Knowledge (XXG)

Horovod (machine learning)

Source 📝

170: 1497: 583: 1477: 26: 352:
Horovod, a distributed training framework for TensorFlow, Keras and PyTorch, improves speed, scale and resource allocation in machine learning training activities. Uber uses Horovod for self-driving vehicles, fraud detection, and trip forecasting. It is also being used by Alibaba, Amazon and
1561: 1371: 1213: 619: 408: 1556: 1546: 1566: 1541: 311: 729: 612: 1402: 428: 417: 401: 226: 1503: 1054: 791: 1551: 1315: 942: 749: 605: 1270: 1536: 394: 1457: 1397: 995: 217:(LF AI). Horovod has the goal of improving the speed, scale, and resource allocation when training a machine learning model. 990: 679: 96: 1432: 786: 739: 734: 516: 1483: 779: 705: 481: 67: 1107: 1042: 643: 231: 1508: 1366: 1005: 836: 659: 1407: 664: 511: 1452: 1437: 1090: 1085: 985: 853: 634: 136: 1531: 1412: 1172: 891: 886: 586: 1442: 1427: 1392: 1080: 980: 848: 1310: 1462: 1417: 863: 808: 654: 649: 536: 438: 187: 169: 1037: 1015: 764: 759: 717: 669: 562: 131: 1476: 1422: 1000: 829: 190: 149: 111: 1488: 1280: 932: 803: 796: 552: 506: 1233: 1223: 1030: 824: 774: 769: 712: 700: 456: 214: 144: 1346: 1290: 1112: 754: 674: 557: 257: 1320: 1285: 1275: 1100: 858: 684: 461: 1525: 1265: 1245: 1162: 841: 194: 183: 1351: 1182: 451: 446: 210: 1447: 1218: 1127: 1122: 744: 722: 312:"Uber brings Horovod project for distributed deep learning to Linux Foundation" 1341: 1300: 1295: 1208: 1117: 1025: 937: 917: 544: 501: 285: 236: 198: 33: 1336: 1305: 1203: 1047: 1010: 947: 901: 896: 881: 526: 466: 597: 1238: 1070: 521: 386: 25: 1361: 1198: 1152: 1075: 975: 970: 922: 547: 496: 206: 124: 1376: 1356: 1228: 1020: 567: 491: 486: 381: 377: 336: 100: 1177: 1157: 1147: 1142: 1137: 1132: 1095: 927: 476: 202: 120: 116: 1167: 471: 104: 38: 601: 390: 370: 160: 1385: 1329: 1258: 1191: 1063: 963: 956: 910: 874: 817: 693: 633: 535: 437: 155: 143: 130: 110: 92: 66: 44: 32: 1562:Python (programming language) scientific libraries 613: 402: 8: 18: 960: 620: 606: 598: 409: 395: 387: 168: 24: 17: 248: 7: 1458:Generative adversarial network (GAN) 227:Comparison of deep learning software 1557:Open-source artificial intelligence 310:Khari Johnson (December 13, 2018). 1547:Free software programmed in Python 258:"Release v0.9.0 · horovod/horovod" 14: 1567:Software using the Apache license 1496: 1495: 1475: 582: 581: 1542:Free software programmed in C++ 256:Alex Sergeev (August 9, 2017). 73:v0.28.1 / June 12, 2023 1408:Recurrent neural network (RNN) 1398:Differentiable neural computer 213:. Horovod is hosted under the 1: 1453:Variational autoencoder (VAE) 1413:Long short-term memory (LSTM) 680:Computational learning theory 1433:Convolutional neural network 286:"Releases · horovod/horovod" 1428:Multilayer perceptron (MLP) 482:Microsoft Cognitive Toolkit 1583: 1504:Artificial neural networks 1418:Gated recurrent unit (GRU) 644:Differentiable programming 232:Differentiable programming 1552:Free statistical software 1471: 837:Artificial neural network 660:Automatic differentiation 577: 424: 88: 62: 23: 665:Neuromorphic engineering 628:Differentiable computing 341:Linux Foundation - LF AI 1438:Residual neural network 854:Artificial Intelligence 137:Artificial intelligence 75:; 15 months ago 1537:Deep learning software 418:Deep learning software 1393:Neural Turing machine 981:Human image synthesis 570:Deep Learning Toolbox 50:; 7 years ago 1484:Computer programming 1463:Graph neural network 1038:Text-to-video models 1016:Text-to-image models 864:Large language model 849:Scientific computing 655:Statistical manifold 650:Information geometry 830:In-context learning 670:Pattern recognition 563:Wolfram Mathematica 215:Linux Foundation AI 48:August 9, 2017 20: 1423:Echo state network 1311:Jürgen Schmidhuber 1006:Facial recognition 1001:Speech recognition 911:Software libraries 337:"Projects - LF AI" 191:software framework 150:Apache License 2.0 1519: 1518: 1281:Stephen Grossberg 1254: 1253: 595: 594: 177: 176: 1574: 1509:Machine learning 1499: 1498: 1479: 1234:Action selection 1224:Self-driving car 1031:Stable Diffusion 996:Speech synthesis 961: 825:Machine learning 701:Gradient descent 622: 615: 608: 599: 585: 584: 411: 404: 397: 388: 374: 373: 371:Official website 356: 355: 349: 347: 333: 327: 326: 324: 322: 307: 301: 300: 298: 296: 282: 276: 275: 270: 268: 253: 193:for distributed 173: 172: 165: 162: 83: 81: 76: 58: 56: 51: 28: 21: 1582: 1581: 1577: 1576: 1575: 1573: 1572: 1571: 1522: 1521: 1520: 1515: 1467: 1381: 1347:Google DeepMind 1325: 1291:Geoffrey Hinton 1250: 1187: 1113:Project Debater 1059: 957:Implementations 952: 906: 870: 813: 755:Backpropagation 689: 675:Tensor calculus 629: 626: 596: 591: 573: 558:Neural Designer 531: 433: 420: 415: 369: 368: 365: 360: 359: 345: 343: 335: 334: 330: 320: 318: 309: 308: 304: 294: 292: 284: 283: 279: 273:Initial release 266: 264: 255: 254: 250: 245: 223: 197:training using 167: 159: 84: 79: 77: 74: 54: 52: 49: 45:Initial release 12: 11: 5: 1580: 1578: 1570: 1569: 1564: 1559: 1554: 1549: 1544: 1539: 1534: 1524: 1523: 1517: 1516: 1514: 1513: 1512: 1511: 1506: 1493: 1492: 1491: 1486: 1472: 1469: 1468: 1466: 1465: 1460: 1455: 1450: 1445: 1440: 1435: 1430: 1425: 1420: 1415: 1410: 1405: 1400: 1395: 1389: 1387: 1383: 1382: 1380: 1379: 1374: 1369: 1364: 1359: 1354: 1349: 1344: 1339: 1333: 1331: 1327: 1326: 1324: 1323: 1321:Ilya Sutskever 1318: 1313: 1308: 1303: 1298: 1293: 1288: 1286:Demis Hassabis 1283: 1278: 1276:Ian Goodfellow 1273: 1268: 1262: 1260: 1256: 1255: 1252: 1251: 1249: 1248: 1243: 1242: 1241: 1231: 1226: 1221: 1216: 1211: 1206: 1201: 1195: 1193: 1189: 1188: 1186: 1185: 1180: 1175: 1170: 1165: 1160: 1155: 1150: 1145: 1140: 1135: 1130: 1125: 1120: 1115: 1110: 1105: 1104: 1103: 1093: 1088: 1083: 1078: 1073: 1067: 1065: 1061: 1060: 1058: 1057: 1052: 1051: 1050: 1045: 1035: 1034: 1033: 1028: 1023: 1013: 1008: 1003: 998: 993: 988: 983: 978: 973: 967: 965: 958: 954: 953: 951: 950: 945: 940: 935: 930: 925: 920: 914: 912: 908: 907: 905: 904: 899: 894: 889: 884: 878: 876: 872: 871: 869: 868: 867: 866: 859:Language model 856: 851: 846: 845: 844: 834: 833: 832: 821: 819: 815: 814: 812: 811: 809:Autoregression 806: 801: 800: 799: 789: 787:Regularization 784: 783: 782: 777: 772: 762: 757: 752: 750:Loss functions 747: 742: 737: 732: 727: 726: 725: 715: 710: 709: 708: 697: 695: 691: 690: 688: 687: 685:Inductive bias 682: 677: 672: 667: 662: 657: 652: 647: 639: 637: 631: 630: 627: 625: 624: 617: 610: 602: 593: 592: 590: 589: 578: 575: 574: 572: 571: 565: 560: 555: 550: 541: 539: 533: 532: 530: 529: 524: 519: 514: 509: 504: 499: 494: 489: 484: 479: 474: 469: 464: 462:Deeplearning4j 459: 454: 449: 443: 441: 435: 434: 432: 431: 425: 422: 421: 416: 414: 413: 406: 399: 391: 385: 384: 375: 364: 363:External links 361: 358: 357: 328: 302: 277: 247: 246: 244: 241: 240: 239: 234: 229: 222: 219: 175: 174: 157: 153: 152: 147: 141: 140: 134: 128: 127: 114: 108: 107: 94: 90: 89: 86: 85: 72: 70: 68:Stable release 64: 63: 60: 59: 46: 42: 41: 36: 30: 29: 13: 10: 9: 6: 4: 3: 2: 1579: 1568: 1565: 1563: 1560: 1558: 1555: 1553: 1550: 1548: 1545: 1543: 1540: 1538: 1535: 1533: 1532:2017 software 1530: 1529: 1527: 1510: 1507: 1505: 1502: 1501: 1494: 1490: 1487: 1485: 1482: 1481: 1478: 1474: 1473: 1470: 1464: 1461: 1459: 1456: 1454: 1451: 1449: 1446: 1444: 1441: 1439: 1436: 1434: 1431: 1429: 1426: 1424: 1421: 1419: 1416: 1414: 1411: 1409: 1406: 1404: 1401: 1399: 1396: 1394: 1391: 1390: 1388: 1386:Architectures 1384: 1378: 1375: 1373: 1370: 1368: 1365: 1363: 1360: 1358: 1355: 1353: 1350: 1348: 1345: 1343: 1340: 1338: 1335: 1334: 1332: 1330:Organizations 1328: 1322: 1319: 1317: 1314: 1312: 1309: 1307: 1304: 1302: 1299: 1297: 1294: 1292: 1289: 1287: 1284: 1282: 1279: 1277: 1274: 1272: 1269: 1267: 1266:Yoshua Bengio 1264: 1263: 1261: 1257: 1247: 1246:Robot control 1244: 1240: 1237: 1236: 1235: 1232: 1230: 1227: 1225: 1222: 1220: 1217: 1215: 1212: 1210: 1207: 1205: 1202: 1200: 1197: 1196: 1194: 1190: 1184: 1181: 1179: 1176: 1174: 1171: 1169: 1166: 1164: 1163:Chinchilla AI 1161: 1159: 1156: 1154: 1151: 1149: 1146: 1144: 1141: 1139: 1136: 1134: 1131: 1129: 1126: 1124: 1121: 1119: 1116: 1114: 1111: 1109: 1106: 1102: 1099: 1098: 1097: 1094: 1092: 1089: 1087: 1084: 1082: 1079: 1077: 1074: 1072: 1069: 1068: 1066: 1062: 1056: 1053: 1049: 1046: 1044: 1041: 1040: 1039: 1036: 1032: 1029: 1027: 1024: 1022: 1019: 1018: 1017: 1014: 1012: 1009: 1007: 1004: 1002: 999: 997: 994: 992: 989: 987: 984: 982: 979: 977: 974: 972: 969: 968: 966: 962: 959: 955: 949: 946: 944: 941: 939: 936: 934: 931: 929: 926: 924: 921: 919: 916: 915: 913: 909: 903: 900: 898: 895: 893: 890: 888: 885: 883: 880: 879: 877: 873: 865: 862: 861: 860: 857: 855: 852: 850: 847: 843: 842:Deep learning 840: 839: 838: 835: 831: 828: 827: 826: 823: 822: 820: 816: 810: 807: 805: 802: 798: 795: 794: 793: 790: 788: 785: 781: 778: 776: 773: 771: 768: 767: 766: 763: 761: 758: 756: 753: 751: 748: 746: 743: 741: 738: 736: 733: 731: 730:Hallucination 728: 724: 721: 720: 719: 716: 714: 711: 707: 704: 703: 702: 699: 698: 696: 692: 686: 683: 681: 678: 676: 673: 671: 668: 666: 663: 661: 658: 656: 653: 651: 648: 646: 645: 641: 640: 638: 636: 632: 623: 618: 616: 611: 609: 604: 603: 600: 588: 580: 579: 576: 569: 566: 564: 561: 559: 556: 554: 551: 549: 546: 543: 542: 540: 538: 534: 528: 525: 523: 520: 518: 515: 513: 510: 508: 505: 503: 500: 498: 495: 493: 490: 488: 485: 483: 480: 478: 475: 473: 470: 468: 465: 463: 460: 458: 455: 453: 450: 448: 445: 444: 442: 440: 436: 430: 427: 426: 423: 419: 412: 407: 405: 400: 398: 393: 392: 389: 383: 379: 376: 372: 367: 366: 362: 354: 342: 338: 332: 329: 317: 313: 306: 303: 291: 287: 281: 278: 274: 263: 259: 252: 249: 242: 238: 235: 233: 230: 228: 225: 224: 220: 218: 216: 212: 208: 204: 200: 196: 195:deep learning 192: 189: 185: 181: 171: 164: 158: 154: 151: 148: 146: 142: 138: 135: 133: 129: 126: 122: 118: 115: 113: 109: 106: 102: 98: 95: 91: 87: 71: 69: 65: 61: 47: 43: 40: 37: 35: 31: 27: 22: 16: 1352:Hugging Face 1316:David Silver 964:Audio–visual 818:Applications 797:Augmentation 642: 452:Apache SINGA 447:Apache MXNet 351: 344:. Retrieved 340: 331: 319:. Retrieved 315: 305: 293:. Retrieved 289: 280: 272: 265:. Retrieved 261: 251: 211:Apache MXNet 179: 178: 34:Developer(s) 15: 1500:Categories 1448:Autoencoder 1403:Transformer 1271:Alex Graves 1219:OpenAI Five 1123:IBM Watsonx 745:Convolution 723:Overfitting 537:Proprietary 439:Open source 316:VentureBeat 188:open-source 1526:Categories 1489:Technology 1342:EleutherAI 1301:Fei-Fei Li 1296:Yann LeCun 1209:Q-learning 1192:Decisional 1118:IBM Watson 1026:Midjourney 918:TensorFlow 765:Activation 718:Regression 713:Clustering 553:IBM Watson 502:TensorFlow 429:Comparison 243:References 237:All-Reduce 199:TensorFlow 93:Written in 80:2023-06-12 55:2017-08-09 1372:MIT CSAIL 1337:Anthropic 1306:Andrew Ng 1204:AlphaZero 1048:VideoPoet 1011:AlphaFold 948:MindSpore 902:SpiNNaker 897:Memristor 804:Diffusion 780:Rectifier 760:Batchnorm 740:Attention 735:Adversary 527:MindSpore 467:DeepSpeed 139:ecosystem 1480:Portals 1239:Auto-GPT 1071:Word2vec 875:Hardware 792:Datasets 694:Concepts 587:Category 522:OpenVINO 295:July 11, 221:See also 112:Platform 1362:Meta AI 1199:AlphaGo 1183:PanGu-Σ 1153:ChatGPT 1128:Granite 1076:Seq2seq 1055:Whisper 976:WaveNet 971:AlexNet 943:Flux.jl 923:PyTorch 775:Sigmoid 770:Softmax 635:General 548:Core ML 497:PyTorch 378:horovod 353:NVIDIA. 346:July 9, 321:July 9, 290:horovod 267:July 9, 262:horovod 207:PyTorch 180:Horovod 161:horovod 156:Website 145:License 125:Windows 78: ( 53: ( 19:Horovod 1377:Huawei 1357:OpenAI 1259:People 1229:MuZero 1091:Gemini 1086:Claude 1021:DALL-E 933:Theano 568:MATLAB 507:Theano 492:OpenNN 487:ML.NET 382:GitHub 209:, and 166:  97:Python 1443:Mamba 1214:SARSA 1178:LLaMA 1173:BLOOM 1158:GPT-J 1148:GPT-4 1143:GPT-3 1138:GPT-2 1133:GPT-1 1096:LaMDA 928:Keras 545:Apple 512:Torch 477:Keras 457:Caffe 203:Keras 182:is a 121:macOS 117:Linux 1367:Mila 1168:PaLM 1101:Bard 1081:BERT 1064:Text 1043:Sora 517:ONNX 472:Dlib 348:2020 323:2020 297:2023 269:2020 186:and 184:free 132:Type 105:CUDA 39:Uber 1108:NMT 991:OCR 986:HWR 938:JAX 892:VPU 887:TPU 882:IPU 706:SGD 380:on 163:.ai 101:C++ 1528:: 350:. 339:. 314:. 288:. 271:. 260:. 205:, 201:, 123:, 119:, 103:, 99:, 621:e 614:t 607:v 410:e 403:t 396:v 325:. 299:. 82:) 57:)

Index


Developer(s)
Uber
Stable release
Python
C++
CUDA
Platform
Linux
macOS
Windows
Type
Artificial intelligence
License
Apache License 2.0
horovod.ai
Edit this on Wikidata
free
open-source
software framework
deep learning
TensorFlow
Keras
PyTorch
Apache MXNet
Linux Foundation AI
Comparison of deep learning software
Differentiable programming
All-Reduce
"Release v0.9.0 · horovod/horovod"

Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.