"He Said She Said" classification challenge (2nd edition)

Give the probability that a text in Polish was written by a man. [ver. 2.0.2]

Git repo URL: git://gonito.net/petite-difference-challenge2 / Branch: master
Run git clone --single-branch git://gonito.net/petite-difference-challenge2 -b master to get the challenge data
Browse at https://gonito.net/gitlist/petite-difference-challenge2.git/master

Leaderboard

# submitter when ver. description test-A Accuracy test-A Likelihood ×
1 kaczla 2020-07-02 08:55 2.0.2 Polish RoBERTa (base), epoch 5, seq_len 512, active dropout fairseq roberta-pl 0.74332 0.62110 30
2 kubapok 2020-06-19 09:29 2.0.2 pl roberta large active dropout avg 12 runs 0.74406 0.61949 28
3 [anonymized] 2020-05-24 14:52 2.0.2 self-made NB with probs ISI-2019-063 probabilities 0.65612 0.52537 5
4 [anonymized] 2020-05-24 14:17 2.0.0 v5 probabilities 0.64427 0.52134 7
5 [anonymized] 2020-06-07 12:45 2.0.0 XGBoost ready-made ready-made xgboost 0.60039 0.52000 5
6 [anonymized] 2020-12-13 22:48 2.0.2 model-size=100k voc-size=100k logistic-regression pytorch-nn 0.60299 0.51313 2
7 wirus wirus 2020-05-23 21:29 2.0.0 null model null-model 0.50000 0.50000 24
8 Adam Wojdyła 2022-04-26 16:15 2.0.2 4444507 self-made 0.50000 0.49749 2
9 [anonymized] 2020-06-03 17:39 2.0.0 2nd logistic-regression word2vec 0.49299 0.49679 2
10 [anonymized] 2020-12-08 15:08 2.0.2 pytorch logistic regression logistic-regression pytorch-nn 0.54176 0.43930 4
11 [anonymized] 2021-02-03 17:42 2.0.2 TAU15 logistic-regression pytorch-nn 0.50074 0.00000 2
12 [anonymized] 2021-02-01 20:38 2.0.2 logistic-regression logistic-regression pytorch-nn 0.63924 0.00000 3
13 [anonymized] 2020-12-15 22:09 2.0.2 added code with output for dev-0, dev-1, test-A logistic-regression pytorch-nn 0.49317 0.00000 1
14 [anonymized] 2020-12-15 21:55 2.0.2 added source file logistic-regression pytorch-nn 0.49728 0.00000 3
15 [anonymized] 2021-01-27 02:20 2.0.2 1try logistic-regression pytorch-nn 0.49914 0.00000 1
16 [anonymized] 2021-01-26 23:25 2.0.2 try again logistic-regression pytorch-nn 0.53999 0.00000 2
17 [anonymized] 2020-12-16 08:42 2.0.2 'final' logistic-regression pytorch-nn 0.50878 0.00000 1
18 [anonymized] 2020-12-07 18:12 2.0.2 add code and fixed test-A logistic-regression pytorch-nn 0.57926 0.00000 2
19 [anonymized] 2020-12-09 07:50 2.0.2 Simple Solution for Logistic Regresion logistic-regression pytorch-nn 0.58355 0.00000 2
20 [anonymized] 2020-12-10 01:01 2.0.2 solution logistic-regression pytorch-nn 0.58344 0.00000 5
21 [anonymized] 2021-03-13 14:48 2.0.2 my brilliant solution logistic-regression pytorch-nn 0.53925 0.00000 1
22 s478839 2022-04-25 22:13 2.0.2 s478839 self-made 0.51284 0.00000 1
23 s444417 2022-04-23 08:45 2.0.2 imbalance words self-made 0.51237 0.00000 2
24 s443930 2022-04-26 21:44 2.0.2 s443930 self-made 0.64618 0.00000 1
25 Jakub Adamski 2022-04-20 06:45 2.0.2 s444341 zadanie self-made 0.58960 0.00000 1
26 s409771 2022-04-22 15:12 2.0.2 multinomial naive bayes self-made 0.65503 0.00000 1
27 s444354 2022-04-26 23:26 2.0.2 s444354 0.51333 0.00000 6
28 s478815 2022-04-27 10:35 2.0.2 478815 self-made 0.52032 0.00000 8
29 ked 2022-04-22 15:09 2.0.2 s449288 - dumb wordlist lookup self-made 0.51464 0.00000 2
30 s478873 2022-04-25 22:18 2.0.2 s478873 self-made 0.51352 0.00000 1
31 Cezary 2022-04-26 22:57 2.0.2 s470623 0.51036 0.00000 2
32 [anonymized] 2022-04-25 23:49 2.0.2 478841 self-made 0.66702 0.00000 1
33 Jakub 2022-04-27 09:04 2.0.2 s434624 self-made 0.51005 0.00000 1
34 [anonymized] 2022-04-26 20:41 2.0.2 s478831 0.51241 0.00000 1
35 [anonymized] 2023-11-03 12:15 2.0.2 s444498 prod test another repo algo 0.51464 0.00000 1
36 s444476 2022-04-22 15:59 2.0.2 s444476 self-made 0.51889 0.00000 2
37 444498 2023-05-19 07:07 2.0.2 test tag request algo attention backoff 0.63615 0.00000 7
38 Mikołaj Pokrywka 2022-04-26 14:14 2.0.2 444463 self-made 0.66439 0.00000 3
39 s444517 2022-04-25 07:39 2.0.2 s444517 - logistic regression self-made 0.66705 0.00000 1
40 s478855 2022-04-20 19:16 2.0.2 478855 - improvement self-made 0.51230 0.00000 2
41 s478840 2022-04-26 05:20 2.0.2 s478840 self-made 0.66531 0.00000 1
42 s478846 2022-04-21 08:55 2.0.2 First solution s478846 self-made 0.51599 0.00000 1
43 s444356 2022-04-20 11:04 2.0.2 s444356 self-made 0.51307 0.00000 3
44 s444386 2022-04-25 15:56 2.0.2 logistic regresion 444386 self-made 0.66769 0.00000 1
45 s444501 2022-04-22 17:31 2.0.2 444501 self-made 0.51405 0.00000 3
46 Kamil Guttmann 2022-04-30 15:42 2.0.2 s444380 logistic regression bigrams self-made 0.67034 0.00000 2
47 [anonymized] 2022-04-25 17:13 2.0.2 444421 self-made 0.66394 0.00000 1
48 s444465 2022-04-25 20:36 2.0.2 Solution 444465 self-made 0.62985 0.00000 4
49 s444452 2022-04-26 21:05 2.0.2 444452 self-made 0.63615 0.00000 2
50 s444018 2022-04-26 14:19 2.0.2 s444018 self-made 0.51307 0.00000 3
51 grzeslaw 2023-11-03 12:19 2.0.2 grzeslaw test another repo beta 5gram base 0.52032 0.00000 1
52 devel 2023-11-03 11:50 2.0.2 front devel test 4 base 0.51885 0.00000 1
# tags test-A Accuracy -C test-A Accuracy +C test-A Accuracy +H test-A Likelihood -C test-A Likelihood +C test-A Likelihood +H test-A Accuracy test-A Likelihood
1 fairseq roberta-pl 0.74244 0.77159 0.77125 0.62032 0.64656 0.64332 0.74332 0.62110
2 0.74349 0.76227 0.77125 0.61881 0.64193 0.63970 0.74406 0.61949
3 roberta-xlm 0.70042 0.72571 0.71500 0.58222 0.60181 0.59521 0.70118 0.58280
4 fairseq roberta N/A N/A N/A N/A N/A N/A 0.69153 0.57068
5 probabilities 0.65572 0.66879 0.64500 0.52524 0.52950 0.52176 0.65612 0.52537
6 ready-made xgboost N/A N/A N/A N/A N/A N/A 0.60039 0.52000
7 logistic-regression pytorch-nn 0.60208 0.63199 0.60500 0.51296 0.51866 0.52272 0.60299 0.51313
8 null-model N/A N/A N/A N/A N/A N/A 0.50000 0.50000
9 self-made 0.50162 0.44799 0.48250 0.49766 0.49233 0.49575 0.50000 0.49749
10 logistic-regression word2vec N/A N/A N/A N/A N/A N/A 0.49299 0.49679
11 baseline N/A N/A N/A N/A N/A N/A 0.50000 0.48990
12 algo 0.51510 0.50000 0.47250 0.00000 0.00000 0.00000 0.51464 0.00000
13 algo attention backoff 0.63534 0.66192 0.64750 0.00000 0.00000 0.00000 0.63615 0.00000
14 beta 5gram base 0.51904 0.56109 0.54375 0.00000 0.00000 0.00000 0.52032 0.00000
15 ready-made svm N/A N/A N/A N/A N/A N/A 0.59623 0.00000