Move info_utils errors to exceptions module

[albertvillanova]

Move info_utils errors to exceptions module.

Additionally rename some of them, deprecate the former ones, and make the deprecation backward compatible (by making the new errors inherit from the former ones).

[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]

is it also marked as deprecated since it inherits from SplitsVerificationException ?

[albertvillanova]

I have renamed SplitsVerificationException to SplitsVerificationError and deprecated the former. The new SplitsVerificationError is not deprecated.

It inherits from SplitsVerificationException for backward compatibility: if a user is catching SplitsVerificationException, now it will also catch SplitsVerificationError.

[lhoestq]

yes but since you subclass from the deprecated class, the subclass is also marked as deprecated

In [1]: from datasets.exceptions import *

In [2]: ChecksumVerificationError()
<ipython-input-2-0953a165cf8e>:1: FutureWarning: ChecksumVerificationException is deprecated and will be removed in the next major version of datasets. Use 'ChecksumVerificationError' instead.

[albertvillanova]

Oups! Thanks for letting me know... 😅

[albertvillanova]

@lhoestq I fixed it.

[lhoestq]

nice !

[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)	0.005232 / 0.011353 (-0.006121)	0.003744 / 0.011008 (-0.007264)	0.064089 / 0.038508 (0.025581)	0.032409 / 0.023109 (0.009300)	0.255886 / 0.275898 (-0.020013)	0.276033 / 0.323480 (-0.047447)	0.004165 / 0.007986 (-0.003821)	0.002741 / 0.004328 (-0.001588)	0.052145 / 0.004250 (0.047894)	0.043863 / 0.037052 (0.006811)	0.258844 / 0.258489 (0.000355)	0.290108 / 0.293841 (-0.003733)	0.027390 / 0.128546 (-0.101156)	0.010543 / 0.075646 (-0.065103)	0.206936 / 0.419271 (-0.212335)	0.036778 / 0.043533 (-0.006755)	0.254331 / 0.255139 (-0.000808)	0.279037 / 0.283200 (-0.004163)	0.018564 / 0.141683 (-0.123119)	1.112765 / 1.452155 (-0.339390)	1.160099 / 1.492716 (-0.332617)

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)	0.092148 / 0.018006 (0.074142)	0.297156 / 0.000490 (0.296667)	0.000211 / 0.000200 (0.000011)	0.000043 / 0.000054 (-0.000011)

Benchmark: benchmark_indices_mapping.json

metric	select	shard	shuffle	sort	train_test_split
new / old (diff)	0.018797 / 0.037411 (-0.018615)	0.062992 / 0.014526 (0.048466)	0.076361 / 0.176557 (-0.100195)	0.121168 / 0.737135 (-0.615968)	0.075845 / 0.296338 (-0.220494)

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)	0.293842 / 0.215209 (0.078633)	2.880720 / 2.077655 (0.803065)	1.477779 / 1.504120 (-0.026341)	1.345136 / 1.541195 (-0.196059)	1.352153 / 1.468490 (-0.116337)	0.574722 / 4.584777 (-4.010055)	2.373925 / 3.745712 (-1.371787)	2.750704 / 5.269862 (-2.519157)	1.725979 / 4.565676 (-2.839697)	0.063006 / 0.424275 (-0.361269)	0.005019 / 0.007607 (-0.002588)	0.341228 / 0.226044 (0.115184)	3.352576 / 2.268929 (1.083647)	1.821363 / 55.444624 (-53.623261)	1.529441 / 6.876477 (-5.347036)	1.543401 / 2.142072 (-0.598671)	0.634282 / 4.805227 (-4.170945)	0.115565 / 6.500664 (-6.385099)	0.042514 / 0.075469 (-0.032956)

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)	0.987532 / 1.841788 (-0.854255)	11.483853 / 8.074308 (3.409545)	9.565657 / 10.191392 (-0.625735)	0.141247 / 0.680424 (-0.539176)	0.015026 / 0.534201 (-0.519175)	0.299905 / 0.579283 (-0.279378)	0.267667 / 0.434364 (-0.166697)	0.320661 / 0.540337 (-0.219676)	0.427368 / 1.386936 (-0.959568)

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)	0.005448 / 0.011353 (-0.005905)	0.003726 / 0.011008 (-0.007283)	0.049776 / 0.038508 (0.011268)	0.032733 / 0.023109 (0.009624)	0.261387 / 0.275898 (-0.014511)	0.280087 / 0.323480 (-0.043393)	0.004351 / 0.007986 (-0.003634)	0.002842 / 0.004328 (-0.001487)	0.049440 / 0.004250 (0.045190)	0.039585 / 0.037052 (0.002533)	0.266331 / 0.258489 (0.007842)	0.299643 / 0.293841 (0.005802)	0.029649 / 0.128546 (-0.098897)	0.010381 / 0.075646 (-0.065265)	0.058596 / 0.419271 (-0.360676)	0.033271 / 0.043533 (-0.010262)	0.251070 / 0.255139 (-0.004069)	0.272850 / 0.283200 (-0.010349)	0.016728 / 0.141683 (-0.124955)	1.146952 / 1.452155 (-0.305202)	1.182602 / 1.492716 (-0.310114)

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)	0.091673 / 0.018006 (0.073667)	0.297228 / 0.000490 (0.296738)	0.000197 / 0.000200 (-0.000003)	0.000051 / 0.000054 (-0.000003)

Benchmark: benchmark_indices_mapping.json

metric	select	shard	shuffle	sort	train_test_split
new / old (diff)	0.023174 / 0.037411 (-0.014237)	0.078866 / 0.014526 (0.064341)	0.088436 / 0.176557 (-0.088121)	0.129650 / 0.737135 (-0.607485)	0.091100 / 0.296338 (-0.205238)

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)	0.293882 / 0.215209 (0.078673)	2.882667 / 2.077655 (0.805012)	1.562949 / 1.504120 (0.058829)	1.435104 / 1.541195 (-0.106090)	1.450815 / 1.468490 (-0.017675)	0.584090 / 4.584777 (-4.000687)	0.984176 / 3.745712 (-2.761536)	2.668740 / 5.269862 (-2.601121)	1.766993 / 4.565676 (-2.798683)	0.064710 / 0.424275 (-0.359565)	0.005329 / 0.007607 (-0.002278)	0.346008 / 0.226044 (0.119964)	3.414576 / 2.268929 (1.145647)	1.911388 / 55.444624 (-53.533236)	1.660357 / 6.876477 (-5.216120)	1.818628 / 2.142072 (-0.323444)	0.659585 / 4.805227 (-4.145643)	0.116980 / 6.500664 (-6.383684)	0.041364 / 0.075469 (-0.034105)

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)	1.005659 / 1.841788 (-0.836129)	12.023761 / 8.074308 (3.949453)	10.351086 / 10.191392 (0.159694)	0.143261 / 0.680424 (-0.537162)	0.016143 / 0.534201 (-0.518058)	0.287793 / 0.579283 (-0.291490)	0.123698 / 0.434364 (-0.310666)	0.325241 / 0.540337 (-0.215097)	0.418772 / 1.386936 (-0.968164)