num_sent_processes: 4 [0] Creating shuffled datasets in /scratch/gz5hp/tfbs_region2vec_models/expr_universe_tile25k/5W100D-0.5000r/shuffled_datasets [1] Creating shuffled datasets in /scratch/gz5hp/tfbs_region2vec_models/expr_universe_tile25k/5W100D-0.5000r/shuffled_datasets [2] Creating shuffled datasets in /scratch/gz5hp/tfbs_region2vec_models/expr_universe_tile25k/5W100D-0.5000r/shuffled_datasets [3] Creating shuffled datasets in /scratch/gz5hp/tfbs_region2vec_models/expr_universe_tile25k/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_tile25k/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 [03/12/24-19:10:44] Start training [03/12/24-19:10:44] Building vocabulary [03/12/24-19:10:58] Vocabulary size is 100053 [Shuffling 1] loss 1330854.2500 lr 0.5000 vocab_size 100053 (6.76s/11.51m) [Shuffling 2] loss 25639.1094 lr 0.4950 vocab_size 100053 (6.88s/11.62m) [Shuffling 3] loss 13143.0117 lr 0.4900 vocab_size 100053 (6.90s/11.66m) [Shuffling 4] loss 8284.6133 lr 0.4850 vocab_size 100053 (6.88s/11.68m) [Shuffling 5] loss 6425.2603 lr 0.4800 vocab_size 100053 (6.89s/11.69m) [Shuffling 6] loss 5221.4395 lr 0.4750 vocab_size 100053 (6.98s/11.72m) [Shuffling 7] loss 3974.6724 lr 0.4700 vocab_size 100053 (6.96s/11.74m) [Shuffling 8] loss 3617.1482 lr 0.4650 vocab_size 100053 (6.97s/11.76m) [Shuffling 9] loss 2895.0222 lr 0.4600 vocab_size 100053 (7.00s/11.77m) [Shuffling 10] loss 2590.2261 lr 0.4550 vocab_size 100053 (6.95s/11.78m) [Shuffling 11] loss 1813.6299 lr 0.4500 vocab_size 100053 (6.91s/11.78m) [Shuffling 12] loss 1902.6003 lr 0.4450 vocab_size 100053 (6.85s/11.77m) [Shuffling 13] loss 1684.7363 lr 0.4400 vocab_size 100053 (7.02s/11.78m) [Shuffling 14] loss 1369.1703 lr 0.4350 vocab_size 100053 (6.96s/11.79m) [Shuffling 15] loss 1233.3844 lr 0.4300 vocab_size 100053 (6.92s/11.79m) [Shuffling 16] loss 1258.8016 lr 0.4250 vocab_size 100053 (6.82s/11.78m) [Shuffling 17] loss 1141.7491 lr 0.4200 vocab_size 100053 (6.94s/11.78m) [Shuffling 18] loss 1164.3254 lr 0.4150 vocab_size 100053 (6.85s/11.77m) [Shuffling 19] loss 1050.1744 lr 0.4100 vocab_size 100053 (7.00s/11.78m) [Shuffling 20] loss 910.3453 lr 0.4050 vocab_size 100053 (6.99s/11.79m) [Shuffling 21] loss 867.1285 lr 0.4000 vocab_size 100053 (6.92s/11.79m) [Shuffling 22] loss 777.0692 lr 0.3950 vocab_size 100053 (6.92s/11.78m) [Shuffling 23] loss 825.7840 lr 0.3900 vocab_size 100053 (6.85s/11.78m) [Shuffling 24] loss 654.6333 lr 0.3850 vocab_size 100053 (6.96s/11.78m) [Shuffling 25] loss 661.7099 lr 0.3800 vocab_size 100053 (6.81s/11.78m) [Shuffling 26] loss 759.1976 lr 0.3750 vocab_size 100053 (6.89s/11.77m) [Shuffling 27] loss 585.4982 lr 0.3700 vocab_size 100053 (7.00s/11.78m) [Shuffling 28] loss 678.0389 lr 0.3650 vocab_size 100053 (6.92s/11.78m) [Shuffling 29] loss 699.3723 lr 0.3600 vocab_size 100053 (6.91s/11.78m) [Shuffling 30] loss 517.1392 lr 0.3550 vocab_size 100053 (7.12s/11.79m) [Shuffling 31] loss 568.7162 lr 0.3500 vocab_size 100053 (6.85s/11.79m) [Shuffling 32] loss 584.4830 lr 0.3450 vocab_size 100053 (6.85s/11.78m) [Shuffling 33] loss 511.0351 lr 0.3400 vocab_size 100053 (6.92s/11.78m) [Shuffling 34] loss 500.3179 lr 0.3350 vocab_size 100053 (6.97s/11.78m) [Shuffling 35] loss 444.5648 lr 0.3300 vocab_size 100053 (7.01s/11.79m) [Shuffling 36] loss 413.2932 lr 0.3250 vocab_size 100053 (6.96s/11.79m) [Shuffling 37] loss 450.2869 lr 0.3200 vocab_size 100053 (6.81s/11.79m) [Shuffling 38] loss 427.4685 lr 0.3150 vocab_size 100053 (6.91s/11.78m) [Shuffling 39] loss 419.9331 lr 0.3100 vocab_size 100053 (6.85s/11.78m) [Shuffling 40] loss 354.4765 lr 0.3050 vocab_size 100053 (6.94s/11.78m) [Shuffling 41] loss 360.3784 lr 0.3000 vocab_size 100053 (6.78s/11.78m) [Shuffling 42] loss 398.0664 lr 0.2950 vocab_size 100053 (6.87s/11.77m) [Shuffling 43] loss 335.4212 lr 0.2900 vocab_size 100053 (6.94s/11.78m) [Shuffling 44] loss 315.3892 lr 0.2850 vocab_size 100053 (7.00s/11.78m) [Shuffling 45] loss 355.1169 lr 0.2800 vocab_size 100053 (6.86s/11.78m) [Shuffling 46] loss 407.7681 lr 0.2750 vocab_size 100053 (6.85s/11.77m) [Shuffling 47] loss 305.9730 lr 0.2700 vocab_size 100053 (6.97s/11.78m) [Shuffling 48] loss 305.1111 lr 0.2650 vocab_size 100053 (6.96s/11.78m) [Shuffling 49] loss 353.7332 lr 0.2600 vocab_size 100053 (6.82s/11.77m) [Shuffling 50] loss 343.9136 lr 0.2550 vocab_size 100053 (6.84s/11.77m) [Shuffling 51] loss 283.5084 lr 0.2500 vocab_size 100053 (6.97s/11.77m) [Shuffling 52] loss 251.4870 lr 0.2451 vocab_size 100053 (6.93s/11.77m) [Shuffling 53] loss 323.9646 lr 0.2401 vocab_size 100053 (6.84s/11.77m) [Shuffling 54] loss 313.9809 lr 0.2351 vocab_size 100053 (6.82s/11.77m) [Shuffling 55] loss 255.6028 lr 0.2301 vocab_size 100053 (6.82s/11.77m) [Shuffling 56] loss 300.4524 lr 0.2251 vocab_size 100053 (6.81s/11.76m) [Shuffling 57] loss 269.9436 lr 0.2201 vocab_size 100053 (6.88s/11.76m) [Shuffling 58] loss 271.5844 lr 0.2151 vocab_size 100053 (6.87s/11.76m) [Shuffling 59] loss 180.9895 lr 0.2101 vocab_size 100053 (7.00s/11.76m) [Shuffling 60] loss 231.2683 lr 0.2051 vocab_size 100053 (6.85s/11.76m) [Shuffling 61] loss 271.9186 lr 0.2001 vocab_size 100053 (6.88s/11.76m) [Shuffling 62] loss 308.6990 lr 0.1951 vocab_size 100053 (6.84s/11.76m) [Shuffling 63] loss 229.0503 lr 0.1901 vocab_size 100053 (6.87s/11.76m) [Shuffling 64] loss 234.9249 lr 0.1851 vocab_size 100053 (6.87s/11.76m) [Shuffling 65] loss 230.2720 lr 0.1801 vocab_size 100053 (6.87s/11.76m) [Shuffling 66] loss 250.0107 lr 0.1751 vocab_size 100053 (6.84s/11.75m) [Shuffling 67] loss 239.9893 lr 0.1701 vocab_size 100053 (6.87s/11.75m) [Shuffling 68] loss 318.6678 lr 0.1651 vocab_size 100053 (6.86s/11.75m) [Shuffling 69] loss 202.4791 lr 0.1601 vocab_size 100053 (6.85s/11.75m) [Shuffling 70] loss 155.1777 lr 0.1551 vocab_size 100053 (6.99s/11.75m) [Shuffling 71] loss 223.7989 lr 0.1501 vocab_size 100053 (6.88s/11.75m) [Shuffling 72] loss 201.0650 lr 0.1451 vocab_size 100053 (6.89s/11.75m) [Shuffling 73] loss 214.5350 lr 0.1401 vocab_size 100053 (6.87s/11.75m) [Shuffling 74] loss 229.4717 lr 0.1351 vocab_size 100053 (6.84s/11.75m) [Shuffling 75] loss 189.5206 lr 0.1301 vocab_size 100053 (6.91s/11.75m) [Shuffling 76] loss 183.1143 lr 0.1251 vocab_size 100053 (6.85s/11.75m) [Shuffling 77] loss 154.9450 lr 0.1201 vocab_size 100053 (6.99s/11.75m) [Shuffling 78] loss 194.5131 lr 0.1151 vocab_size 100053 (6.87s/11.75m) [Shuffling 79] loss 131.4584 lr 0.1101 vocab_size 100053 (6.96s/11.75m) [Shuffling 80] loss 184.4336 lr 0.1051 vocab_size 100053 (6.87s/11.75m) [Shuffling 81] loss 159.2261 lr 0.1001 vocab_size 100053 (6.82s/11.75m) [Shuffling 82] loss 173.7853 lr 0.0951 vocab_size 100053 (6.85s/11.75m) [Shuffling 83] loss 186.8891 lr 0.0901 vocab_size 100053 (6.84s/11.75m) [Shuffling 84] loss 185.3372 lr 0.0851 vocab_size 100053 (6.88s/11.75m) [Shuffling 85] loss 202.5962 lr 0.0801 vocab_size 100053 (6.86s/11.75m) [Shuffling 86] loss 160.5797 lr 0.0751 vocab_size 100053 (6.87s/11.75m) [Shuffling 87] loss 133.8623 lr 0.0701 vocab_size 100053 (6.86s/11.74m) [Shuffling 88] loss 166.0749 lr 0.0651 vocab_size 100053 (6.84s/11.74m) [Shuffling 89] loss 199.6577 lr 0.0601 vocab_size 100053 (6.86s/11.74m) [Shuffling 90] loss 147.9556 lr 0.0551 vocab_size 100053 (6.84s/11.74m) [Shuffling 91] loss 211.9348 lr 0.0501 vocab_size 100053 (6.85s/11.74m) [Shuffling 92] loss 182.5257 lr 0.0451 vocab_size 100053 (6.86s/11.74m) [Shuffling 93] loss 137.2178 lr 0.0401 vocab_size 100053 (6.84s/11.74m) [Shuffling 94] loss 195.0915 lr 0.0351 vocab_size 100053 (6.82s/11.74m) [Shuffling 95] loss 188.8390 lr 0.0301 vocab_size 100053 (6.91s/11.74m) [Shuffling 96] loss 153.7831 lr 0.0251 vocab_size 100053 (11.05s/11.81m) [Shuffling 97] loss 142.9752 lr 0.0201 vocab_size 100053 (6.92s/11.81m) [Shuffling 98] loss 181.9661 lr 0.0151 vocab_size 100053 (10.01s/11.86m) [Shuffling 99] loss 119.3859 lr 0.0101 vocab_size 100053 (7.84s/11.88m) [Shuffling 100] loss 198.6681 lr 0.0051 vocab_size 100053 (16.77s/12.04m) [03/12/24-19:22:46] Training finished, training Time 12.04m