385:
344:
In 2016 Machine
Learning Infrastructure team led by Anna Dorogush started working on Gradient Boosting in Yandex, including Matrixnet and Tensornet. They implemented and open-sourced the next version of Gradient Boosting library called CatBoost, which has support of categorical and text data, GPU
285:
listed CatBoost as one of the most frequently used machine learning (ML) frameworks in the world. It was listed as the top-8 most frequently used ML framework in the 2020 survey and as the top-7 most frequently used ML framework in the 2021 survey.
27:
334:, a proprietary gradient boosting library that was used in Yandex to rank search results. Since 2009 MatrixNet has been used in different projects in Yandex, including recommendation systems and weather prediction.
886:
896:
568:
Prokhorenkova, Liudmila; Gusev, Gleb; Vorobev, Aleksandr; Dorogush, Anna
Veronika; Gulin, Andrey (2019-01-20). "CatBoost: unbiased boosting with categorical features".
207:
framework which among other features attempts to solve for categorical features using a permutation driven alternative compared to the classical algorithm. It works on
876:
337:
In 2014–2015 Andrey Gulin with a team of researchers has started a new project called
Tensornet that was aimed at solving the problem of "how to work with
906:
881:
871:
474:
901:
390:
500:
743:
Dorogush, Anna
Veronika; Ershov, Vasily; Gulin, Andrey (2018-10-24). "CatBoost: gradient boosting with categorical features support".
891:
654:
236:
341:". It resulted in several proprietary Gradient Boosting libraries with different approaches to handling categorical data.
220:
111:
789:
301:
CatBoost has gained popularity compared to other gradient boosting algorithms primarily due to the following features
240:
232:
123:
82:
764:
807:
348:
CatBoost was open-sourced in July 2017 and is under active development in Yandex and the open-source community.
916:
315:
224:
115:
719:
911:
825:
629:
338:
193:
196:
150:
744:
569:
167:
403:
204:
398:
162:
155:
130:
256:
865:
418:
267:
magazine awarded the library "The best machine learning tools" in 2017. along with
524:
609:
380:
363:
268:
47:
34:
686:
672:
475:"Yandex open sources CatBoost, a gradient boosting machine learning library"
438:
357:
331:
263:
700:
26:
589:
413:
289:
As of April 2022, CatBoost is installed about 100000 times per day from
851:
551:
456:
408:
276:
272:
244:
212:
143:
765:"CatBoost Enables Fast Gradient Boosting on Decision Trees Using GPUs"
369:
282:
200:
52:
40:
856:
749:
574:
228:
119:
216:
208:
139:
135:
826:"How Careem's Destination Prediction Service speeds up your ride"
227:, and models built using CatBoost can be used for predictions in
290:
252:
248:
563:
561:
630:"Maven Repository: ai.catboost » catboost-prediction"
372:
uses CatBoost to predict future destinations of the rides
846:
501:"Yandex open sources CatBoost machine learning library"
311:
Visualizations and tools for model and feature analysis
178:
808:"Stop the Bots: Practical Lessons in Machine Learning"
655:"Bossie Awards 2017: The best machine learning tools"
173:
161:
149:
129:
107:
81:
59:
46:
33:
887:Python (programming language) scientific libraries
687:"State of Data Science and Machine Learning 2021"
673:"State of Data Science and Machine Learning 2020"
16:Open-source software library developed by Yandex
345:training, model analysis, visualisation tools.
546:
544:
318:trees or symmetric trees for faster execution
8:
790:"Code Completion, Episode 4: Model Training"
439:"Andrey Gulin - People - Research at Yandex"
19:
25:
18:
897:Data mining and machine learning software
748:
573:
321:Ordered boosting to overcome overfitting
305:Native handling for categorical features
430:
653:staff, InfoWorld (27 September 2017).
590:"Python Package Index PYPI: catboost"
554:. August 30, 2020 – via GitHub.
7:
391:Free and open-source software portal
255:. The source code is licensed under
877:Open-source artificial intelligence
720:"The Gradient Boosters V: CatBoost"
14:
882:Software using the Apache license
360:uses CatBoost for code completion
88:1.2.3 / February 23, 2024
610:"Conda force package catboost-r"
383:
907:Free software programmed in C++
499:Yegulalp, Serdar (2017-07-18).
366:uses CatBoost for bot detection
330:In 2009 Andrey Gulin developed
525:"Releases · catboost/catboost"
1:
872:Free and open-source software
857:CatBoost - Yandex Technology
902:Free data analysis software
718:Joseph, Manu (2020-02-29).
933:
852:GitHub - catboost/catboost
259:and available on GitHub.
103:
77:
55:and CatBoost Contributors
24:
892:Applied machine learning
794:JetBrains Developer Blog
279:and 8 other libraries.
219:, and is available in
90:; 6 months ago
769:NVIDIA Developer Blog
701:"PyPI Stats catboost"
65:; 7 years ago
812:The Cloudflare Blog
552:"catboost/catboost"
457:"catboost/catboost"
443:research.yandex.com
21:
724:Deep & Shallow
168:Apache License 2.0
63:July 18, 2017
35:Original author(s)
634:mvnrepository.com
404:Gradient boosting
308:Fast GPU training
205:gradient boosting
187:
186:
924:
834:
833:
822:
816:
815:
804:
798:
797:
786:
780:
779:
777:
776:
761:
755:
754:
752:
740:
734:
733:
731:
730:
715:
709:
708:
697:
691:
690:
683:
677:
676:
669:
663:
662:
650:
644:
643:
641:
640:
626:
620:
619:
617:
616:
606:
600:
599:
597:
596:
586:
580:
579:
577:
565:
556:
555:
548:
539:
538:
536:
535:
521:
515:
514:
512:
511:
496:
490:
489:
487:
486:
471:
465:
464:
453:
447:
446:
435:
399:Machine learning
393:
388:
387:
386:
339:categorical data
203:. It provides a
197:software library
183:
180:
156:Machine learning
131:Operating system
98:
96:
91:
73:
71:
66:
39:Andrey Gulin: /
29:
22:
932:
931:
927:
926:
925:
923:
922:
921:
917:Yandex software
862:
861:
843:
838:
837:
824:
823:
819:
806:
805:
801:
788:
787:
783:
774:
772:
763:
762:
758:
742:
741:
737:
728:
726:
717:
716:
712:
699:
698:
694:
685:
684:
680:
671:
670:
666:
652:
651:
647:
638:
636:
628:
627:
623:
614:
612:
608:
607:
603:
594:
592:
588:
587:
583:
567:
566:
559:
550:
549:
542:
533:
531:
523:
522:
518:
509:
507:
498:
497:
493:
484:
482:
473:
472:
468:
455:
454:
450:
437:
436:
432:
427:
389:
384:
382:
379:
354:
328:
299:
177:
99:
94:
92:
89:
69:
67:
64:
60:Initial release
17:
12:
11:
5:
930:
928:
920:
919:
914:
909:
904:
899:
894:
889:
884:
879:
874:
864:
863:
860:
859:
854:
849:
842:
841:External links
839:
836:
835:
817:
799:
781:
756:
735:
710:
692:
678:
664:
645:
621:
601:
581:
557:
540:
516:
491:
481:. 18 July 2017
466:
448:
429:
428:
426:
423:
422:
421:
416:
411:
406:
401:
395:
394:
378:
375:
374:
373:
367:
361:
353:
350:
327:
324:
323:
322:
319:
312:
309:
306:
298:
295:
257:Apache License
185:
184:
175:
171:
170:
165:
159:
158:
153:
147:
146:
133:
127:
126:
109:
105:
104:
101:
100:
87:
85:
83:Stable release
79:
78:
75:
74:
61:
57:
56:
50:
44:
43:
37:
31:
30:
15:
13:
10:
9:
6:
4:
3:
2:
929:
918:
915:
913:
912:2017 software
910:
908:
905:
903:
900:
898:
895:
893:
890:
888:
885:
883:
880:
878:
875:
873:
870:
869:
867:
858:
855:
853:
850:
848:
845:
844:
840:
832:. 2019-02-19.
831:
827:
821:
818:
814:. 2019-02-20.
813:
809:
803:
800:
796:. 2021-08-20.
795:
791:
785:
782:
770:
766:
760:
757:
751:
746:
739:
736:
725:
721:
714:
711:
706:
702:
696:
693:
688:
682:
679:
674:
668:
665:
660:
656:
649:
646:
635:
631:
625:
622:
611:
605:
602:
591:
585:
582:
576:
571:
564:
562:
558:
553:
547:
545:
541:
530:
526:
520:
517:
506:
502:
495:
492:
480:
476:
470:
467:
462:
458:
452:
449:
444:
440:
434:
431:
424:
420:
417:
415:
412:
410:
407:
405:
402:
400:
397:
396:
392:
381:
376:
371:
368:
365:
362:
359:
356:
355:
351:
349:
346:
342:
340:
335:
333:
325:
320:
317:
313:
310:
307:
304:
303:
302:
296:
294:
292:
287:
284:
280:
278:
274:
270:
266:
265:
260:
258:
254:
250:
246:
242:
238:
234:
230:
226:
222:
218:
214:
210:
206:
202:
199:developed by
198:
195:
191:
182:
176:
172:
169:
166:
164:
160:
157:
154:
152:
148:
145:
141:
137:
134:
132:
128:
125:
121:
117:
113:
110:
106:
102:
86:
84:
80:
76:
62:
58:
54:
51:
49:
45:
42:
38:
36:
32:
28:
23:
829:
820:
811:
802:
793:
784:
773:. Retrieved
771:. 2018-12-13
768:
759:
738:
727:. Retrieved
723:
713:
704:
695:
681:
667:
658:
648:
637:. Retrieved
633:
624:
613:. Retrieved
604:
593:. Retrieved
584:
532:. Retrieved
528:
519:
508:. Retrieved
504:
494:
483:. Retrieved
478:
469:
460:
451:
442:
433:
419:scikit-learn
347:
343:
336:
329:
300:
288:
281:
262:
261:
189:
188:
48:Developer(s)
352:Application
293:repository
194:open-source
866:Categories
775:2020-08-30
750:1810.11363
729:2020-08-30
705:PyPI Stats
639:2020-08-30
615:2020-08-30
595:2020-08-20
575:1706.09516
534:2024-03-14
510:2020-08-30
485:2020-08-30
479:TechCrunch
425:References
364:Cloudflare
269:TensorFlow
108:Written in
95:2024-02-23
70:2017-07-18
659:InfoWorld
505:InfoWorld
358:JetBrains
332:MatrixNet
316:oblivious
264:InfoWorld
847:CatBoost
414:LightGBM
377:See also
297:Features
190:CatBoost
179:catboost
20:CatBoost
409:XGBoost
326:History
277:XGBoost
273:Pytorch
245:Core ML
213:Windows
174:Website
163:License
144:Windows
93: (
68: (
830:Careem
529:GitHub
461:GitHub
370:Careem
314:Using
283:Kaggle
251:, and
221:Python
201:Yandex
192:is an
112:Python
53:Yandex
41:Yandex
745:arXiv
570:arXiv
217:macOS
209:Linux
140:macOS
136:Linux
291:PyPI
253:PMML
249:ONNX
241:Rust
233:Java
151:Type
124:Java
229:C++
223:,
181:.ai
120:C++
868::
828:.
810:.
792:.
767:.
722:.
703:.
657:.
632:.
560:^
543:^
527:.
503:.
477:.
459:.
441:.
275:,
271:,
247:,
243:,
239:,
237:C#
235:,
231:,
215:,
211:,
142:,
138:,
122:,
118:,
114:,
778:.
753:.
747::
732:.
707:.
689:.
675:.
661:.
642:.
618:.
598:.
578:.
572::
537:.
513:.
488:.
463:.
445:.
225:R
116:R
97:)
72:)
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.
