num_sent_processes: 4 [3] Creating shuffled datasets in /scratch/gz5hp/tfbs_region2vec_models/expr_universe_tile5k/5W100D-0.5000r/shuffled_datasets [2] Creating shuffled datasets in /scratch/gz5hp/tfbs_region2vec_models/expr_universe_tile5k/5W100D-0.5000r/shuffled_datasets [0] Creating shuffled datasets in /scratch/gz5hp/tfbs_region2vec_models/expr_universe_tile5k/5W100D-0.5000r/shuffled_datasets [1] Creating shuffled datasets in /scratch/gz5hp/tfbs_region2vec_models/expr_universe_tile5k/5W100D-0.5000r/shuffled_datasets {'num_shuffle': 100, 'embed_dim': 100, 'context_len': 5, 'nworkers': 6, 'save_freq': -1, 'save_dir': '/scratch/gz5hp/tfbs_region2vec_models/expr_universe_tile5k/5W100D-0.5000r', 'resume': '', 'train_alg': 'cbow', 'min_count': 5, 'neg_samples': 5, 'init_lr': 0.5, 'min_lr': 0.0001, 'lr_mode': 'linear', 'milestones': [], 'hier_softmax': False, 'update_vocab': 'once', 'seed': 0} Using cbow, negative sampling with 5 negative samples [03/12/24-20:49:33] Start training [03/12/24-20:49:33] Building vocabulary [03/12/24-20:50:20] Vocabulary size is 251084 [Shuffling 1] loss 2456195.0000 lr 0.5000 vocab_size 251084 (13.05s/22.53m) [Shuffling 2] loss 104599.4844 lr 0.4950 vocab_size 251084 (11.11s/20.93m) [Shuffling 3] loss 43089.6836 lr 0.4900 vocab_size 251084 (11.58s/20.66m) [Shuffling 4] loss 27407.3320 lr 0.4850 vocab_size 251084 (11.64s/20.55m) [Shuffling 5] loss 17726.8652 lr 0.4800 vocab_size 251084 (11.34s/20.39m) [Shuffling 6] loss 14095.8379 lr 0.4750 vocab_size 251084 (11.15s/20.22m) [Shuffling 7] loss 12006.0625 lr 0.4700 vocab_size 251084 (11.58s/20.21m) [Shuffling 8] loss 9470.8018 lr 0.4650 vocab_size 251084 (11.16s/20.11m) [Shuffling 9] loss 7224.0098 lr 0.4600 vocab_size 251084 (10.93s/19.99m) [Shuffling 10] loss 7067.5063 lr 0.4550 vocab_size 251084 (11.79s/20.04m) [Shuffling 11] loss 5749.8257 lr 0.4500 vocab_size 251084 (11.34s/20.01m) [Shuffling 12] loss 5223.3828 lr 0.4450 vocab_size 251084 (10.86s/19.92m) [Shuffling 13] loss 4864.0063 lr 0.4400 vocab_size 251084 (10.77s/19.83m) [Shuffling 14] loss 4439.8608 lr 0.4350 vocab_size 251084 (11.18s/19.80m) [Shuffling 15] loss 4180.6655 lr 0.4300 vocab_size 251084 (11.05s/19.76m) [Shuffling 16] loss 3617.9607 lr 0.4250 vocab_size 251084 (12.00s/19.83m) [Shuffling 17] loss 3218.5100 lr 0.4200 vocab_size 251084 (10.96s/19.78m) [Shuffling 18] loss 3251.9504 lr 0.4150 vocab_size 251084 (11.03s/19.75m) [Shuffling 19] loss 2966.3467 lr 0.4100 vocab_size 251084 (11.23s/19.74m) [Shuffling 20] loss 2985.1082 lr 0.4050 vocab_size 251084 (10.68s/19.68m) [Shuffling 21] loss 2584.6106 lr 0.4000 vocab_size 251084 (11.03s/19.66m) [Shuffling 22] loss 2646.2993 lr 0.3950 vocab_size 251084 (12.27s/19.73m) [Shuffling 23] loss 2446.6914 lr 0.3900 vocab_size 251084 (11.07s/19.71m) [Shuffling 24] loss 2297.1453 lr 0.3850 vocab_size 251084 (11.66s/19.74m) [Shuffling 25] loss 2047.9167 lr 0.3800 vocab_size 251084 (10.75s/19.69m) [Shuffling 26] loss 2099.0664 lr 0.3750 vocab_size 251084 (10.77s/19.66m) [Shuffling 27] loss 1911.0421 lr 0.3700 vocab_size 251084 (11.31s/19.66m) [Shuffling 28] loss 1812.6973 lr 0.3650 vocab_size 251084 (12.14s/19.71m) [Shuffling 29] loss 1852.6920 lr 0.3600 vocab_size 251084 (10.61s/19.67m) [Shuffling 30] loss 1742.0497 lr 0.3550 vocab_size 251084 (11.63s/19.68m) [Shuffling 31] loss 1479.8383 lr 0.3500 vocab_size 251084 (11.13s/19.67m) [Shuffling 32] loss 1448.3596 lr 0.3450 vocab_size 251084 (11.21s/19.67m) [Shuffling 33] loss 1343.0952 lr 0.3400 vocab_size 251084 (10.78s/19.64m) [Shuffling 34] loss 1778.9907 lr 0.3350 vocab_size 251084 (11.10s/19.63m) [Shuffling 35] loss 1392.9266 lr 0.3300 vocab_size 251084 (10.81s/19.61m) [Shuffling 36] loss 1383.3906 lr 0.3250 vocab_size 251084 (11.89s/19.64m) [Shuffling 37] loss 1100.6746 lr 0.3200 vocab_size 251084 (12.04s/19.67m) [Shuffling 38] loss 1403.8368 lr 0.3150 vocab_size 251084 (10.71s/19.65m) [Shuffling 39] loss 1106.8835 lr 0.3100 vocab_size 251084 (10.60s/19.62m) [Shuffling 40] loss 1176.3523 lr 0.3050 vocab_size 251084 (10.67s/19.59m) [Shuffling 41] loss 1087.1010 lr 0.3000 vocab_size 251084 (13.46s/19.68m) [Shuffling 42] loss 954.1144 lr 0.2950 vocab_size 251084 (11.40s/19.68m) [Shuffling 43] loss 952.5590 lr 0.2900 vocab_size 251084 (11.21s/19.68m) [Shuffling 44] loss 1097.9375 lr 0.2850 vocab_size 251084 (10.82s/19.66m) [Shuffling 45] loss 1014.6675 lr 0.2800 vocab_size 251084 (10.88s/19.64m) [Shuffling 46] loss 975.7088 lr 0.2750 vocab_size 251084 (11.19s/19.64m) [Shuffling 47] loss 976.2994 lr 0.2700 vocab_size 251084 (11.98s/19.67m) [Shuffling 48] loss 943.4694 lr 0.2650 vocab_size 251084 (10.93s/19.65m) [Shuffling 49] loss 943.0305 lr 0.2600 vocab_size 251084 (11.25s/19.65m) [Shuffling 50] loss 1096.9633 lr 0.2550 vocab_size 251084 (10.85s/19.64m) [Shuffling 51] loss 887.6270 lr 0.2500 vocab_size 251084 (12.83s/19.69m) [Shuffling 52] loss 910.3066 lr 0.2451 vocab_size 251084 (10.67s/19.66m) [Shuffling 53] loss 765.9875 lr 0.2401 vocab_size 251084 (11.29s/19.66m) [Shuffling 54] loss 999.5359 lr 0.2351 vocab_size 251084 (10.63s/19.64m) [Shuffling 55] loss 759.3408 lr 0.2301 vocab_size 251084 (12.10s/19.67m) [Shuffling 56] loss 778.1039 lr 0.2251 vocab_size 251084 (11.61s/19.68m) [Shuffling 57] loss 867.2405 lr 0.2201 vocab_size 251084 (10.95s/19.67m) [Shuffling 58] loss 742.4241 lr 0.2151 vocab_size 251084 (11.11s/19.66m) [Shuffling 59] loss 697.8636 lr 0.2101 vocab_size 251084 (10.95s/19.65m) [Shuffling 60] loss 806.7392 lr 0.2051 vocab_size 251084 (11.20s/19.65m) [Shuffling 61] loss 671.5496 lr 0.2001 vocab_size 251084 (11.23s/19.65m) [Shuffling 62] loss 603.1574 lr 0.1951 vocab_size 251084 (10.98s/19.64m) [Shuffling 63] loss 770.0544 lr 0.1901 vocab_size 251084 (11.56s/19.65m) [Shuffling 64] loss 673.6166 lr 0.1851 vocab_size 251084 (11.01s/19.64m) [Shuffling 65] loss 791.8885 lr 0.1801 vocab_size 251084 (10.99s/19.63m) [Shuffling 66] loss 680.6115 lr 0.1751 vocab_size 251084 (10.59s/19.61m) [Shuffling 67] loss 755.3920 lr 0.1701 vocab_size 251084 (11.87s/19.63m) [Shuffling 68] loss 593.0894 lr 0.1651 vocab_size 251084 (12.09s/19.65m) [Shuffling 69] loss 612.0858 lr 0.1601 vocab_size 251084 (10.92s/19.64m) [Shuffling 70] loss 634.3549 lr 0.1551 vocab_size 251084 (10.93s/19.63m) [Shuffling 71] loss 624.5287 lr 0.1501 vocab_size 251084 (11.11s/19.62m) [Shuffling 72] loss 518.7010 lr 0.1451 vocab_size 251084 (12.31s/19.65m) [Shuffling 73] loss 593.1895 lr 0.1401 vocab_size 251084 (11.35s/19.65m) [Shuffling 74] loss 639.5143 lr 0.1351 vocab_size 251084 (11.51s/19.65m) [Shuffling 75] loss 609.4088 lr 0.1301 vocab_size 251084 (11.15s/19.65m) [Shuffling 76] loss 579.9636 lr 0.1251 vocab_size 251084 (10.62s/19.64m) [Shuffling 77] loss 598.2985 lr 0.1201 vocab_size 251084 (11.88s/19.65m) [Shuffling 78] loss 601.9803 lr 0.1151 vocab_size 251084 (11.96s/19.66m) [Shuffling 79] loss 453.9713 lr 0.1101 vocab_size 251084 (11.23s/19.66m) [Shuffling 80] loss 539.0922 lr 0.1051 vocab_size 251084 (10.67s/19.65m) [Shuffling 81] loss 520.2746 lr 0.1001 vocab_size 251084 (10.81s/19.64m) [Shuffling 82] loss 472.6655 lr 0.0951 vocab_size 251084 (13.29s/19.68m) [Shuffling 83] loss 558.3383 lr 0.0901 vocab_size 251084 (10.83s/19.67m) [Shuffling 84] loss 503.6202 lr 0.0851 vocab_size 251084 (10.64s/19.65m) [Shuffling 85] loss 537.7441 lr 0.0801 vocab_size 251084 (10.57s/19.64m) [Shuffling 86] loss 560.8794 lr 0.0751 vocab_size 251084 (11.83s/19.65m) [Shuffling 87] loss 532.8023 lr 0.0701 vocab_size 251084 (12.93s/19.68m) [Shuffling 88] loss 554.0289 lr 0.0651 vocab_size 251084 (12.13s/19.70m) [Shuffling 89] loss 459.1537 lr 0.0601 vocab_size 251084 (11.81s/19.71m) [Shuffling 90] loss 483.2545 lr 0.0551 vocab_size 251084 (11.19s/19.70m) [Shuffling 91] loss 518.9536 lr 0.0501 vocab_size 251084 (11.34s/19.70m) [Shuffling 92] loss 482.2791 lr 0.0451 vocab_size 251084 (10.81s/19.69m) [Shuffling 93] loss 639.9487 lr 0.0401 vocab_size 251084 (11.43s/19.70m) [Shuffling 94] loss 425.6749 lr 0.0351 vocab_size 251084 (12.25s/19.71m) [Shuffling 95] loss 378.1344 lr 0.0301 vocab_size 251084 (11.01s/19.71m) [Shuffling 96] loss 452.0578 lr 0.0251 vocab_size 251084 (11.07s/19.70m) [Shuffling 97] loss 486.5109 lr 0.0201 vocab_size 251084 (10.84s/19.69m) [Shuffling 98] loss 497.5575 lr 0.0151 vocab_size 251084 (12.01s/19.71m) [Shuffling 99] loss 445.2628 lr 0.0101 vocab_size 251084 (11.91s/19.72m) [Shuffling 100] loss 434.2335 lr 0.0051 vocab_size 251084 (12.17s/19.73m) [03/12/24-21:09:17] Training finished, training Time 19.73m