eval-harness

dataset

cnn_dailymail_dm200

200 items · 5 arms. Headline: each arm's own primary metric, mean over its items; arms with different primary metrics are not ranked against each other.

Inspect the 5 arms on coverage, scored against gold/cnn_dailymail_dm200Open its sheet

goals: Short news summaries (CNN)

01arms

scored against silverscored by a judgewrote the reference it is scored againstscored against a label set, not a reference setunmarked rows are scored against gold

5 shown
armmetricmeanitemscost per itemspeed per itemrunsscored against
cnn_bart_l_dm200_v1coverage0.443200$0.000000 · 934 tokp50 10,251 · p95 16,948 ms1gold/cnn_dailymail_dm200 goldinspect
cnn_bart_m_dm200_v1coverage0.443200$0.000000 · 940 tokp50 6,927 · p95 11,473 ms1gold/cnn_dailymail_dm200 goldinspect
cnn_bart_s_dm200_v1coverage0.405200$0.000000 · 914 tokp50 4,207 · p95 5,571 ms1gold/cnn_dailymail_dm200 goldinspect
cnn_lead3_dm200_v1coverage0.370200$0.000000p50 0.1 · p95 0.27 ms1gold/cnn_dailymail_dm200 goldinspect
cnn_bart_l_xsum_dm200_v1coverage0.152200$0.000000 · 874 tokp50 5,103 · p95 8,017 ms1gold/cnn_dailymail_dm200 goldinspect

02what's inside

200 items, text, 110–1844 words (median 605, middle half 422–859).

Source: abisee/cnn_dailymail (test split). The text stays there; each item below links to its row.

02550110–283 words: 20284–457 words: 45458–631 words: 44632–805 words: 31806–979 words: 23980–1153 words: 191154–1327 words: 61328–1501 words: 31502–1675 words: 41676–1849 words: 5110–283284–457458–631632–805806–979980–11531154–13271328–15011502–16751676–1849

the gold

gold/cnn_dailymail_dm200: text, 22–130 words, median 54

200 shown
itemwordsgoldopen
00bfcb9d07fa52342 wordssource ↗ · inspect
01205b42980433041 wordssource ↗ · inspect
017738d8c5fc44249 wordssource ↗ · inspect
051db32933d747845 wordssource ↗ · inspect
0538c4e50be688666 wordssource ↗ · inspect
070cff525ba5679101 wordssource ↗ · inspect
086d9542139453154 wordssource ↗ · inspect
096dda15ca2767078 wordssource ↗ · inspect
0a09b7008372110880 wordssource ↗ · inspect
0fa083cb7da134183 wordssource ↗ · inspect
128939500473867130 wordssource ↗ · inspect
14514f8448a122752 wordssource ↗ · inspect
18a5d47e83e925942 wordssource ↗ · inspect
19d910abd292141184 wordssource ↗ · inspect
1a343914e24342039 wordssource ↗ · inspect
1bd44b6e16bf166581 wordssource ↗ · inspect
1c69b5c1b2fb88370 wordssource ↗ · inspect
1d965b7ef3e328161 wordssource ↗ · inspect
1ddc1d165b7442756 wordssource ↗ · inspect
1e717ed4d3b380934 wordssource ↗ · inspect
21030573ff6b89390 wordssource ↗ · inspect
22224ab649ac110152 wordssource ↗ · inspect
225f2330a64722859 wordssource ↗ · inspect
226e8811bb8b85264 wordssource ↗ · inspect
24f53c358c6755638 wordssource ↗ · inspect
289e33599f4386654 wordssource ↗ · inspect
2aa60a7b1f0233064 wordssource ↗ · inspect
2dfa0295b48256438 wordssource ↗ · inspect
2e2e6c66137561351 wordssource ↗ · inspect
2e40e0934dde85250 wordssource ↗ · inspect
2f13a3fa89fa41735 wordssource ↗ · inspect
304dfe02b1bd76251 wordssource ↗ · inspect
32076aaf737427551 wordssource ↗ · inspect
324799aced1d29059 wordssource ↗ · inspect
32b1f281483c42253 wordssource ↗ · inspect
349d234d54b555659 wordssource ↗ · inspect
36c34f4ffa8816840 wordssource ↗ · inspect
36d0dfb24f4542536 wordssource ↗ · inspect
37a8f67bffe7114862 wordssource ↗ · inspect
37b73620531773966 wordssource ↗ · inspect
388e3456094f33065 wordssource ↗ · inspect
3b8dc47341aa75453 wordssource ↗ · inspect
3c92b3c36e051813127 wordssource ↗ · inspect
3d06967875a874745 wordssource ↗ · inspect
3da318564d1142834 wordssource ↗ · inspect
3da4c2801fcc34364 wordssource ↗ · inspect
3dd47c7a4db126674 wordssource ↗ · inspect
4006ed714a5062548 wordssource ↗ · inspect
409d09b6f4fd74252 wordssource ↗ · inspect
41ea154a71ba130656 wordssource ↗ · inspect
43a900fa3160705106 wordssource ↗ · inspect
45ca61ee5a26125037 wordssource ↗ · inspect
46931c4bd3a6112837 wordssource ↗ · inspect
47b1af381aac64578 wordssource ↗ · inspect
48ba76cd72d134153 wordssource ↗ · inspect
48f52d03996f63450 wordssource ↗ · inspect
4ee59a5bf1ef35475 wordssource ↗ · inspect
518a7c88200120336 wordssource ↗ · inspect
53473bfa17eb82541 wordssource ↗ · inspect
53e798bb1a4f28722 wordssource ↗ · inspect
5584b67ffb2a170056 wordssource ↗ · inspect
55b8192b89cc27742 wordssource ↗ · inspect
58ca33afaedb95755 wordssource ↗ · inspect
5acf7aba835137886 wordssource ↗ · inspect
5b286830ff3579456 wordssource ↗ · inspect
5c0a90c2e1a5852100 wordssource ↗ · inspect
5fdd0e36b6c941473 wordssource ↗ · inspect
60f4f74a0e7749354 wordssource ↗ · inspect
62271db1dd4859850 wordssource ↗ · inspect
625c26dea34c89365 wordssource ↗ · inspect
63b17fa94a3145669 wordssource ↗ · inspect
685573e9aaed102065 wordssource ↗ · inspect
6a54725e5444852110 wordssource ↗ · inspect
6bcdaaa4ce3f104447 wordssource ↗ · inspect
6c1341bedf9223436 wordssource ↗ · inspect
6c6f1dbdfc2450876 wordssource ↗ · inspect
6caae1d0976583750 wordssource ↗ · inspect
6d1328c719e649450 wordssource ↗ · inspect
6d580f4aef621058106 wordssource ↗ · inspect
6d65c417c88931454 wordssource ↗ · inspect
6e201472a23685987 wordssource ↗ · inspect
6e405588649947070 wordssource ↗ · inspect
700b448e4556140757 wordssource ↗ · inspect
7428b0892cc149366 wordssource ↗ · inspect
74468085fc3969070 wordssource ↗ · inspect
7691a520145a782107 wordssource ↗ · inspect
78492369ef4f27231 wordssource ↗ · inspect
7afdcaba2fd047137 wordssource ↗ · inspect
7b27f7ffb9c946449 wordssource ↗ · inspect
7d15a0b9f8c470152 wordssource ↗ · inspect
7d8389d498d7130729 wordssource ↗ · inspect
7df38beb49dd68967 wordssource ↗ · inspect
822fd6ef846354738 wordssource ↗ · inspect
831ad546d0ac39749 wordssource ↗ · inspect
84c8f7cbb37c78554 wordssource ↗ · inspect
854e8429695871043 wordssource ↗ · inspect
862bbd27abe675075 wordssource ↗ · inspect
87626b6f72f283632 wordssource ↗ · inspect
894695f3151759667 wordssource ↗ · inspect
89baf55b264e49060 wordssource ↗ · inspect
8a5d6113450367332 wordssource ↗ · inspect
8a9a4d93589362360 wordssource ↗ · inspect
8c65ea56cda045131 wordssource ↗ · inspect
8f8c6175fd1429055 wordssource ↗ · inspect
9107fbb6760555546 wordssource ↗ · inspect
9130155af3f9184443 wordssource ↗ · inspect
91b7d231152711035 wordssource ↗ · inspect
92ce51c6b1b2116061 wordssource ↗ · inspect
9307675298b567572 wordssource ↗ · inspect
94e9c63bb76174250 wordssource ↗ · inspect
98274af48dff40855 wordssource ↗ · inspect
992a8f57a53360560 wordssource ↗ · inspect
9a31ceff27ac95533 wordssource ↗ · inspect
9aba5324e32e46633 wordssource ↗ · inspect
9d5a64da66af127443 wordssource ↗ · inspect
9dd43f7237d0617105 wordssource ↗ · inspect
9ef9bcf7b10259248 wordssource ↗ · inspect
a290765c547b34783 wordssource ↗ · inspect
a3238a368672107464 wordssource ↗ · inspect
a36ff64c7a7f154244 wordssource ↗ · inspect
a3b6201aca7742563 wordssource ↗ · inspect
a726904222fc170077 wordssource ↗ · inspect
a7486968048b53672 wordssource ↗ · inspect
a80add45ec4042235 wordssource ↗ · inspect
a864b02f2f8078646 wordssource ↗ · inspect
a9d468537ca843577 wordssource ↗ · inspect
a9fdb9e984bb93075 wordssource ↗ · inspect
ae49ea8ca26344253 wordssource ↗ · inspect
b0348c9c7e2625349 wordssource ↗ · inspect
b1585b611a9043437 wordssource ↗ · inspect
b18caeefd6e339189 wordssource ↗ · inspect
b23343f2af2a54252 wordssource ↗ · inspect
b45ab07ab745108538 wordssource ↗ · inspect
b52cb9c3722154450 wordssource ↗ · inspect
b5596ad0d22c60353 wordssource ↗ · inspect
b57ea4a4c707113338 wordssource ↗ · inspect
b5a377afaf9312038 wordssource ↗ · inspect
b773c7f158ae76342 wordssource ↗ · inspect
b77e01dfab7647950 wordssource ↗ · inspect
b8139bbbf3d739872 wordssource ↗ · inspect
b8f746629678109447 wordssource ↗ · inspect
b96ef482f5e6102958 wordssource ↗ · inspect
bcbdb86d4eba94352 wordssource ↗ · inspect
bcefd57fbf8c41861 wordssource ↗ · inspect
be575882b45f61950 wordssource ↗ · inspect
bf45bdab5ec890853 wordssource ↗ · inspect
bffca412fd1775580 wordssource ↗ · inspect
c0daee326074102767 wordssource ↗ · inspect
c3396652f6e7112561 wordssource ↗ · inspect
c3608b0a954364049 wordssource ↗ · inspect
c37431a19d8449170 wordssource ↗ · inspect
c4b1d91028a255169 wordssource ↗ · inspect
c4f93b339ca525136 wordssource ↗ · inspect
c57e46857d5a102858 wordssource ↗ · inspect
c58477f5ac31113788 wordssource ↗ · inspect
c5dcaf5c23ff159554 wordssource ↗ · inspect
c715c8dcc0fc61744 wordssource ↗ · inspect
cb7a72d78d5f29181 wordssource ↗ · inspect
cbf34d5ddfac48953 wordssource ↗ · inspect
cd064649d10639236 wordssource ↗ · inspect
d0767fea905857863 wordssource ↗ · inspect
d0b4d71e6d0232833 wordssource ↗ · inspect
d18f4335544e74595 wordssource ↗ · inspect
d243e0bdc63d109676 wordssource ↗ · inspect
d297355a018630445 wordssource ↗ · inspect
d453bd848a1458453 wordssource ↗ · inspect
d4bcbba8a28064077 wordssource ↗ · inspect
d4f5e19dfa49152859 wordssource ↗ · inspect
d5e3add9f88796742 wordssource ↗ · inspect
d6e1dbb112931357109 wordssource ↗ · inspect
d7972cbb7f52118646 wordssource ↗ · inspect
d987f48e741340339 wordssource ↗ · inspect
da653ae6641316538 wordssource ↗ · inspect
dc4c106fba9c41576 wordssource ↗ · inspect
dc6af035130335843 wordssource ↗ · inspect
e0f1759fd45249652 wordssource ↗ · inspect
e12a6d81e71578058 wordssource ↗ · inspect
e33105857efc106463 wordssource ↗ · inspect
e8fd26eb7eef18348 wordssource ↗ · inspect
eaa26b0b349b27455 wordssource ↗ · inspect
eb98777b970454150 wordssource ↗ · inspect
ec8df2c3e90062148 wordssource ↗ · inspect
ef581231850a52054 wordssource ↗ · inspect
f069049a037450753 wordssource ↗ · inspect
f0976e50cbdd106955 wordssource ↗ · inspect
f1a5b86c369927336 wordssource ↗ · inspect
f2c5101a077366854 wordssource ↗ · inspect
f30645fc7f73174066 wordssource ↗ · inspect
f3716b374462450111 wordssource ↗ · inspect
f48a1745c08f77698 wordssource ↗ · inspect
f5aa0a5183f736550 wordssource ↗ · inspect
f5ab37eb415d43552 wordssource ↗ · inspect
f60256b9f99343039 wordssource ↗ · inspect
f91a6e3839c781364 wordssource ↗ · inspect
f99f7c90a4ef68150 wordssource ↗ · inspect
f9a1388b5b5389852 wordssource ↗ · inspect
fa6a8b996f1a26037 wordssource ↗ · inspect
fc2d95d4b9e152641 wordssource ↗ · inspect
fd48dbfe117260769 wordssource ↗ · inspect
fe4531d4b8d741957 wordssource ↗ · inspect

Each row is the profile evh dataset-items --dataset-id cnn_dailymail_dm200 prints; inspect opens the item in the inspector with the dataset's largest group of comparable arms, and its text where the inspector runs on a machine that has it.

03reference sets

04labels

No labels yet. A sheet to fill: evh label-sheet --dataset-id cnn_dailymail_dm200 --out sheet.csv; a run's outputs as a model set: evh label-from-run --run R --set S.