File size: 95,242 Bytes
81b7e35 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 |
2022-04-09 01:40:41,909 INFO [decode_test.py:583] Decoding started
2022-04-09 01:40:41,910 INFO [decode_test.py:584] {'subsampling_factor': 4, 'vgg_frontend': False, 'use_feat_batchnorm': True, 'feature_dim': 80, 'nhead': 8, 'attention_dim': 512, 'num_decoder_layers': 6, 'search_beam': 20, 'output_beam': 8, 'min_active_states': 30, 'max_active_states': 10000, 'use_double_scores': True, 'env_info': {'k2-version': '1.14', 'k2-build-type': 'Release', 'k2-with-cuda': True, 'k2-git-sha1': '6833270cb228aba7bf9681fccd41e2b52f7d984c', 'k2-git-date': 'Wed Mar 16 11:16:05 2022', 'lhotse-version': '1.0.0.dev+git.d917411.clean', 'torch-cuda-available': True, 'torch-cuda-version': '11.1', 'python-version': '3.7', 'icefall-git-branch': 'gigaspeech_recipe', 'icefall-git-sha1': 'c3993a5-dirty', 'icefall-git-date': 'Mon Mar 21 13:49:39 2022', 'icefall-path': '/userhome/user/guanbo/icefall_decode', 'k2-path': '/opt/conda/lib/python3.7/site-packages/k2-1.14.dev20220408+cuda11.1.torch1.10.0-py3.7-linux-x86_64.egg/k2/__init__.py', 'lhotse-path': '/userhome/user/guanbo/lhotse/lhotse/__init__.py', 'hostname': 'c8861f400b70d011ec0a3ee069db84328338-chenx8564-0', 'IP address': '10.9.150.55'}, 'epoch': 18, 'avg': 6, 'method': 'attention-decoder', 'num_paths': 1000, 'nbest_scale': 0.5, 'exp_dir': PosixPath('conformer_ctc/exp_500_8_2'), 'lang_dir': PosixPath('data/lang_bpe_500'), 'lm_dir': PosixPath('data/lm'), 'manifest_dir': PosixPath('data/fbank'), 'max_duration': 20, 'bucketing_sampler': True, 'num_buckets': 30, 'concatenate_cuts': False, 'duration_factor': 1.0, 'gap': 1.0, 'on_the_fly_feats': False, 'shuffle': True, 'return_cuts': True, 'num_workers': 1, 'enable_spec_aug': True, 'spec_aug_time_warp_factor': 80, 'enable_musan': True, 'subset': 'XL', 'lazy_load': True, 'small_dev': False}
2022-04-09 01:40:42,371 INFO [lexicon.py:176] Loading pre-compiled data/lang_bpe_500/Linv.pt
2022-04-09 01:40:42,473 INFO [decode_test.py:594] device: cuda:0
2022-04-09 01:40:46,249 INFO [decode_test.py:656] Loading pre-compiled G_4_gram.pt
2022-04-09 01:40:47,406 INFO [decode_test.py:692] averaging ['conformer_ctc/exp_500_8_2/epoch-13.pt', 'conformer_ctc/exp_500_8_2/epoch-14.pt', 'conformer_ctc/exp_500_8_2/epoch-15.pt', 'conformer_ctc/exp_500_8_2/epoch-16.pt', 'conformer_ctc/exp_500_8_2/epoch-17.pt', 'conformer_ctc/exp_500_8_2/epoch-18.pt']
2022-04-09 01:40:53,065 INFO [decode_test.py:699] Number of model parameters: 109226120
2022-04-09 01:40:53,065 INFO [asr_datamodule.py:381] About to get test cuts
2022-04-09 01:40:56,361 INFO [decode_test.py:497] batch 0/?, cuts processed until now is 3
2022-04-09 01:41:24,462 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 5.93 GiB (GPU 0; 31.75 GiB total capacity; 27.23 GiB already allocated; 1.90 GiB free; 28.49 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 01:41:24,462 INFO [decode.py:743] num_arcs before pruning: 324363
2022-04-09 01:41:24,462 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 01:41:24,473 INFO [decode.py:757] num_arcs after pruning: 7174
2022-04-09 01:41:40,284 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 4.67 GiB (GPU 0; 31.75 GiB total capacity; 25.69 GiB already allocated; 2.92 GiB free; 27.47 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 01:41:40,285 INFO [decode.py:743] num_arcs before pruning: 368362
2022-04-09 01:41:40,285 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 01:41:40,305 INFO [decode.py:757] num_arcs after pruning: 8521
2022-04-09 01:42:38,727 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 2.18 GiB (GPU 0; 31.75 GiB total capacity; 26.05 GiB already allocated; 1.42 GiB free; 28.98 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 01:42:38,727 INFO [decode.py:743] num_arcs before pruning: 432616
2022-04-09 01:42:38,728 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 01:42:38,736 INFO [decode.py:757] num_arcs after pruning: 9233
2022-04-09 01:43:13,573 INFO [decode_test.py:497] batch 100/?, cuts processed until now is 297
2022-04-09 01:43:48,362 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 25.34 GiB already allocated; 2.20 GiB free; 28.20 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 01:43:48,363 INFO [decode.py:743] num_arcs before pruning: 319907
2022-04-09 01:43:48,363 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 01:43:48,372 INFO [decode.py:757] num_arcs after pruning: 6358
2022-04-09 01:43:59,713 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 2.74 GiB (GPU 0; 31.75 GiB total capacity; 27.51 GiB already allocated; 2.19 GiB free; 28.20 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 01:43:59,713 INFO [decode.py:743] num_arcs before pruning: 313596
2022-04-09 01:43:59,713 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 01:43:59,724 INFO [decode.py:757] num_arcs after pruning: 8252
2022-04-09 01:44:54,463 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 25.25 GiB already allocated; 3.07 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 01:44:54,463 INFO [decode.py:743] num_arcs before pruning: 353355
2022-04-09 01:44:54,463 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 01:44:54,485 INFO [decode.py:757] num_arcs after pruning: 7520
2022-04-09 01:45:20,716 INFO [decode_test.py:497] batch 200/?, cuts processed until now is 570
2022-04-09 01:47:19,457 INFO [decode_test.py:497] batch 300/?, cuts processed until now is 806
2022-04-09 01:47:38,292 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 2.28 GiB (GPU 0; 31.75 GiB total capacity; 26.28 GiB already allocated; 1.48 GiB free; 28.92 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 01:47:38,293 INFO [decode.py:743] num_arcs before pruning: 596002
2022-04-09 01:47:38,293 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 01:47:38,312 INFO [decode.py:757] num_arcs after pruning: 10745
2022-04-09 01:49:18,493 INFO [decode.py:736] Caught exception:
Some bad things happened. Please read the above error messages and stack
trace. If you are using Python, the following command may be helpful:
gdb --args python /path/to/your/code.py
(You can use `gdb` to debug the code. Please consider compiling
a debug version of k2.).
If you are unable to fix it, please open an issue at:
https://github.com/k2-fsa/k2/issues/new
2022-04-09 01:49:18,494 INFO [decode.py:743] num_arcs before pruning: 398202
2022-04-09 01:49:18,494 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 01:49:18,541 INFO [decode.py:757] num_arcs after pruning: 14003
2022-04-09 01:49:21,800 INFO [decode_test.py:497] batch 400/?, cuts processed until now is 1082
2022-04-09 01:50:58,700 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 4.85 GiB (GPU 0; 31.75 GiB total capacity; 25.89 GiB already allocated; 1.48 GiB free; 28.92 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 01:50:58,701 INFO [decode.py:743] num_arcs before pruning: 398349
2022-04-09 01:50:58,701 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 01:50:58,709 INFO [decode.py:757] num_arcs after pruning: 10321
2022-04-09 01:51:31,627 INFO [decode_test.py:497] batch 500/?, cuts processed until now is 1334
2022-04-09 01:52:05,232 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.62 GiB already allocated; 1.47 GiB free; 28.93 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 01:52:05,232 INFO [decode.py:743] num_arcs before pruning: 212665
2022-04-09 01:52:05,232 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 01:52:05,241 INFO [decode.py:757] num_arcs after pruning: 6301
2022-04-09 01:53:29,890 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 1.91 GiB (GPU 0; 31.75 GiB total capacity; 25.66 GiB already allocated; 1.48 GiB free; 28.92 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 01:53:29,891 INFO [decode.py:743] num_arcs before pruning: 883555
2022-04-09 01:53:29,891 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 01:53:29,905 INFO [decode.py:757] num_arcs after pruning: 14819
2022-04-09 01:53:38,676 INFO [decode_test.py:497] batch 600/?, cuts processed until now is 1651
2022-04-09 01:54:57,438 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 25.34 GiB already allocated; 1.48 GiB free; 28.92 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 01:54:57,438 INFO [decode.py:743] num_arcs before pruning: 515795
2022-04-09 01:54:57,438 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 01:54:57,447 INFO [decode.py:757] num_arcs after pruning: 10132
2022-04-09 01:55:28,356 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.46 GiB already allocated; 1.48 GiB free; 28.92 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 01:55:28,356 INFO [decode.py:743] num_arcs before pruning: 670748
2022-04-09 01:55:28,356 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 01:55:28,365 INFO [decode.py:757] num_arcs after pruning: 10497
2022-04-09 01:55:42,238 INFO [decode_test.py:497] batch 700/?, cuts processed until now is 1956
2022-04-09 01:57:57,456 INFO [decode_test.py:497] batch 800/?, cuts processed until now is 2238
2022-04-09 01:58:04,281 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.45 GiB already allocated; 3.07 GiB free; 27.33 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 01:58:04,282 INFO [decode.py:743] num_arcs before pruning: 175423
2022-04-09 01:58:04,282 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 01:58:04,296 INFO [decode.py:757] num_arcs after pruning: 7926
2022-04-09 01:59:07,916 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 4.68 GiB (GPU 0; 31.75 GiB total capacity; 24.40 GiB already allocated; 3.06 GiB free; 27.33 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 01:59:07,917 INFO [decode.py:743] num_arcs before pruning: 259758
2022-04-09 01:59:07,917 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 01:59:07,928 INFO [decode.py:757] num_arcs after pruning: 6026
2022-04-09 02:00:00,623 INFO [decode_test.py:497] batch 900/?, cuts processed until now is 2536
2022-04-09 02:01:22,959 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.44 GiB already allocated; 3.08 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:01:22,959 INFO [decode.py:743] num_arcs before pruning: 749228
2022-04-09 02:01:22,959 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:01:22,968 INFO [decode.py:757] num_arcs after pruning: 23868
2022-04-09 02:01:59,449 INFO [decode_test.py:497] batch 1000/?, cuts processed until now is 2824
2022-04-09 02:03:05,494 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.38 GiB already allocated; 3.06 GiB free; 27.33 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:03:05,494 INFO [decode.py:743] num_arcs before pruning: 255135
2022-04-09 02:03:05,494 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:03:05,504 INFO [decode.py:757] num_arcs after pruning: 5955
2022-04-09 02:03:48,017 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.61 GiB already allocated; 3.08 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:03:48,017 INFO [decode.py:743] num_arcs before pruning: 517077
2022-04-09 02:03:48,017 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:03:48,026 INFO [decode.py:757] num_arcs after pruning: 7695
2022-04-09 02:04:09,806 INFO [decode_test.py:497] batch 1100/?, cuts processed until now is 3105
2022-04-09 02:04:31,410 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.34 GiB already allocated; 3.08 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:04:31,411 INFO [decode.py:743] num_arcs before pruning: 859561
2022-04-09 02:04:31,411 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:04:31,422 INFO [decode.py:757] num_arcs after pruning: 13014
2022-04-09 02:06:11,496 INFO [decode_test.py:497] batch 1200/?, cuts processed until now is 3401
2022-04-09 02:08:10,727 INFO [decode_test.py:497] batch 1300/?, cuts processed until now is 3730
2022-04-09 02:10:17,677 INFO [decode_test.py:497] batch 1400/?, cuts processed until now is 4067
2022-04-09 02:12:13,175 INFO [decode_test.py:497] batch 1500/?, cuts processed until now is 4329
2022-04-09 02:13:02,842 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.55 GiB already allocated; 3.08 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:13:02,843 INFO [decode.py:743] num_arcs before pruning: 475511
2022-04-09 02:13:02,843 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:13:02,849 INFO [decode.py:757] num_arcs after pruning: 8439
2022-04-09 02:13:46,588 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 2.37 GiB (GPU 0; 31.75 GiB total capacity; 26.83 GiB already allocated; 1.45 GiB free; 28.94 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:13:46,588 INFO [decode.py:743] num_arcs before pruning: 595488
2022-04-09 02:13:46,588 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:13:46,598 INFO [decode.py:757] num_arcs after pruning: 13475
2022-04-09 02:14:21,206 INFO [decode_test.py:497] batch 1600/?, cuts processed until now is 4598
2022-04-09 02:16:42,740 INFO [decode_test.py:497] batch 1700/?, cuts processed until now is 4969
2022-04-09 02:17:13,672 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 25.39 GiB already allocated; 1.45 GiB free; 28.94 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:17:13,673 INFO [decode.py:743] num_arcs before pruning: 615734
2022-04-09 02:17:13,673 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:17:13,685 INFO [decode.py:757] num_arcs after pruning: 8684
2022-04-09 02:18:54,514 INFO [decode_test.py:497] batch 1800/?, cuts processed until now is 5260
2022-04-09 02:18:59,938 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.36 GiB already allocated; 3.08 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:18:59,938 INFO [decode.py:743] num_arcs before pruning: 360099
2022-04-09 02:18:59,938 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:18:59,949 INFO [decode.py:757] num_arcs after pruning: 6898
2022-04-09 02:19:48,186 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 6.00 GiB (GPU 0; 31.75 GiB total capacity; 27.15 GiB already allocated; 967.75 MiB free; 29.45 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:19:48,186 INFO [decode.py:743] num_arcs before pruning: 168720
2022-04-09 02:19:48,186 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:19:48,201 INFO [decode.py:757] num_arcs after pruning: 5346
2022-04-09 02:20:52,049 INFO [decode_test.py:497] batch 1900/?, cuts processed until now is 5585
2022-04-09 02:22:12,107 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.45 GiB already allocated; 973.75 MiB free; 29.44 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:22:12,107 INFO [decode.py:743] num_arcs before pruning: 1151735
2022-04-09 02:22:12,107 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:22:12,120 INFO [decode.py:757] num_arcs after pruning: 8335
2022-04-09 02:23:01,497 INFO [decode_test.py:497] batch 2000/?, cuts processed until now is 5902
2022-04-09 02:25:26,356 INFO [decode_test.py:497] batch 2100/?, cuts processed until now is 6219
2022-04-09 02:25:56,466 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.34 GiB already allocated; 973.75 MiB free; 29.44 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:25:56,467 INFO [decode.py:743] num_arcs before pruning: 612804
2022-04-09 02:25:56,467 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:25:56,477 INFO [decode.py:757] num_arcs after pruning: 10853
2022-04-09 02:27:26,441 INFO [decode_test.py:497] batch 2200/?, cuts processed until now is 6480
2022-04-09 02:29:28,073 INFO [decode_test.py:497] batch 2300/?, cuts processed until now is 6768
2022-04-09 02:31:41,553 INFO [decode_test.py:497] batch 2400/?, cuts processed until now is 7120
2022-04-09 02:31:55,632 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.42 GiB already allocated; 3.07 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:31:55,632 INFO [decode.py:743] num_arcs before pruning: 411490
2022-04-09 02:31:55,632 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:31:55,638 INFO [decode.py:757] num_arcs after pruning: 8626
2022-04-09 02:33:22,034 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.42 GiB already allocated; 3.07 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:33:22,034 INFO [decode.py:743] num_arcs before pruning: 625728
2022-04-09 02:33:22,035 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:33:22,043 INFO [decode.py:757] num_arcs after pruning: 9502
2022-04-09 02:33:37,663 INFO [decode_test.py:497] batch 2500/?, cuts processed until now is 7387
2022-04-09 02:34:18,300 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.51 GiB already allocated; 3.07 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:34:18,301 INFO [decode.py:743] num_arcs before pruning: 1015956
2022-04-09 02:34:18,301 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:34:18,314 INFO [decode.py:757] num_arcs after pruning: 14404
2022-04-09 02:34:20,220 INFO [decode.py:841] Caught exception:
CUDA out of memory. Tried to allocate 5.58 GiB (GPU 0; 31.75 GiB total capacity; 24.87 GiB already allocated; 3.07 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:34:20,221 INFO [decode.py:843] num_paths before decreasing: 1000
2022-04-09 02:34:20,221 INFO [decode.py:852] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:34:20,221 INFO [decode.py:858] num_paths after decreasing: 500
2022-04-09 02:34:40,089 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.38 GiB already allocated; 3.07 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:34:40,089 INFO [decode.py:743] num_arcs before pruning: 570686
2022-04-09 02:34:40,089 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:34:40,098 INFO [decode.py:757] num_arcs after pruning: 9182
2022-04-09 02:35:50,624 INFO [decode_test.py:497] batch 2600/?, cuts processed until now is 7764
2022-04-09 02:36:44,519 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.61 GiB already allocated; 3.08 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:36:44,519 INFO [decode.py:743] num_arcs before pruning: 1066267
2022-04-09 02:36:44,519 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:36:44,530 INFO [decode.py:757] num_arcs after pruning: 6963
2022-04-09 02:38:18,717 INFO [decode_test.py:497] batch 2700/?, cuts processed until now is 8078
2022-04-09 02:40:07,021 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.42 GiB already allocated; 3.07 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:40:07,022 INFO [decode.py:743] num_arcs before pruning: 1023667
2022-04-09 02:40:07,022 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:40:07,034 INFO [decode.py:757] num_arcs after pruning: 13090
2022-04-09 02:40:25,184 INFO [decode_test.py:497] batch 2800/?, cuts processed until now is 8444
2022-04-09 02:41:27,080 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.32 GiB already allocated; 3.08 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:41:27,080 INFO [decode.py:743] num_arcs before pruning: 739744
2022-04-09 02:41:27,080 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:41:27,093 INFO [decode.py:757] num_arcs after pruning: 9791
2022-04-09 02:42:44,319 INFO [decode_test.py:497] batch 2900/?, cuts processed until now is 8765
2022-04-09 02:42:44,656 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.73 GiB already allocated; 3.08 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:42:44,656 INFO [decode.py:743] num_arcs before pruning: 666168
2022-04-09 02:42:44,656 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:42:44,665 INFO [decode.py:757] num_arcs after pruning: 17223
2022-04-09 02:43:05,748 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 5.60 GiB (GPU 0; 31.75 GiB total capacity; 26.18 GiB already allocated; 1.14 GiB free; 29.26 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:43:05,748 INFO [decode.py:743] num_arcs before pruning: 188729
2022-04-09 02:43:05,748 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:43:05,762 INFO [decode.py:757] num_arcs after pruning: 8688
2022-04-09 02:44:54,469 INFO [decode_test.py:497] batch 3000/?, cuts processed until now is 9050
2022-04-09 02:46:55,167 INFO [decode_test.py:497] batch 3100/?, cuts processed until now is 9296
2022-04-09 02:47:28,418 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 20.00 GiB already allocated; 3.07 GiB free; 27.33 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:47:28,419 INFO [decode.py:743] num_arcs before pruning: 160153
2022-04-09 02:47:28,419 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:47:28,448 INFO [decode.py:757] num_arcs after pruning: 7778
2022-04-09 02:49:21,448 INFO [decode_test.py:497] batch 3200/?, cuts processed until now is 9652
2022-04-09 02:50:17,558 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 6.13 GiB (GPU 0; 31.75 GiB total capacity; 27.60 GiB already allocated; 895.75 MiB free; 29.52 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:50:17,558 INFO [decode.py:743] num_arcs before pruning: 388116
2022-04-09 02:50:17,559 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:50:17,565 INFO [decode.py:757] num_arcs after pruning: 10555
2022-04-09 02:51:30,675 INFO [decode_test.py:497] batch 3300/?, cuts processed until now is 10071
2022-04-09 02:53:49,565 INFO [decode_test.py:497] batch 3400/?, cuts processed until now is 10342
2022-04-09 02:55:49,392 INFO [decode_test.py:497] batch 3500/?, cuts processed until now is 10642
2022-04-09 02:58:07,518 INFO [decode_test.py:497] batch 3600/?, cuts processed until now is 10951
2022-04-09 02:58:16,360 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.29 GiB already allocated; 3.07 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 02:58:16,361 INFO [decode.py:743] num_arcs before pruning: 396714
2022-04-09 02:58:16,361 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 02:58:16,374 INFO [decode.py:757] num_arcs after pruning: 9543
2022-04-09 03:00:00,485 INFO [decode_test.py:497] batch 3700/?, cuts processed until now is 11231
2022-04-09 03:00:17,600 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.45 GiB already allocated; 3.07 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:00:17,601 INFO [decode.py:743] num_arcs before pruning: 854366
2022-04-09 03:00:17,601 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:00:17,612 INFO [decode.py:757] num_arcs after pruning: 10487
2022-04-09 03:00:20,098 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.68 GiB already allocated; 3.07 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:00:20,098 INFO [decode.py:743] num_arcs before pruning: 442824
2022-04-09 03:00:20,098 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:00:20,108 INFO [decode.py:757] num_arcs after pruning: 5265
2022-04-09 03:02:00,114 INFO [decode_test.py:497] batch 3800/?, cuts processed until now is 11509
2022-04-09 03:02:11,570 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.19 GiB already allocated; 3.07 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:02:11,571 INFO [decode.py:743] num_arcs before pruning: 285638
2022-04-09 03:02:11,571 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:02:11,579 INFO [decode.py:757] num_arcs after pruning: 5903
2022-04-09 03:04:02,757 INFO [decode_test.py:497] batch 3900/?, cuts processed until now is 11774
2022-04-09 03:05:19,989 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.73 GiB already allocated; 3.08 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:05:19,990 INFO [decode.py:743] num_arcs before pruning: 637327
2022-04-09 03:05:19,990 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:05:19,999 INFO [decode.py:757] num_arcs after pruning: 6357
2022-04-09 03:06:01,953 INFO [decode_test.py:497] batch 4000/?, cuts processed until now is 12045
2022-04-09 03:07:49,854 INFO [decode_test.py:497] batch 4100/?, cuts processed until now is 12300
2022-04-09 03:09:15,137 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.45 GiB already allocated; 3.08 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:09:15,138 INFO [decode.py:743] num_arcs before pruning: 507733
2022-04-09 03:09:15,138 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:09:15,148 INFO [decode.py:757] num_arcs after pruning: 4196
2022-04-09 03:09:47,397 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 5.86 GiB (GPU 0; 31.75 GiB total capacity; 27.78 GiB already allocated; 925.75 MiB free; 29.49 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:09:47,397 INFO [decode.py:743] num_arcs before pruning: 514118
2022-04-09 03:09:47,397 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:09:47,407 INFO [decode.py:757] num_arcs after pruning: 7168
2022-04-09 03:10:00,013 INFO [decode_test.py:497] batch 4200/?, cuts processed until now is 12580
2022-04-09 03:10:33,411 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 2.80 GiB (GPU 0; 31.75 GiB total capacity; 27.70 GiB already allocated; 925.75 MiB free; 29.49 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:10:33,411 INFO [decode.py:743] num_arcs before pruning: 374935
2022-04-09 03:10:33,411 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:10:33,418 INFO [decode.py:757] num_arcs after pruning: 10023
2022-04-09 03:12:04,333 INFO [decode_test.py:497] batch 4300/?, cuts processed until now is 12807
2022-04-09 03:14:06,889 INFO [decode_test.py:497] batch 4400/?, cuts processed until now is 13050
2022-04-09 03:14:34,787 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.47 GiB already allocated; 925.75 MiB free; 29.49 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:14:34,788 INFO [decode.py:743] num_arcs before pruning: 767465
2022-04-09 03:14:34,788 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:14:34,797 INFO [decode.py:757] num_arcs after pruning: 19151
2022-04-09 03:15:08,864 INFO [decode.py:736] Caught exception:
Some bad things happened. Please read the above error messages and stack
trace. If you are using Python, the following command may be helpful:
gdb --args python /path/to/your/code.py
(You can use `gdb` to debug the code. Please consider compiling
a debug version of k2.).
If you are unable to fix it, please open an issue at:
https://github.com/k2-fsa/k2/issues/new
2022-04-09 03:15:08,864 INFO [decode.py:743] num_arcs before pruning: 123833
2022-04-09 03:15:08,864 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:15:08,913 INFO [decode.py:757] num_arcs after pruning: 4150
2022-04-09 03:15:34,899 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 25.64 GiB already allocated; 3.07 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:15:34,899 INFO [decode.py:743] num_arcs before pruning: 444800
2022-04-09 03:15:34,899 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:15:34,908 INFO [decode.py:757] num_arcs after pruning: 11839
2022-04-09 03:16:08,462 INFO [decode_test.py:497] batch 4500/?, cuts processed until now is 13295
2022-04-09 03:17:56,946 INFO [decode_test.py:497] batch 4600/?, cuts processed until now is 13593
2022-04-09 03:18:16,099 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 5.53 GiB (GPU 0; 31.75 GiB total capacity; 26.53 GiB already allocated; 1.12 GiB free; 29.28 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:18:16,099 INFO [decode.py:743] num_arcs before pruning: 350609
2022-04-09 03:18:16,100 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:18:16,105 INFO [decode.py:757] num_arcs after pruning: 9262
2022-04-09 03:19:57,230 INFO [decode_test.py:497] batch 4700/?, cuts processed until now is 13858
2022-04-09 03:20:19,775 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 4.87 GiB (GPU 0; 31.75 GiB total capacity; 25.78 GiB already allocated; 1.12 GiB free; 29.28 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:20:19,775 INFO [decode.py:743] num_arcs before pruning: 375071
2022-04-09 03:20:19,775 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:20:19,785 INFO [decode.py:757] num_arcs after pruning: 6365
2022-04-09 03:21:29,481 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.42 GiB already allocated; 1.12 GiB free; 29.27 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:21:29,481 INFO [decode.py:743] num_arcs before pruning: 872088
2022-04-09 03:21:29,481 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:21:29,492 INFO [decode.py:757] num_arcs after pruning: 10043
2022-04-09 03:22:01,760 INFO [decode_test.py:497] batch 4800/?, cuts processed until now is 14079
2022-04-09 03:24:10,370 INFO [decode_test.py:497] batch 4900/?, cuts processed until now is 14298
2022-04-09 03:26:10,811 INFO [decode_test.py:497] batch 5000/?, cuts processed until now is 14515
2022-04-09 03:27:46,191 INFO [decode.py:736] Caught exception:
Some bad things happened. Please read the above error messages and stack
trace. If you are using Python, the following command may be helpful:
gdb --args python /path/to/your/code.py
(You can use `gdb` to debug the code. Please consider compiling
a debug version of k2.).
If you are unable to fix it, please open an issue at:
https://github.com/k2-fsa/k2/issues/new
2022-04-09 03:27:46,192 INFO [decode.py:743] num_arcs before pruning: 246382
2022-04-09 03:27:46,192 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:27:46,253 INFO [decode.py:757] num_arcs after pruning: 6775
2022-04-09 03:28:15,199 INFO [decode_test.py:497] batch 5100/?, cuts processed until now is 14718
2022-04-09 03:29:19,807 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 6.15 GiB (GPU 0; 31.75 GiB total capacity; 26.67 GiB already allocated; 1.11 GiB free; 29.29 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:29:19,808 INFO [decode.py:743] num_arcs before pruning: 220820
2022-04-09 03:29:19,808 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:29:19,815 INFO [decode.py:757] num_arcs after pruning: 13482
2022-04-09 03:30:16,045 INFO [decode_test.py:497] batch 5200/?, cuts processed until now is 14930
2022-04-09 03:32:12,235 INFO [decode_test.py:497] batch 5300/?, cuts processed until now is 15128
2022-04-09 03:33:06,358 INFO [decode.py:736] Caught exception:
Some bad things happened. Please read the above error messages and stack
trace. If you are using Python, the following command may be helpful:
gdb --args python /path/to/your/code.py
(You can use `gdb` to debug the code. Please consider compiling
a debug version of k2.).
If you are unable to fix it, please open an issue at:
https://github.com/k2-fsa/k2/issues/new
2022-04-09 03:33:06,359 INFO [decode.py:743] num_arcs before pruning: 190203
2022-04-09 03:33:06,359 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:33:06,413 INFO [decode.py:757] num_arcs after pruning: 6202
2022-04-09 03:34:14,862 INFO [decode_test.py:497] batch 5400/?, cuts processed until now is 15327
2022-04-09 03:36:18,973 INFO [decode_test.py:497] batch 5500/?, cuts processed until now is 15531
2022-04-09 03:38:18,633 INFO [decode_test.py:497] batch 5600/?, cuts processed until now is 15724
2022-04-09 03:38:48,490 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.52 GiB already allocated; 3.07 GiB free; 27.32 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:38:48,491 INFO [decode.py:743] num_arcs before pruning: 554330
2022-04-09 03:38:48,491 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:38:48,500 INFO [decode.py:757] num_arcs after pruning: 10730
2022-04-09 03:39:51,281 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 4.83 GiB (GPU 0; 31.75 GiB total capacity; 25.96 GiB already allocated; 1.31 GiB free; 29.08 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:39:51,281 INFO [decode.py:743] num_arcs before pruning: 160031
2022-04-09 03:39:51,281 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:39:51,288 INFO [decode.py:757] num_arcs after pruning: 4270
2022-04-09 03:40:28,016 INFO [decode_test.py:497] batch 5700/?, cuts processed until now is 15908
2022-04-09 03:40:46,608 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 2.58 GiB (GPU 0; 31.75 GiB total capacity; 27.28 GiB already allocated; 1.32 GiB free; 29.07 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:40:46,608 INFO [decode.py:743] num_arcs before pruning: 406026
2022-04-09 03:40:46,608 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:40:46,616 INFO [decode.py:757] num_arcs after pruning: 11179
2022-04-09 03:42:16,464 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 2.29 GiB (GPU 0; 31.75 GiB total capacity; 26.71 GiB already allocated; 1.32 GiB free; 29.07 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:42:16,464 INFO [decode.py:743] num_arcs before pruning: 639824
2022-04-09 03:42:16,464 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:42:16,476 INFO [decode.py:757] num_arcs after pruning: 5520
2022-04-09 03:42:52,683 INFO [decode_test.py:497] batch 5800/?, cuts processed until now is 16094
2022-04-09 03:44:51,754 INFO [decode_test.py:497] batch 5900/?, cuts processed until now is 16289
2022-04-09 03:46:52,121 INFO [decode_test.py:497] batch 6000/?, cuts processed until now is 16488
2022-04-09 03:48:54,739 INFO [decode_test.py:497] batch 6100/?, cuts processed until now is 16661
2022-04-09 03:49:24,829 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 1.84 GiB (GPU 0; 31.75 GiB total capacity; 28.87 GiB already allocated; 409.75 MiB free; 29.99 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:49:24,830 INFO [decode.py:743] num_arcs before pruning: 443401
2022-04-09 03:49:24,830 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:49:24,837 INFO [decode.py:757] num_arcs after pruning: 5211
2022-04-09 03:50:27,492 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.35 GiB already allocated; 2.15 GiB free; 28.24 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:50:27,493 INFO [decode.py:743] num_arcs before pruning: 361598
2022-04-09 03:50:27,493 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:50:27,507 INFO [decode.py:757] num_arcs after pruning: 8660
2022-04-09 03:51:02,856 INFO [decode_test.py:497] batch 6200/?, cuts processed until now is 16828
2022-04-09 03:53:03,912 INFO [decode_test.py:497] batch 6300/?, cuts processed until now is 17002
2022-04-09 03:55:04,964 INFO [decode_test.py:497] batch 6400/?, cuts processed until now is 17181
2022-04-09 03:55:08,345 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 4.89 GiB (GPU 0; 31.75 GiB total capacity; 26.28 GiB already allocated; 2.16 GiB free; 28.24 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:55:08,345 INFO [decode.py:743] num_arcs before pruning: 867262
2022-04-09 03:55:08,345 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:55:08,356 INFO [decode.py:757] num_arcs after pruning: 6494
2022-04-09 03:56:03,884 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 1.90 GiB (GPU 0; 31.75 GiB total capacity; 28.97 GiB already allocated; 1.16 GiB free; 29.23 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:56:03,885 INFO [decode.py:743] num_arcs before pruning: 233755
2022-04-09 03:56:03,885 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:56:03,910 INFO [decode.py:757] num_arcs after pruning: 5823
2022-04-09 03:57:08,774 INFO [decode_test.py:497] batch 6500/?, cuts processed until now is 17347
2022-04-09 03:59:01,245 INFO [decode_test.py:497] batch 6600/?, cuts processed until now is 17502
2022-04-09 03:59:13,147 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 5.80 GiB (GPU 0; 31.75 GiB total capacity; 26.73 GiB already allocated; 1.17 GiB free; 29.22 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 03:59:13,147 INFO [decode.py:743] num_arcs before pruning: 174004
2022-04-09 03:59:13,147 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 03:59:13,155 INFO [decode.py:757] num_arcs after pruning: 6857
2022-04-09 04:00:59,687 INFO [decode_test.py:497] batch 6700/?, cuts processed until now is 17661
2022-04-09 04:03:01,660 INFO [decode_test.py:497] batch 6800/?, cuts processed until now is 17823
2022-04-09 04:04:55,219 INFO [decode_test.py:497] batch 6900/?, cuts processed until now is 17997
2022-04-09 04:07:05,841 INFO [decode_test.py:497] batch 7000/?, cuts processed until now is 18159
2022-04-09 04:09:04,994 INFO [decode_test.py:497] batch 7100/?, cuts processed until now is 18299
2022-04-09 04:11:07,439 INFO [decode_test.py:497] batch 7200/?, cuts processed until now is 18432
2022-04-09 04:13:18,126 INFO [decode_test.py:497] batch 7300/?, cuts processed until now is 18552
2022-04-09 04:15:23,102 INFO [decode_test.py:497] batch 7400/?, cuts processed until now is 18656
2022-04-09 04:17:49,550 INFO [decode_test.py:497] batch 7500/?, cuts processed until now is 18798
2022-04-09 04:19:16,128 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.34 GiB already allocated; 2.12 GiB free; 28.27 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 04:19:16,129 INFO [decode.py:743] num_arcs before pruning: 1155990
2022-04-09 04:19:16,129 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 04:19:16,143 INFO [decode.py:757] num_arcs after pruning: 9141
2022-04-09 04:20:19,961 INFO [decode_test.py:497] batch 7600/?, cuts processed until now is 18945
2022-04-09 04:22:44,642 INFO [decode_test.py:497] batch 7700/?, cuts processed until now is 19084
2022-04-09 04:23:18,184 INFO [decode.py:841] Caught exception:
CUDA out of memory. Tried to allocate 1.26 GiB (GPU 0; 31.75 GiB total capacity; 27.36 GiB already allocated; 881.75 MiB free; 29.53 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 04:23:18,184 INFO [decode.py:843] num_paths before decreasing: 1000
2022-04-09 04:23:18,184 INFO [decode.py:852] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 04:23:18,184 INFO [decode.py:858] num_paths after decreasing: 500
2022-04-09 04:24:52,959 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.53 GiB already allocated; 2.12 GiB free; 28.27 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 04:24:52,960 INFO [decode.py:743] num_arcs before pruning: 624026
2022-04-09 04:24:52,960 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 04:24:52,972 INFO [decode.py:757] num_arcs after pruning: 10008
2022-04-09 04:25:07,718 INFO [decode_test.py:497] batch 7800/?, cuts processed until now is 19232
2022-04-09 04:25:31,876 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.51 GiB already allocated; 2.12 GiB free; 28.27 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 04:25:31,876 INFO [decode.py:743] num_arcs before pruning: 688909
2022-04-09 04:25:31,877 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 04:25:31,887 INFO [decode.py:757] num_arcs after pruning: 8886
2022-04-09 04:25:57,970 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 5.04 GiB (GPU 0; 31.75 GiB total capacity; 25.95 GiB already allocated; 2.12 GiB free; 28.27 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 04:25:57,971 INFO [decode.py:743] num_arcs before pruning: 891176
2022-04-09 04:25:57,971 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 04:25:57,982 INFO [decode.py:757] num_arcs after pruning: 10106
2022-04-09 04:26:19,609 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 2.63 GiB (GPU 0; 31.75 GiB total capacity; 27.60 GiB already allocated; 327.75 MiB free; 30.07 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 04:26:19,609 INFO [decode.py:743] num_arcs before pruning: 415376
2022-04-09 04:26:19,609 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 04:26:19,620 INFO [decode.py:757] num_arcs after pruning: 7771
2022-04-09 04:27:33,059 INFO [decode_test.py:497] batch 7900/?, cuts processed until now is 19375
2022-04-09 04:29:43,649 INFO [decode_test.py:497] batch 8000/?, cuts processed until now is 19510
2022-04-09 04:30:20,590 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.65 GiB already allocated; 2.12 GiB free; 28.27 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 04:30:20,591 INFO [decode.py:743] num_arcs before pruning: 330767
2022-04-09 04:30:20,591 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 04:30:20,606 INFO [decode.py:757] num_arcs after pruning: 5820
2022-04-09 04:31:55,818 INFO [decode_test.py:497] batch 8100/?, cuts processed until now is 19643
2022-04-09 04:34:11,720 INFO [decode_test.py:497] batch 8200/?, cuts processed until now is 19776
2022-04-09 04:35:04,147 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 4.49 GiB (GPU 0; 31.75 GiB total capacity; 24.38 GiB already allocated; 2.12 GiB free; 28.27 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 04:35:04,147 INFO [decode.py:743] num_arcs before pruning: 533967
2022-04-09 04:35:04,147 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 04:35:04,157 INFO [decode.py:757] num_arcs after pruning: 3449
2022-04-09 04:36:15,595 INFO [decode.py:736] Caught exception:
CUDA out of memory. Tried to allocate 8.00 GiB (GPU 0; 31.75 GiB total capacity; 19.67 GiB already allocated; 2.12 GiB free; 28.27 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF
2022-04-09 04:36:15,595 INFO [decode.py:743] num_arcs before pruning: 397138
2022-04-09 04:36:15,596 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 04:36:15,605 INFO [decode.py:757] num_arcs after pruning: 6775
2022-04-09 04:36:31,844 INFO [decode_test.py:497] batch 8300/?, cuts processed until now is 19882
2022-04-09 04:37:04,130 INFO [decode.py:736] Caught exception:
Some bad things happened. Please read the above error messages and stack
trace. If you are using Python, the following command may be helpful:
gdb --args python /path/to/your/code.py
(You can use `gdb` to debug the code. Please consider compiling
a debug version of k2.).
If you are unable to fix it, please open an issue at:
https://github.com/k2-fsa/k2/issues/new
2022-04-09 04:37:04,130 INFO [decode.py:743] num_arcs before pruning: 456591
2022-04-09 04:37:04,130 INFO [decode.py:746] This OOM is not an error. You can ignore it. If your model does not converge well, or --max-duration is too large, or the input sound file is difficult to decode, you will meet this exception.
2022-04-09 04:37:04,180 INFO [decode.py:757] num_arcs after pruning: 5275
2022-04-09 04:57:33,432 INFO [decode_test.py:567]
For test, WER of different settings are:
ngram_lm_scale_0.3_attention_scale_0.7 10.58 best for test
ngram_lm_scale_0.5_attention_scale_1.3 10.58
ngram_lm_scale_0.3_attention_scale_0.5 10.59
ngram_lm_scale_0.3_attention_scale_0.6 10.59
ngram_lm_scale_0.3_attention_scale_0.9 10.59
ngram_lm_scale_0.3_attention_scale_1.0 10.59
ngram_lm_scale_0.3_attention_scale_1.1 10.59
ngram_lm_scale_0.3_attention_scale_1.2 10.59
ngram_lm_scale_0.3_attention_scale_1.3 10.59
ngram_lm_scale_0.5_attention_scale_1.0 10.59
ngram_lm_scale_0.5_attention_scale_1.1 10.59
ngram_lm_scale_0.5_attention_scale_1.2 10.59
ngram_lm_scale_0.5_attention_scale_1.5 10.59
ngram_lm_scale_0.5_attention_scale_1.7 10.59
ngram_lm_scale_0.5_attention_scale_1.9 10.59
ngram_lm_scale_0.5_attention_scale_2.0 10.59
ngram_lm_scale_0.5_attention_scale_2.1 10.59
ngram_lm_scale_0.5_attention_scale_2.2 10.59
ngram_lm_scale_0.5_attention_scale_2.3 10.59
ngram_lm_scale_0.6_attention_scale_1.9 10.59
ngram_lm_scale_0.6_attention_scale_2.0 10.59
ngram_lm_scale_0.6_attention_scale_2.1 10.59
ngram_lm_scale_0.6_attention_scale_2.2 10.59
ngram_lm_scale_0.6_attention_scale_2.3 10.59
ngram_lm_scale_0.6_attention_scale_2.5 10.59
ngram_lm_scale_0.3_attention_scale_1.5 10.6
ngram_lm_scale_0.3_attention_scale_1.7 10.6
ngram_lm_scale_0.3_attention_scale_1.9 10.6
ngram_lm_scale_0.3_attention_scale_2.0 10.6
ngram_lm_scale_0.3_attention_scale_2.1 10.6
ngram_lm_scale_0.3_attention_scale_2.2 10.6
ngram_lm_scale_0.3_attention_scale_2.3 10.6
ngram_lm_scale_0.3_attention_scale_2.5 10.6
ngram_lm_scale_0.5_attention_scale_0.9 10.6
ngram_lm_scale_0.5_attention_scale_2.5 10.6
ngram_lm_scale_0.5_attention_scale_3.0 10.6
ngram_lm_scale_0.6_attention_scale_1.3 10.6
ngram_lm_scale_0.6_attention_scale_1.5 10.6
ngram_lm_scale_0.6_attention_scale_1.7 10.6
ngram_lm_scale_0.6_attention_scale_3.0 10.6
ngram_lm_scale_0.3_attention_scale_0.3 10.61
ngram_lm_scale_0.3_attention_scale_3.0 10.61
ngram_lm_scale_0.5_attention_scale_4.0 10.61
ngram_lm_scale_0.5_attention_scale_5.0 10.61
ngram_lm_scale_0.6_attention_scale_1.2 10.61
ngram_lm_scale_0.6_attention_scale_4.0 10.61
ngram_lm_scale_0.6_attention_scale_5.0 10.61
ngram_lm_scale_0.7_attention_scale_1.7 10.61
ngram_lm_scale_0.7_attention_scale_1.9 10.61
ngram_lm_scale_0.7_attention_scale_2.0 10.61
ngram_lm_scale_0.7_attention_scale_2.1 10.61
ngram_lm_scale_0.7_attention_scale_2.2 10.61
ngram_lm_scale_0.7_attention_scale_2.3 10.61
ngram_lm_scale_0.7_attention_scale_2.5 10.61
ngram_lm_scale_0.7_attention_scale_3.0 10.61
ngram_lm_scale_0.7_attention_scale_4.0 10.61
ngram_lm_scale_0.7_attention_scale_5.0 10.61
ngram_lm_scale_0.1_attention_scale_1.1 10.62
ngram_lm_scale_0.3_attention_scale_4.0 10.62
ngram_lm_scale_0.3_attention_scale_5.0 10.62
ngram_lm_scale_0.5_attention_scale_0.7 10.62
ngram_lm_scale_0.6_attention_scale_1.0 10.62
ngram_lm_scale_0.6_attention_scale_1.1 10.62
ngram_lm_scale_0.7_attention_scale_1.5 10.62
ngram_lm_scale_0.9_attention_scale_3.0 10.62
ngram_lm_scale_0.9_attention_scale_4.0 10.62
ngram_lm_scale_0.9_attention_scale_5.0 10.62
ngram_lm_scale_1.0_attention_scale_4.0 10.62
ngram_lm_scale_1.1_attention_scale_5.0 10.62
ngram_lm_scale_0.05_attention_scale_1.1 10.63
ngram_lm_scale_0.05_attention_scale_1.2 10.63
ngram_lm_scale_0.08_attention_scale_0.9 10.63
ngram_lm_scale_0.08_attention_scale_1.0 10.63
ngram_lm_scale_0.08_attention_scale_1.1 10.63
ngram_lm_scale_0.08_attention_scale_1.2 10.63
ngram_lm_scale_0.08_attention_scale_1.3 10.63
ngram_lm_scale_0.08_attention_scale_1.9 10.63
ngram_lm_scale_0.08_attention_scale_2.0 10.63
ngram_lm_scale_0.08_attention_scale_2.1 10.63
ngram_lm_scale_0.08_attention_scale_2.2 10.63
ngram_lm_scale_0.08_attention_scale_2.3 10.63
ngram_lm_scale_0.08_attention_scale_3.0 10.63
ngram_lm_scale_0.1_attention_scale_0.5 10.63
ngram_lm_scale_0.1_attention_scale_0.6 10.63
ngram_lm_scale_0.1_attention_scale_0.7 10.63
ngram_lm_scale_0.1_attention_scale_0.9 10.63
ngram_lm_scale_0.1_attention_scale_1.0 10.63
ngram_lm_scale_0.1_attention_scale_1.2 10.63
ngram_lm_scale_0.1_attention_scale_1.3 10.63
ngram_lm_scale_0.1_attention_scale_1.7 10.63
ngram_lm_scale_0.1_attention_scale_1.9 10.63
ngram_lm_scale_0.1_attention_scale_2.0 10.63
ngram_lm_scale_0.1_attention_scale_2.1 10.63
ngram_lm_scale_0.1_attention_scale_2.2 10.63
ngram_lm_scale_0.1_attention_scale_2.3 10.63
ngram_lm_scale_0.1_attention_scale_2.5 10.63
ngram_lm_scale_0.1_attention_scale_3.0 10.63
ngram_lm_scale_0.1_attention_scale_5.0 10.63
ngram_lm_scale_0.5_attention_scale_0.6 10.63
ngram_lm_scale_0.6_attention_scale_0.9 10.63
ngram_lm_scale_0.9_attention_scale_2.3 10.63
ngram_lm_scale_0.9_attention_scale_2.5 10.63
ngram_lm_scale_1.0_attention_scale_5.0 10.63
ngram_lm_scale_1.2_attention_scale_5.0 10.63
ngram_lm_scale_0.01_attention_scale_0.9 10.64
ngram_lm_scale_0.01_attention_scale_1.0 10.64
ngram_lm_scale_0.01_attention_scale_1.1 10.64
ngram_lm_scale_0.01_attention_scale_1.2 10.64
ngram_lm_scale_0.01_attention_scale_4.0 10.64
ngram_lm_scale_0.01_attention_scale_5.0 10.64
ngram_lm_scale_0.05_attention_scale_0.5 10.64
ngram_lm_scale_0.05_attention_scale_0.6 10.64
ngram_lm_scale_0.05_attention_scale_0.7 10.64
ngram_lm_scale_0.05_attention_scale_0.9 10.64
ngram_lm_scale_0.05_attention_scale_1.0 10.64
ngram_lm_scale_0.05_attention_scale_1.3 10.64
ngram_lm_scale_0.05_attention_scale_1.5 10.64
ngram_lm_scale_0.05_attention_scale_1.7 10.64
ngram_lm_scale_0.05_attention_scale_1.9 10.64
ngram_lm_scale_0.05_attention_scale_2.0 10.64
ngram_lm_scale_0.05_attention_scale_2.1 10.64
ngram_lm_scale_0.05_attention_scale_2.2 10.64
ngram_lm_scale_0.05_attention_scale_2.3 10.64
ngram_lm_scale_0.05_attention_scale_2.5 10.64
ngram_lm_scale_0.05_attention_scale_3.0 10.64
ngram_lm_scale_0.05_attention_scale_4.0 10.64
ngram_lm_scale_0.05_attention_scale_5.0 10.64
ngram_lm_scale_0.08_attention_scale_0.5 10.64
ngram_lm_scale_0.08_attention_scale_0.6 10.64
ngram_lm_scale_0.08_attention_scale_0.7 10.64
ngram_lm_scale_0.08_attention_scale_1.5 10.64
ngram_lm_scale_0.08_attention_scale_1.7 10.64
ngram_lm_scale_0.08_attention_scale_2.5 10.64
ngram_lm_scale_0.08_attention_scale_4.0 10.64
ngram_lm_scale_0.08_attention_scale_5.0 10.64
ngram_lm_scale_0.1_attention_scale_0.3 10.64
ngram_lm_scale_0.1_attention_scale_1.5 10.64
ngram_lm_scale_0.1_attention_scale_4.0 10.64
ngram_lm_scale_0.7_attention_scale_1.3 10.64
ngram_lm_scale_0.9_attention_scale_2.2 10.64
ngram_lm_scale_1.0_attention_scale_3.0 10.64
ngram_lm_scale_1.1_attention_scale_4.0 10.64
ngram_lm_scale_1.3_attention_scale_5.0 10.64
ngram_lm_scale_0.01_attention_scale_0.6 10.65
ngram_lm_scale_0.01_attention_scale_0.7 10.65
ngram_lm_scale_0.01_attention_scale_1.3 10.65
ngram_lm_scale_0.01_attention_scale_1.5 10.65
ngram_lm_scale_0.01_attention_scale_1.7 10.65
ngram_lm_scale_0.01_attention_scale_1.9 10.65
ngram_lm_scale_0.01_attention_scale_2.0 10.65
ngram_lm_scale_0.01_attention_scale_2.1 10.65
ngram_lm_scale_0.01_attention_scale_2.2 10.65
ngram_lm_scale_0.01_attention_scale_2.3 10.65
ngram_lm_scale_0.01_attention_scale_2.5 10.65
ngram_lm_scale_0.01_attention_scale_3.0 10.65
ngram_lm_scale_0.08_attention_scale_0.3 10.65
ngram_lm_scale_0.5_attention_scale_0.5 10.65
ngram_lm_scale_0.6_attention_scale_0.7 10.65
ngram_lm_scale_0.7_attention_scale_1.1 10.65
ngram_lm_scale_0.7_attention_scale_1.2 10.65
ngram_lm_scale_0.9_attention_scale_2.1 10.65
ngram_lm_scale_1.2_attention_scale_4.0 10.65
ngram_lm_scale_0.05_attention_scale_0.3 10.66
ngram_lm_scale_0.7_attention_scale_1.0 10.66
ngram_lm_scale_0.9_attention_scale_1.9 10.66
ngram_lm_scale_0.9_attention_scale_2.0 10.66
ngram_lm_scale_1.0_attention_scale_2.5 10.66
ngram_lm_scale_1.1_attention_scale_3.0 10.66
ngram_lm_scale_0.01_attention_scale_0.5 10.67
ngram_lm_scale_0.1_attention_scale_0.08 10.67
ngram_lm_scale_0.1_attention_scale_0.1 10.67
ngram_lm_scale_0.6_attention_scale_0.6 10.67
ngram_lm_scale_0.9_attention_scale_1.7 10.67
ngram_lm_scale_1.0_attention_scale_2.2 10.67
ngram_lm_scale_1.0_attention_scale_2.3 10.67
ngram_lm_scale_1.3_attention_scale_4.0 10.67
ngram_lm_scale_1.5_attention_scale_5.0 10.67
ngram_lm_scale_0.01_attention_scale_0.3 10.68
ngram_lm_scale_0.08_attention_scale_0.08 10.68
ngram_lm_scale_0.08_attention_scale_0.1 10.68
ngram_lm_scale_0.3_attention_scale_0.08 10.68
ngram_lm_scale_0.3_attention_scale_0.1 10.68
ngram_lm_scale_0.7_attention_scale_0.9 10.68
ngram_lm_scale_1.0_attention_scale_2.0 10.68
ngram_lm_scale_1.0_attention_scale_2.1 10.68
ngram_lm_scale_1.1_attention_scale_2.5 10.68
ngram_lm_scale_1.2_attention_scale_3.0 10.68
ngram_lm_scale_0.1_attention_scale_0.05 10.69
ngram_lm_scale_0.5_attention_scale_0.3 10.69
ngram_lm_scale_0.9_attention_scale_1.5 10.69
ngram_lm_scale_1.0_attention_scale_1.9 10.69
ngram_lm_scale_1.1_attention_scale_2.3 10.69
ngram_lm_scale_0.05_attention_scale_0.1 10.7
ngram_lm_scale_0.08_attention_scale_0.05 10.7
ngram_lm_scale_0.3_attention_scale_0.05 10.7
ngram_lm_scale_0.6_attention_scale_0.5 10.7
ngram_lm_scale_1.1_attention_scale_2.2 10.7
ngram_lm_scale_1.5_attention_scale_4.0 10.7
ngram_lm_scale_1.7_attention_scale_5.0 10.7
ngram_lm_scale_0.05_attention_scale_0.08 10.71
ngram_lm_scale_1.1_attention_scale_2.1 10.71
ngram_lm_scale_1.2_attention_scale_2.5 10.71
ngram_lm_scale_1.3_attention_scale_3.0 10.71
ngram_lm_scale_0.01_attention_scale_0.1 10.72
ngram_lm_scale_0.05_attention_scale_0.05 10.72
ngram_lm_scale_0.08_attention_scale_0.01 10.72
ngram_lm_scale_0.1_attention_scale_0.01 10.72
ngram_lm_scale_0.3_attention_scale_0.01 10.72
ngram_lm_scale_0.7_attention_scale_0.7 10.72
ngram_lm_scale_0.9_attention_scale_1.3 10.72
ngram_lm_scale_1.0_attention_scale_1.7 10.72
ngram_lm_scale_1.1_attention_scale_2.0 10.72
ngram_lm_scale_0.01_attention_scale_0.08 10.73
ngram_lm_scale_0.9_attention_scale_1.2 10.73
ngram_lm_scale_1.1_attention_scale_1.9 10.73
ngram_lm_scale_1.2_attention_scale_2.3 10.73
ngram_lm_scale_1.0_attention_scale_1.5 10.74
ngram_lm_scale_1.2_attention_scale_2.2 10.74
ngram_lm_scale_1.3_attention_scale_2.5 10.74
ngram_lm_scale_1.9_attention_scale_5.0 10.74
ngram_lm_scale_0.01_attention_scale_0.05 10.75
ngram_lm_scale_0.05_attention_scale_0.01 10.75
ngram_lm_scale_0.7_attention_scale_0.6 10.75
ngram_lm_scale_0.9_attention_scale_1.1 10.75
ngram_lm_scale_1.1_attention_scale_1.7 10.75
ngram_lm_scale_1.2_attention_scale_2.1 10.75
ngram_lm_scale_1.7_attention_scale_4.0 10.75
ngram_lm_scale_1.2_attention_scale_2.0 10.76
ngram_lm_scale_1.3_attention_scale_2.3 10.76
ngram_lm_scale_2.0_attention_scale_5.0 10.76
ngram_lm_scale_1.0_attention_scale_1.3 10.77
ngram_lm_scale_1.2_attention_scale_1.9 10.77
ngram_lm_scale_1.5_attention_scale_3.0 10.77
ngram_lm_scale_0.01_attention_scale_0.01 10.78
ngram_lm_scale_0.6_attention_scale_0.3 10.78
ngram_lm_scale_0.7_attention_scale_0.5 10.78
ngram_lm_scale_0.9_attention_scale_1.0 10.78
ngram_lm_scale_2.1_attention_scale_5.0 10.78
ngram_lm_scale_1.1_attention_scale_1.5 10.79
ngram_lm_scale_1.3_attention_scale_2.2 10.79
ngram_lm_scale_0.5_attention_scale_0.1 10.8
ngram_lm_scale_1.0_attention_scale_1.2 10.8
ngram_lm_scale_1.3_attention_scale_2.1 10.8
ngram_lm_scale_1.9_attention_scale_4.0 10.8
ngram_lm_scale_2.2_attention_scale_5.0 10.8
ngram_lm_scale_0.5_attention_scale_0.08 10.81
ngram_lm_scale_0.9_attention_scale_0.9 10.81
ngram_lm_scale_1.2_attention_scale_1.7 10.81
ngram_lm_scale_1.3_attention_scale_2.0 10.81
ngram_lm_scale_1.0_attention_scale_1.1 10.82
ngram_lm_scale_0.5_attention_scale_0.05 10.83
ngram_lm_scale_1.1_attention_scale_1.3 10.83
ngram_lm_scale_1.3_attention_scale_1.9 10.83
ngram_lm_scale_1.5_attention_scale_2.5 10.84
ngram_lm_scale_2.3_attention_scale_5.0 10.84
ngram_lm_scale_1.0_attention_scale_1.0 10.85
ngram_lm_scale_1.2_attention_scale_1.5 10.85
ngram_lm_scale_2.0_attention_scale_4.0 10.85
ngram_lm_scale_1.1_attention_scale_1.2 10.86
ngram_lm_scale_1.7_attention_scale_3.0 10.86
ngram_lm_scale_0.5_attention_scale_0.01 10.87
ngram_lm_scale_1.5_attention_scale_2.3 10.87
ngram_lm_scale_0.7_attention_scale_0.3 10.88
ngram_lm_scale_0.9_attention_scale_0.7 10.88
ngram_lm_scale_1.3_attention_scale_1.7 10.88
ngram_lm_scale_1.0_attention_scale_0.9 10.89
ngram_lm_scale_1.5_attention_scale_2.2 10.89
ngram_lm_scale_2.1_attention_scale_4.0 10.89
ngram_lm_scale_1.1_attention_scale_1.1 10.91
ngram_lm_scale_0.6_attention_scale_0.1 10.92
ngram_lm_scale_0.9_attention_scale_0.6 10.92
ngram_lm_scale_1.5_attention_scale_2.1 10.92
ngram_lm_scale_1.2_attention_scale_1.3 10.93
ngram_lm_scale_2.5_attention_scale_5.0 10.93
ngram_lm_scale_0.6_attention_scale_0.08 10.94
ngram_lm_scale_2.2_attention_scale_4.0 10.94
ngram_lm_scale_1.1_attention_scale_1.0 10.95
ngram_lm_scale_1.3_attention_scale_1.5 10.95
ngram_lm_scale_1.5_attention_scale_2.0 10.96
ngram_lm_scale_1.2_attention_scale_1.2 10.97
ngram_lm_scale_1.7_attention_scale_2.5 10.97
ngram_lm_scale_0.6_attention_scale_0.05 10.98
ngram_lm_scale_1.9_attention_scale_3.0 10.98
ngram_lm_scale_1.0_attention_scale_0.7 10.99
ngram_lm_scale_1.5_attention_scale_1.9 10.99
ngram_lm_scale_2.3_attention_scale_4.0 10.99
ngram_lm_scale_0.9_attention_scale_0.5 11.0
ngram_lm_scale_1.1_attention_scale_0.9 11.0
ngram_lm_scale_0.6_attention_scale_0.01 11.02
ngram_lm_scale_1.2_attention_scale_1.1 11.02
ngram_lm_scale_1.7_attention_scale_2.3 11.03
ngram_lm_scale_1.3_attention_scale_1.3 11.05
ngram_lm_scale_2.0_attention_scale_3.0 11.05
ngram_lm_scale_1.7_attention_scale_2.2 11.07
ngram_lm_scale_1.0_attention_scale_0.6 11.08
ngram_lm_scale_1.5_attention_scale_1.7 11.08
ngram_lm_scale_1.2_attention_scale_1.0 11.09
ngram_lm_scale_0.7_attention_scale_0.1 11.1
ngram_lm_scale_1.3_attention_scale_1.2 11.1
ngram_lm_scale_1.7_attention_scale_2.1 11.11
ngram_lm_scale_2.1_attention_scale_3.0 11.12
ngram_lm_scale_2.5_attention_scale_4.0 11.12
ngram_lm_scale_0.7_attention_scale_0.08 11.13
ngram_lm_scale_1.9_attention_scale_2.5 11.13
ngram_lm_scale_1.7_attention_scale_2.0 11.14
ngram_lm_scale_1.2_attention_scale_0.9 11.16
ngram_lm_scale_1.1_attention_scale_0.7 11.17
ngram_lm_scale_1.3_attention_scale_1.1 11.17
ngram_lm_scale_3.0_attention_scale_5.0 11.17
ngram_lm_scale_0.7_attention_scale_0.05 11.18
ngram_lm_scale_1.5_attention_scale_1.5 11.18
ngram_lm_scale_1.0_attention_scale_0.5 11.19
ngram_lm_scale_1.7_attention_scale_1.9 11.2
ngram_lm_scale_2.2_attention_scale_3.0 11.21
ngram_lm_scale_1.9_attention_scale_2.3 11.22
ngram_lm_scale_2.0_attention_scale_2.5 11.23
ngram_lm_scale_0.9_attention_scale_0.3 11.25
ngram_lm_scale_1.3_attention_scale_1.0 11.26
ngram_lm_scale_0.7_attention_scale_0.01 11.27
ngram_lm_scale_1.9_attention_scale_2.2 11.27
ngram_lm_scale_1.1_attention_scale_0.6 11.29
ngram_lm_scale_2.3_attention_scale_3.0 11.31
ngram_lm_scale_1.7_attention_scale_1.7 11.33
ngram_lm_scale_1.5_attention_scale_1.3 11.34
ngram_lm_scale_1.9_attention_scale_2.1 11.34
ngram_lm_scale_2.0_attention_scale_2.3 11.34
ngram_lm_scale_2.1_attention_scale_2.5 11.35
ngram_lm_scale_1.3_attention_scale_0.9 11.36
ngram_lm_scale_1.2_attention_scale_0.7 11.39
ngram_lm_scale_1.9_attention_scale_2.0 11.4
ngram_lm_scale_2.0_attention_scale_2.2 11.4
ngram_lm_scale_1.5_attention_scale_1.2 11.43
ngram_lm_scale_1.1_attention_scale_0.5 11.44
ngram_lm_scale_2.0_attention_scale_2.1 11.47
ngram_lm_scale_2.1_attention_scale_2.3 11.47
ngram_lm_scale_2.2_attention_scale_2.5 11.47
ngram_lm_scale_1.9_attention_scale_1.9 11.48
ngram_lm_scale_1.7_attention_scale_1.5 11.5
ngram_lm_scale_2.5_attention_scale_3.0 11.51
ngram_lm_scale_3.0_attention_scale_4.0 11.51
ngram_lm_scale_1.0_attention_scale_0.3 11.53
ngram_lm_scale_1.2_attention_scale_0.6 11.53
ngram_lm_scale_1.5_attention_scale_1.1 11.54
ngram_lm_scale_2.1_attention_scale_2.2 11.54
ngram_lm_scale_2.0_attention_scale_2.0 11.55
ngram_lm_scale_2.3_attention_scale_2.5 11.59
ngram_lm_scale_2.2_attention_scale_2.3 11.61
ngram_lm_scale_2.1_attention_scale_2.1 11.62
ngram_lm_scale_1.3_attention_scale_0.7 11.63
ngram_lm_scale_2.0_attention_scale_1.9 11.63
ngram_lm_scale_1.9_attention_scale_1.7 11.66
ngram_lm_scale_1.5_attention_scale_1.0 11.67
ngram_lm_scale_2.2_attention_scale_2.2 11.69
ngram_lm_scale_0.9_attention_scale_0.1 11.7
ngram_lm_scale_2.1_attention_scale_2.0 11.71
ngram_lm_scale_1.2_attention_scale_0.5 11.72
ngram_lm_scale_1.7_attention_scale_1.3 11.72
ngram_lm_scale_2.3_attention_scale_2.3 11.75
ngram_lm_scale_0.9_attention_scale_0.08 11.77
ngram_lm_scale_2.2_attention_scale_2.1 11.78
ngram_lm_scale_2.1_attention_scale_1.9 11.82
ngram_lm_scale_1.3_attention_scale_0.6 11.83
ngram_lm_scale_1.5_attention_scale_0.9 11.85
ngram_lm_scale_2.0_attention_scale_1.7 11.85
ngram_lm_scale_2.3_attention_scale_2.2 11.86
ngram_lm_scale_0.9_attention_scale_0.05 11.87
ngram_lm_scale_1.1_attention_scale_0.3 11.87
ngram_lm_scale_1.7_attention_scale_1.2 11.88
ngram_lm_scale_1.9_attention_scale_1.5 11.9
ngram_lm_scale_2.2_attention_scale_2.0 11.9
ngram_lm_scale_2.5_attention_scale_2.5 11.9
ngram_lm_scale_4.0_attention_scale_5.0 11.93
ngram_lm_scale_2.3_attention_scale_2.1 11.97
ngram_lm_scale_0.9_attention_scale_0.01 12.0
ngram_lm_scale_2.2_attention_scale_1.9 12.02
ngram_lm_scale_1.7_attention_scale_1.1 12.05
ngram_lm_scale_1.3_attention_scale_0.5 12.07
ngram_lm_scale_2.1_attention_scale_1.7 12.07
ngram_lm_scale_2.3_attention_scale_2.0 12.09
ngram_lm_scale_1.0_attention_scale_0.1 12.11
ngram_lm_scale_2.5_attention_scale_2.3 12.11
ngram_lm_scale_2.0_attention_scale_1.5 12.14
ngram_lm_scale_1.0_attention_scale_0.08 12.19
ngram_lm_scale_3.0_attention_scale_3.0 12.19
ngram_lm_scale_1.9_attention_scale_1.3 12.22
ngram_lm_scale_1.7_attention_scale_1.0 12.23
ngram_lm_scale_2.3_attention_scale_1.9 12.23
ngram_lm_scale_2.5_attention_scale_2.2 12.23
ngram_lm_scale_1.5_attention_scale_0.7 12.27
ngram_lm_scale_1.2_attention_scale_0.3 12.28
ngram_lm_scale_2.2_attention_scale_1.7 12.3
ngram_lm_scale_1.0_attention_scale_0.05 12.32
ngram_lm_scale_2.5_attention_scale_2.1 12.37
ngram_lm_scale_2.1_attention_scale_1.5 12.39
ngram_lm_scale_1.9_attention_scale_1.2 12.41
ngram_lm_scale_1.7_attention_scale_0.9 12.46
ngram_lm_scale_1.0_attention_scale_0.01 12.49
ngram_lm_scale_2.0_attention_scale_1.3 12.5
ngram_lm_scale_2.5_attention_scale_2.0 12.51
ngram_lm_scale_2.3_attention_scale_1.7 12.54
ngram_lm_scale_1.5_attention_scale_0.6 12.55
ngram_lm_scale_1.1_attention_scale_0.1 12.58
ngram_lm_scale_1.9_attention_scale_1.1 12.62
ngram_lm_scale_2.2_attention_scale_1.5 12.64
ngram_lm_scale_1.1_attention_scale_0.08 12.67
ngram_lm_scale_2.5_attention_scale_1.9 12.67
ngram_lm_scale_4.0_attention_scale_4.0 12.67
ngram_lm_scale_1.3_attention_scale_0.3 12.71
ngram_lm_scale_2.0_attention_scale_1.2 12.71
ngram_lm_scale_2.1_attention_scale_1.3 12.78
ngram_lm_scale_3.0_attention_scale_2.5 12.8
ngram_lm_scale_1.1_attention_scale_0.05 12.81
ngram_lm_scale_1.9_attention_scale_1.0 12.85
ngram_lm_scale_1.5_attention_scale_0.5 12.86
ngram_lm_scale_2.3_attention_scale_1.5 12.91
ngram_lm_scale_2.0_attention_scale_1.1 12.92
ngram_lm_scale_1.7_attention_scale_0.7 12.99
ngram_lm_scale_2.1_attention_scale_1.2 12.99
ngram_lm_scale_5.0_attention_scale_5.0 13.01
ngram_lm_scale_1.1_attention_scale_0.01 13.02
ngram_lm_scale_2.5_attention_scale_1.7 13.02
ngram_lm_scale_2.2_attention_scale_1.3 13.05
ngram_lm_scale_3.0_attention_scale_2.3 13.09
ngram_lm_scale_1.2_attention_scale_0.1 13.1
ngram_lm_scale_1.9_attention_scale_0.9 13.11
ngram_lm_scale_2.0_attention_scale_1.0 13.17
ngram_lm_scale_1.2_attention_scale_0.08 13.2
ngram_lm_scale_2.1_attention_scale_1.1 13.22
ngram_lm_scale_3.0_attention_scale_2.2 13.24
ngram_lm_scale_2.2_attention_scale_1.2 13.28
ngram_lm_scale_1.7_attention_scale_0.6 13.33
ngram_lm_scale_2.3_attention_scale_1.3 13.34
ngram_lm_scale_1.2_attention_scale_0.05 13.36
ngram_lm_scale_3.0_attention_scale_2.1 13.42
ngram_lm_scale_2.5_attention_scale_1.5 13.43
ngram_lm_scale_2.0_attention_scale_0.9 13.48
ngram_lm_scale_2.1_attention_scale_1.0 13.51
ngram_lm_scale_2.2_attention_scale_1.1 13.56
ngram_lm_scale_1.2_attention_scale_0.01 13.6
ngram_lm_scale_2.3_attention_scale_1.2 13.6
ngram_lm_scale_3.0_attention_scale_2.0 13.62
ngram_lm_scale_1.3_attention_scale_0.1 13.65
ngram_lm_scale_1.5_attention_scale_0.3 13.68
ngram_lm_scale_1.7_attention_scale_0.5 13.72
ngram_lm_scale_1.3_attention_scale_0.08 13.76
ngram_lm_scale_1.9_attention_scale_0.7 13.78
ngram_lm_scale_3.0_attention_scale_1.9 13.81
ngram_lm_scale_2.1_attention_scale_0.9 13.82
ngram_lm_scale_2.2_attention_scale_1.0 13.85
ngram_lm_scale_4.0_attention_scale_3.0 13.85
ngram_lm_scale_2.3_attention_scale_1.1 13.89
ngram_lm_scale_1.3_attention_scale_0.05 13.94
ngram_lm_scale_2.5_attention_scale_1.3 13.94
ngram_lm_scale_5.0_attention_scale_4.0 13.97
ngram_lm_scale_1.9_attention_scale_0.6 14.15
ngram_lm_scale_2.0_attention_scale_0.7 14.16
ngram_lm_scale_2.2_attention_scale_0.9 14.17
ngram_lm_scale_2.3_attention_scale_1.0 14.19
ngram_lm_scale_1.3_attention_scale_0.01 14.2
ngram_lm_scale_2.5_attention_scale_1.2 14.2
ngram_lm_scale_3.0_attention_scale_1.7 14.26
ngram_lm_scale_2.5_attention_scale_1.1 14.48
ngram_lm_scale_2.3_attention_scale_0.9 14.5
ngram_lm_scale_2.1_attention_scale_0.7 14.53
ngram_lm_scale_2.0_attention_scale_0.6 14.54
ngram_lm_scale_1.9_attention_scale_0.5 14.57
ngram_lm_scale_4.0_attention_scale_2.5 14.63
ngram_lm_scale_1.7_attention_scale_0.3 14.64
ngram_lm_scale_3.0_attention_scale_1.5 14.71
ngram_lm_scale_1.5_attention_scale_0.1 14.75
ngram_lm_scale_2.5_attention_scale_1.0 14.79
ngram_lm_scale_2.2_attention_scale_0.7 14.86
ngram_lm_scale_1.5_attention_scale_0.08 14.87
ngram_lm_scale_2.1_attention_scale_0.6 14.91
ngram_lm_scale_2.0_attention_scale_0.5 14.95
ngram_lm_scale_4.0_attention_scale_2.3 14.98
ngram_lm_scale_1.5_attention_scale_0.05 15.05
ngram_lm_scale_2.5_attention_scale_0.9 15.12
ngram_lm_scale_4.0_attention_scale_2.2 15.17
ngram_lm_scale_2.3_attention_scale_0.7 15.21
ngram_lm_scale_3.0_attention_scale_1.3 15.22
ngram_lm_scale_2.2_attention_scale_0.6 15.27
ngram_lm_scale_1.5_attention_scale_0.01 15.3
ngram_lm_scale_5.0_attention_scale_3.0 15.32
ngram_lm_scale_2.1_attention_scale_0.5 15.33
ngram_lm_scale_4.0_attention_scale_2.1 15.37
ngram_lm_scale_1.9_attention_scale_0.3 15.5
ngram_lm_scale_3.0_attention_scale_1.2 15.51
ngram_lm_scale_4.0_attention_scale_2.0 15.57
ngram_lm_scale_2.3_attention_scale_0.6 15.61
ngram_lm_scale_2.2_attention_scale_0.5 15.68
ngram_lm_scale_1.7_attention_scale_0.1 15.72
ngram_lm_scale_4.0_attention_scale_1.9 15.79
ngram_lm_scale_3.0_attention_scale_1.1 15.82
ngram_lm_scale_1.7_attention_scale_0.08 15.83
ngram_lm_scale_2.5_attention_scale_0.7 15.85
ngram_lm_scale_2.0_attention_scale_0.3 15.87
ngram_lm_scale_2.3_attention_scale_0.5 16.0
ngram_lm_scale_1.7_attention_scale_0.05 16.01
ngram_lm_scale_3.0_attention_scale_1.0 16.11
ngram_lm_scale_5.0_attention_scale_2.5 16.12
ngram_lm_scale_2.5_attention_scale_0.6 16.19
ngram_lm_scale_2.1_attention_scale_0.3 16.2
ngram_lm_scale_4.0_attention_scale_1.7 16.22
ngram_lm_scale_1.7_attention_scale_0.01 16.23
ngram_lm_scale_3.0_attention_scale_0.9 16.4
ngram_lm_scale_5.0_attention_scale_2.3 16.44
ngram_lm_scale_1.9_attention_scale_0.1 16.5
ngram_lm_scale_2.2_attention_scale_0.3 16.53
ngram_lm_scale_2.5_attention_scale_0.5 16.54
ngram_lm_scale_1.9_attention_scale_0.08 16.6
ngram_lm_scale_5.0_attention_scale_2.2 16.6
ngram_lm_scale_4.0_attention_scale_1.5 16.63
ngram_lm_scale_1.9_attention_scale_0.05 16.74
ngram_lm_scale_5.0_attention_scale_2.1 16.77
ngram_lm_scale_2.3_attention_scale_0.3 16.81
ngram_lm_scale_2.0_attention_scale_0.1 16.83
ngram_lm_scale_2.0_attention_scale_0.08 16.92
ngram_lm_scale_5.0_attention_scale_2.0 16.94
ngram_lm_scale_1.9_attention_scale_0.01 16.95
ngram_lm_scale_3.0_attention_scale_0.7 16.96
ngram_lm_scale_2.0_attention_scale_0.05 17.05
ngram_lm_scale_4.0_attention_scale_1.3 17.05
ngram_lm_scale_2.1_attention_scale_0.1 17.11
ngram_lm_scale_5.0_attention_scale_1.9 17.11
ngram_lm_scale_2.1_attention_scale_0.08 17.21
ngram_lm_scale_2.0_attention_scale_0.01 17.24
ngram_lm_scale_3.0_attention_scale_0.6 17.26
ngram_lm_scale_4.0_attention_scale_1.2 17.27
ngram_lm_scale_2.5_attention_scale_0.3 17.28
ngram_lm_scale_2.1_attention_scale_0.05 17.34
ngram_lm_scale_2.2_attention_scale_0.1 17.38
ngram_lm_scale_5.0_attention_scale_1.7 17.44
ngram_lm_scale_2.2_attention_scale_0.08 17.46
ngram_lm_scale_4.0_attention_scale_1.1 17.5
ngram_lm_scale_2.1_attention_scale_0.01 17.52
ngram_lm_scale_3.0_attention_scale_0.5 17.57
ngram_lm_scale_2.2_attention_scale_0.05 17.59
ngram_lm_scale_2.3_attention_scale_0.1 17.62
ngram_lm_scale_2.3_attention_scale_0.08 17.7
ngram_lm_scale_4.0_attention_scale_1.0 17.72
ngram_lm_scale_2.2_attention_scale_0.01 17.76
ngram_lm_scale_5.0_attention_scale_1.5 17.8
ngram_lm_scale_2.3_attention_scale_0.05 17.82
ngram_lm_scale_4.0_attention_scale_0.9 17.94
ngram_lm_scale_2.3_attention_scale_0.01 17.98
ngram_lm_scale_2.5_attention_scale_0.1 18.03
ngram_lm_scale_2.5_attention_scale_0.08 18.1
ngram_lm_scale_5.0_attention_scale_1.3 18.12
ngram_lm_scale_3.0_attention_scale_0.3 18.17
ngram_lm_scale_2.5_attention_scale_0.05 18.2
ngram_lm_scale_5.0_attention_scale_1.2 18.29
ngram_lm_scale_2.5_attention_scale_0.01 18.33
ngram_lm_scale_4.0_attention_scale_0.7 18.36
ngram_lm_scale_5.0_attention_scale_1.1 18.48
ngram_lm_scale_4.0_attention_scale_0.6 18.58
ngram_lm_scale_5.0_attention_scale_1.0 18.65
ngram_lm_scale_3.0_attention_scale_0.1 18.75
ngram_lm_scale_4.0_attention_scale_0.5 18.79
ngram_lm_scale_3.0_attention_scale_0.08 18.81
ngram_lm_scale_5.0_attention_scale_0.9 18.81
ngram_lm_scale_3.0_attention_scale_0.05 18.89
ngram_lm_scale_3.0_attention_scale_0.01 18.99
ngram_lm_scale_5.0_attention_scale_0.7 19.11
ngram_lm_scale_4.0_attention_scale_0.3 19.18
ngram_lm_scale_5.0_attention_scale_0.6 19.25
ngram_lm_scale_5.0_attention_scale_0.5 19.41
ngram_lm_scale_4.0_attention_scale_0.1 19.57
ngram_lm_scale_4.0_attention_scale_0.08 19.61
ngram_lm_scale_4.0_attention_scale_0.05 19.67
ngram_lm_scale_5.0_attention_scale_0.3 19.71
ngram_lm_scale_4.0_attention_scale_0.01 19.73
ngram_lm_scale_5.0_attention_scale_0.1 19.99
ngram_lm_scale_5.0_attention_scale_0.08 20.01
ngram_lm_scale_5.0_attention_scale_0.05 20.05
ngram_lm_scale_5.0_attention_scale_0.01 20.11
2022-04-09 04:57:33,455 INFO [decode_test.py:730] Done!
|