Fix CI benchmarks by temporarily pinning Docker image version

[albertvillanova]

This PR fixes CI benchmarks, by temporarily pinning Docker image version, instead of "latest" tag.

It also updates deprecated cml-send-comment command and using cml comment create instead.

Fix #5431.

[HuggingFaceDocBuilderDev]

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

[github-actions]

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)	0.008519 / 0.011353 (-0.002834)	0.004451 / 0.011008 (-0.006558)	0.102401 / 0.038508 (0.063893)	0.029779 / 0.023109 (0.006669)	0.302654 / 0.275898 (0.026756)	0.366002 / 0.323480 (0.042522)	0.007044 / 0.007986 (-0.000942)	0.003350 / 0.004328 (-0.000978)	0.078213 / 0.004250 (0.073963)	0.035208 / 0.037052 (-0.001844)	0.312980 / 0.258489 (0.054491)	0.344217 / 0.293841 (0.050376)	0.033089 / 0.128546 (-0.095457)	0.011443 / 0.075646 (-0.064203)	0.353143 / 0.419271 (-0.066128)	0.040851 / 0.043533 (-0.002682)	0.304501 / 0.255139 (0.049362)	0.329118 / 0.283200 (0.045918)	0.087399 / 0.141683 (-0.054284)	1.500200 / 1.452155 (0.048046)	1.536176 / 1.492716 (0.043459)

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)	0.209626 / 0.018006 (0.191619)	0.425551 / 0.000490 (0.425061)	0.001168 / 0.000200 (0.000968)	0.000069 / 0.000054 (0.000014)

Benchmark: benchmark_indices_mapping.json

metric	select	shard	shuffle	sort	train_test_split
new / old (diff)	0.023664 / 0.037411 (-0.013748)	0.096792 / 0.014526 (0.082266)	0.105652 / 0.176557 (-0.070905)	0.140796 / 0.737135 (-0.596340)	0.109319 / 0.296338 (-0.187019)

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)	0.414802 / 0.215209 (0.199593)	4.152619 / 2.077655 (2.074964)	1.814403 / 1.504120 (0.310283)	1.611392 / 1.541195 (0.070198)	1.667350 / 1.468490 (0.198860)	0.691855 / 4.584777 (-3.892922)	3.406584 / 3.745712 (-0.339128)	1.940332 / 5.269862 (-3.329530)	1.279061 / 4.565676 (-3.286615)	0.082938 / 0.424275 (-0.341337)	0.012388 / 0.007607 (0.004781)	0.521738 / 0.226044 (0.295693)	5.233764 / 2.268929 (2.964835)	2.306573 / 55.444624 (-53.138051)	1.954631 / 6.876477 (-4.921845)	2.048315 / 2.142072 (-0.093757)	0.816921 / 4.805227 (-3.988306)	0.150983 / 6.500664 (-6.349681)	0.066628 / 0.075469 (-0.008842)

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)	1.235939 / 1.841788 (-0.605849)	14.047114 / 8.074308 (5.972806)	14.149842 / 10.191392 (3.958450)	0.152836 / 0.680424 (-0.527588)	0.028837 / 0.534201 (-0.505364)	0.396232 / 0.579283 (-0.183051)	0.409950 / 0.434364 (-0.024414)	0.460296 / 0.540337 (-0.080041)	0.556787 / 1.386936 (-0.830149)

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)	0.006582 / 0.011353 (-0.004771)	0.004491 / 0.011008 (-0.006518)	0.100093 / 0.038508 (0.061585)	0.026826 / 0.023109 (0.003717)	0.413971 / 0.275898 (0.138073)	0.445625 / 0.323480 (0.122145)	0.004892 / 0.007986 (-0.003094)	0.003295 / 0.004328 (-0.001034)	0.077879 / 0.004250 (0.073628)	0.039177 / 0.037052 (0.002125)	0.353299 / 0.258489 (0.094810)	0.406566 / 0.293841 (0.112725)	0.031633 / 0.128546 (-0.096913)	0.011517 / 0.075646 (-0.064130)	0.320939 / 0.419271 (-0.098332)	0.041487 / 0.043533 (-0.002046)	0.353735 / 0.255139 (0.098596)	0.434786 / 0.283200 (0.151586)	0.087722 / 0.141683 (-0.053961)	1.515134 / 1.452155 (0.062979)	1.588908 / 1.492716 (0.096191)

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)	0.225312 / 0.018006 (0.207305)	0.398324 / 0.000490 (0.397834)	0.000453 / 0.000200 (0.000253)	0.000064 / 0.000054 (0.000009)

Benchmark: benchmark_indices_mapping.json

metric	select	shard	shuffle	sort	train_test_split
new / old (diff)	0.024645 / 0.037411 (-0.012766)	0.099399 / 0.014526 (0.084873)	0.107006 / 0.176557 (-0.069550)	0.145090 / 0.737135 (-0.592045)	0.110046 / 0.296338 (-0.186292)

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)	0.450573 / 0.215209 (0.235364)	4.498030 / 2.077655 (2.420375)	2.193164 / 1.504120 (0.689044)	1.940103 / 1.541195 (0.398908)	1.957137 / 1.468490 (0.488647)	0.697599 / 4.584777 (-3.887178)	3.465146 / 3.745712 (-0.280566)	1.918209 / 5.269862 (-3.351653)	1.183921 / 4.565676 (-3.381756)	0.082540 / 0.424275 (-0.341735)	0.012495 / 0.007607 (0.004888)	0.549702 / 0.226044 (0.323658)	5.526841 / 2.268929 (3.257912)	2.658611 / 55.444624 (-52.786014)	2.259542 / 6.876477 (-4.616935)	2.310139 / 2.142072 (0.168066)	0.810550 / 4.805227 (-3.994677)	0.152369 / 6.500664 (-6.348295)	0.066295 / 0.075469 (-0.009174)

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)	1.289240 / 1.841788 (-0.552547)	14.032143 / 8.074308 (5.957834)	13.973492 / 10.191392 (3.782100)	0.140082 / 0.680424 (-0.540342)	0.017113 / 0.534201 (-0.517088)	0.386534 / 0.579283 (-0.192749)	0.393723 / 0.434364 (-0.040641)	0.448891 / 0.540337 (-0.091446)	0.533085 / 1.386936 (-0.853851)