Fix use_dataset.mdx

[angel-luis]

The current example isn't working because it can't find labels inside the Dataset object. So I've added an extra step to the process. Tested and working in Colab.

[HuggingFaceDocBuilderDev]

The documentation is not available anymore as the PR was closed or merged.

[mariosasko]

Good catch!

[mariosasko]

We can avoid introducing a new step by replacing labels mentions with label in the existing ones.

[mariosasko]

I made the changes myself. Looks good now. Thanks!

[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)	0.007718 / 0.011353 (-0.003635)	0.004730 / 0.011008 (-0.006278)	0.097262 / 0.038508 (0.058754)	0.077880 / 0.023109 (0.054771)	0.363855 / 0.275898 (0.087957)	0.394470 / 0.323480 (0.070990)	0.006416 / 0.007986 (-0.001570)	0.003596 / 0.004328 (-0.000732)	0.076494 / 0.004250 (0.072243)	0.062656 / 0.037052 (0.025603)	0.366160 / 0.258489 (0.107671)	0.421383 / 0.293841 (0.127542)	0.035756 / 0.128546 (-0.092791)	0.009430 / 0.075646 (-0.066217)	0.327722 / 0.419271 (-0.091550)	0.061252 / 0.043533 (0.017719)	0.352167 / 0.255139 (0.097028)	0.385166 / 0.283200 (0.101966)	0.026656 / 0.141683 (-0.115027)	1.718533 / 1.452155 (0.266378)	1.886646 / 1.492716 (0.393930)

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)	0.254564 / 0.018006 (0.236558)	0.490942 / 0.000490 (0.490452)	0.011656 / 0.000200 (0.011456)	0.000313 / 0.000054 (0.000259)

Benchmark: benchmark_indices_mapping.json

metric	select	shard	shuffle	sort	train_test_split
new / old (diff)	0.028753 / 0.037411 (-0.008659)	0.093076 / 0.014526 (0.078550)	0.096441 / 0.176557 (-0.080116)	0.154848 / 0.737135 (-0.582287)	0.092903 / 0.296338 (-0.203435)

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)	0.395611 / 0.215209 (0.180402)	3.860736 / 2.077655 (1.783082)	1.908808 / 1.504120 (0.404688)	1.708975 / 1.541195 (0.167781)	1.848173 / 1.468490 (0.379683)	0.527022 / 4.584777 (-4.057755)	3.815171 / 3.745712 (0.069459)	3.621132 / 5.269862 (-1.648730)	2.220238 / 4.565676 (-2.345439)	0.063169 / 0.424275 (-0.361106)	0.008906 / 0.007607 (0.001299)	0.510478 / 0.226044 (0.284433)	4.828116 / 2.268929 (2.559187)	2.340801 / 55.444624 (-53.103824)	2.040834 / 6.876477 (-4.835642)	2.092316 / 2.142072 (-0.049757)	0.579194 / 4.805227 (-4.226033)	0.135525 / 6.500664 (-6.365139)	0.062720 / 0.075469 (-0.012749)

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)	1.393091 / 1.841788 (-0.448697)	19.751526 / 8.074308 (11.677218)	14.161795 / 10.191392 (3.970403)	0.163340 / 0.680424 (-0.517084)	0.021504 / 0.534201 (-0.512697)	0.393183 / 0.579283 (-0.186100)	0.448407 / 0.434364 (0.014043)	0.504169 / 0.540337 (-0.036169)	0.663698 / 1.386936 (-0.723238)

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)	0.007390 / 0.011353 (-0.003962)	0.004381 / 0.011008 (-0.006628)	0.074501 / 0.038508 (0.035993)	0.078242 / 0.023109 (0.055133)	0.481108 / 0.275898 (0.205210)	0.512111 / 0.323480 (0.188631)	0.006280 / 0.007986 (-0.001705)	0.003820 / 0.004328 (-0.000509)	0.071602 / 0.004250 (0.067351)	0.068359 / 0.037052 (0.031307)	0.478484 / 0.258489 (0.219995)	0.519543 / 0.293841 (0.225702)	0.036211 / 0.128546 (-0.092335)	0.009433 / 0.075646 (-0.066213)	0.086140 / 0.419271 (-0.333132)	0.054177 / 0.043533 (0.010644)	0.466726 / 0.255139 (0.211587)	0.514085 / 0.283200 (0.230885)	0.026729 / 0.141683 (-0.114954)	1.743770 / 1.452155 (0.291615)	1.833469 / 1.492716 (0.340753)

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)	0.251339 / 0.018006 (0.233333)	0.472294 / 0.000490 (0.471804)	0.013381 / 0.000200 (0.013181)	0.000117 / 0.000054 (0.000062)

Benchmark: benchmark_indices_mapping.json

metric	select	shard	shuffle	sort	train_test_split
new / old (diff)	0.037845 / 0.037411 (0.000433)	0.105977 / 0.014526 (0.091451)	0.124446 / 0.176557 (-0.052111)	0.180432 / 0.737135 (-0.556703)	0.120844 / 0.296338 (-0.175495)

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)	0.470928 / 0.215209 (0.255719)	4.738154 / 2.077655 (2.660499)	2.558618 / 1.504120 (1.054498)	2.359745 / 1.541195 (0.818550)	2.458438 / 1.468490 (0.989948)	0.548580 / 4.584777 (-4.036197)	3.912145 / 3.745712 (0.166433)	3.764174 / 5.269862 (-1.505687)	2.325265 / 4.565676 (-2.240411)	0.078022 / 0.424275 (-0.346254)	0.008279 / 0.007607 (0.000672)	0.571635 / 0.226044 (0.345590)	5.672445 / 2.268929 (3.403517)	2.760577 / 55.444624 (-52.684047)	2.544229 / 6.876477 (-4.332248)	2.537509 / 2.142072 (0.395436)	0.609858 / 4.805227 (-4.195369)	0.131053 / 6.500664 (-6.369611)	0.056433 / 0.075469 (-0.019036)

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)	1.567231 / 1.841788 (-0.274556)	21.415586 / 8.074308 (13.341278)	15.982328 / 10.191392 (5.790936)	0.167648 / 0.680424 (-0.512776)	0.023562 / 0.534201 (-0.510639)	0.477307 / 0.579283 (-0.101976)	0.471929 / 0.434364 (0.037566)	0.549996 / 0.540337 (0.009659)	0.753927 / 1.386936 (-0.633009)