)	1.155583 / 1.452155 (-0.296572)	1.225611 / 1.492716 (-0.267106)

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)	0.104302 / 0.018006 (0.086295)	0.307324 / 0.000490 (0.306834)	0.000219 / 0.000200 (0.000019)	0.000045 / 0.000054 (-0.000009)

Benchmark: benchmark_indices_mapping.json

metric	select	shard	shuffle	sort	train_test_split
new / old (diff)	0.023672 / 0.037411 (-0.013739)	0.081110 / 0.014526 (0.066584)	0.091783 / 0.176557 (-0.084773)	0.131738 / 0.737135 (-0.605397)	0.092391 / 0.296338 (-0.203948)

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)	0.289341 / 0.215209 (0.074132)	2.849894 / 2.077655 (0.772239)	1.539679 / 1.504120 (0.035559)	1.417975 / 1.541195 (-0.123220)	1.473631 / 1.468490 (0.005141)	0.583013 / 4.584777 (-4.001764)	0.960106 / 3.745712 (-2.785606)	2.962785 / 5.269862 (-2.307077)	1.827539 / 4.565676 (-2.738138)	0.063875 / 0.424275 (-0.360400)	0.005251 / 0.007607 (-0.002356)	0.347127 / 0.226044 (0.121082)	3.417364 / 2.268929 (1.148435)	1.965901 / 55.444624 (-53.478723)	1.632337 / 6.876477 (-5.244140)	1.683100 / 2.142072 (-0.458972)	0.664951 / 4.805227 (-4.140277)	0.119046 / 6.500664 (-6.381618)	0.042828 / 0.075469 (-0.032641)

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)	0.999569 / 1.841788 (-0.842218)	13.366482 / 8.074308 (5.292174)	10.635396 / 10.191392 (0.444004)	0.133840 / 0.680424 (-0.546584)	0.016232 / 0.534201 (-0.517969)	0.292764 / 0.579283 (-0.286519)	0.128558 / 0.434364 (-0.305806)	0.405596 / 0.540337 (-0.134741)	0.429633 / 1.386936 (-0.957303)