FinalMLP: An Enhanced Two-Stream MLP Model for CTR Prediction
Paper • 2304.00902 • Published
label int64 0 1 | user_id int64 0 90.4k | item_id int64 1 90.4k | tag_id int64 2 69.1k |
|---|---|---|---|
0 | 51,798 | 2,473 | 37,583 |
0 | 66,335 | 61,344 | 29,842 |
0 | 89,085 | 60,033 | 47,050 |
1 | 61,293 | 8,073 | 3,903 |
0 | 81,335 | 56,575 | 50,067 |
0 | 65,166 | 48,181 | 12,510 |
0 | 75,300 | 26,027 | 38,510 |
1 | 10,219 | 2,122 | 383 |
1 | 80,855 | 80,856 | 24,728 |
1 | 67,033 | 721 | 19,495 |
0 | 80,574 | 4,352 | 1,885 |
0 | 90,194 | 13,389 | 13,229 |
0 | 81,469 | 34,789 | 24,788 |
1 | 76,049 | 56,307 | 51,431 |
0 | 63,553 | 1,662 | 46,915 |
0 | 86,940 | 13,820 | 44,681 |
1 | 39,479 | 39,496 | 972 |
0 | 68,854 | 57,527 | 41,338 |
0 | 85,648 | 488 | 24,424 |
1 | 86,148 | 10,155 | 82 |
0 | 65,166 | 25,552 | 27,746 |
1 | 68,854 | 63,464 | 9,130 |
0 | 67,886 | 800 | 46,637 |
0 | 74,824 | 17,296 | 8,180 |
0 | 86,498 | 41,080 | 28,484 |
0 | 65,166 | 60,615 | 28,617 |
1 | 71,708 | 25,715 | 764 |
0 | 85,231 | 16,221 | 16,180 |
0 | 80,954 | 62,140 | 43,799 |
0 | 80,874 | 1,656 | 34,769 |
0 | 84,982 | 603 | 26,084 |
0 | 80,034 | 11,717 | 54,781 |
1 | 58,995 | 13,820 | 12,684 |
0 | 66,217 | 57,129 | 24,385 |
1 | 82,033 | 6,833 | 10,044 |
1 | 80,179 | 12,914 | 17,734 |
0 | 68,658 | 8,171 | 28,680 |
1 | 81,557 | 26,261 | 10,504 |
0 | 88,928 | 67,064 | 27,341 |
0 | 71,683 | 29,781 | 6,713 |
1 | 66,893 | 501 | 14,782 |
0 | 85,606 | 721 | 24,507 |
0 | 78,897 | 40,815 | 30,825 |
0 | 87,624 | 8,171 | 35,649 |
0 | 67,270 | 15,255 | 39,558 |
0 | 59,125 | 603 | 22,563 |
0 | 86,301 | 21,868 | 2,516 |
0 | 86,299 | 46,892 | 49,832 |
1 | 85,783 | 19,482 | 11,347 |
1 | 90,194 | 10,099 | 477 |
1 | 57,686 | 33,143 | 1,814 |
0 | 61,861 | 54,364 | 35,746 |
0 | 78,005 | 13,687 | 25,123 |
1 | 68,220 | 65,248 | 6,802 |
1 | 81,792 | 28,147 | 10,444 |
0 | 66,590 | 12,833 | 23,703 |
0 | 68,854 | 60,433 | 3,848 |
0 | 87,689 | 1,630 | 40,066 |
1 | 38,499 | 8,965 | 13,046 |
0 | 74,856 | 7,523 | 19,303 |
0 | 75,092 | 19,389 | 33,433 |
1 | 60,494 | 53,608 | 33,158 |
0 | 82,036 | 73,796 | 58,562 |
1 | 55,712 | 392 | 409 |
0 | 74,724 | 47,209 | 29,720 |
1 | 52,064 | 11,105 | 11,878 |
1 | 84,833 | 5,150 | 5,157 |
1 | 84,910 | 1,846 | 35,538 |
0 | 49,624 | 10,355 | 45,183 |
1 | 86,072 | 70,460 | 3,831 |
0 | 87,509 | 1,508 | 10,010 |
0 | 76,824 | 15,783 | 19,734 |
0 | 72,038 | 11,495 | 45,541 |
0 | 85,911 | 85,939 | 20,046 |
1 | 78,559 | 68,148 | 14,525 |
0 | 77,375 | 1,678 | 47,379 |
0 | 54,423 | 653 | 21,760 |
0 | 90,077 | 23,748 | 6,355 |
0 | 68,854 | 69,666 | 36,331 |
1 | 64,128 | 20,609 | 13,481 |
1 | 87,906 | 16,842 | 2,012 |
0 | 72,038 | 10,099 | 33,513 |
0 | 75,774 | 5,573 | 15,708 |
1 | 86,940 | 36,699 | 1,898 |
1 | 78,559 | 13,577 | 3,302 |
0 | 87,742 | 30,648 | 60,288 |
1 | 62,154 | 60,463 | 914 |
1 | 76,877 | 33,569 | 3,885 |
0 | 68,854 | 22,624 | 9,140 |
1 | 66,779 | 10,099 | 477 |
1 | 59,709 | 59,917 | 8,534 |
0 | 64,920 | 7,504 | 55,530 |
0 | 43,874 | 1,052 | 36,230 |
1 | 67,393 | 50,423 | 1,923 |
0 | 49,624 | 12,512 | 13,530 |
1 | 77,567 | 63,583 | 12,223 |
0 | 80,103 | 49,406 | 29,856 |
0 | 62,154 | 3,160 | 28,672 |
0 | 46,657 | 7,537 | 3,166 |
1 | 63,329 | 1,508 | 4,553 |
YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
The MovieLens dataset consists of users' tagging records on movies. The task is formulated as personalized tag recommendation with each tagging record (user_id, item_id, tag_id) as an data instance. The target value denotes whether the user has assigned a particular tag to the movie. We provide the reusable, processed dataset released by the BARS benchmark, which are randomly split into 7:2:1 as the training set, validation set, and test set, respectively.
Repository: https://github.com/reczoo/BARS/blob/main/datasets/MovieLens/README.md#movielenslatest_x1
Used by papers:
Check the md5sum for data integrity:
```bash
$ md5sum train.csv valid.csv test.csv
efc8bceeaa0e895d566470fc99f3f271 train.csv
e1930223a5026e910ed5a48687de8af1 valid.csv
54e8c6baff2e059fe067fb9b69e692d0 test.csv
```