num_sent_processes: 4 [0] Creating shuffled datasets in /scratch/gz5hp/tfbs_region2vec_models/expr_universe_tile25k/5W100D-0.0250r/shuffled_datasets [1] Creating shuffled datasets in /scratch/gz5hp/tfbs_region2vec_models/expr_universe_tile25k/5W100D-0.0250r/shuffled_datasets [2] Creating shuffled datasets in /scratch/gz5hp/tfbs_region2vec_models/expr_universe_tile25k/5W100D-0.0250r/shuffled_datasets [3] Creating shuffled datasets in /scratch/gz5hp/tfbs_region2vec_models/expr_universe_tile25k/5W100D-0.0250r/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-0.0250r', 'resume': '', 'train_alg': 'cbow', 'min_count': 5, 'neg_samples': 5, 'init_lr': 0.025, 'min_lr': 0.0001, 'lr_mode': 'linear', 'milestones': [], 'hier_softmax': False, 'update_vocab': 'once', 'seed': 0} Using cbow, negative sampling with 5 negative samples [03/12/24-17:31:30] Start training [03/12/24-17:31:30] Building vocabulary [03/12/24-17:31:45] Vocabulary size is 100053 [Shuffling 1] loss 5544917.5000 lr 0.0250 vocab_size 100053 (6.83s/11.62m) [Shuffling 2] loss 4419701.5000 lr 0.0248 vocab_size 100053 (7.04s/11.80m) [Shuffling 3] loss 4399375.0000 lr 0.0245 vocab_size 100053 (7.09s/11.89m) [Shuffling 4] loss 4254142.5000 lr 0.0243 vocab_size 100053 (7.09s/11.93m) [Shuffling 5] loss 4035336.7500 lr 0.0240 vocab_size 100053 (7.15s/11.98m) [Shuffling 6] loss 4079432.7500 lr 0.0238 vocab_size 100053 (7.00s/11.97m) [Shuffling 7] loss 4056783.7500 lr 0.0235 vocab_size 100053 (7.01s/11.96m) [Shuffling 8] loss 4108757.5000 lr 0.0233 vocab_size 100053 (7.00s/11.96m) [Shuffling 9] loss 4042045.7500 lr 0.0230 vocab_size 100053 (6.95s/11.94m) [Shuffling 10] loss 4170227.0000 lr 0.0228 vocab_size 100053 (7.01s/11.94m) [Shuffling 11] loss 4101290.5000 lr 0.0225 vocab_size 100053 (7.00s/11.94m) [Shuffling 12] loss 3996420.2500 lr 0.0223 vocab_size 100053 (7.16s/11.96m) [Shuffling 13] loss 3958420.5000 lr 0.0220 vocab_size 100053 (7.16s/11.98m) [Shuffling 14] loss 4053123.2500 lr 0.0218 vocab_size 100053 (7.00s/11.97m) [Shuffling 15] loss 3978231.0000 lr 0.0215 vocab_size 100053 (6.96s/11.96m) [Shuffling 16] loss 3988039.0000 lr 0.0213 vocab_size 100053 (7.01s/11.96m) [Shuffling 17] loss 4101636.2500 lr 0.0210 vocab_size 100053 (7.06s/11.96m) [Shuffling 18] loss 3979162.5000 lr 0.0208 vocab_size 100053 (7.07s/11.97m) [Shuffling 19] loss 3862451.2500 lr 0.0205 vocab_size 100053 (7.15s/11.98m) [Shuffling 20] loss 4074324.7500 lr 0.0203 vocab_size 100053 (6.93s/11.97m) [Shuffling 21] loss 4153712.0000 lr 0.0200 vocab_size 100053 (6.93s/11.96m) [Shuffling 22] loss 4017930.7500 lr 0.0198 vocab_size 100053 (6.90s/11.95m) [Shuffling 23] loss 4042241.2500 lr 0.0195 vocab_size 100053 (6.91s/11.94m) [Shuffling 24] loss 3979756.2500 lr 0.0193 vocab_size 100053 (6.92s/11.94m) [Shuffling 25] loss 3924361.2500 lr 0.0190 vocab_size 100053 (6.98s/11.93m) [Shuffling 26] loss 3977266.7500 lr 0.0188 vocab_size 100053 (7.06s/11.94m) [Shuffling 27] loss 3793319.0000 lr 0.0185 vocab_size 100053 (7.10s/11.94m) [Shuffling 28] loss 3932350.0000 lr 0.0183 vocab_size 100053 (7.05s/11.94m) [Shuffling 29] loss 4014094.0000 lr 0.0180 vocab_size 100053 (7.03s/11.95m) [Shuffling 30] loss 3922998.2500 lr 0.0178 vocab_size 100053 (7.02s/11.95m) [Shuffling 31] loss 3806575.0000 lr 0.0175 vocab_size 100053 (7.00s/11.94m) [Shuffling 32] loss 4037771.7500 lr 0.0173 vocab_size 100053 (7.04s/11.94m) [Shuffling 33] loss 3879443.5000 lr 0.0170 vocab_size 100053 (7.06s/11.95m) [Shuffling 34] loss 3992922.5000 lr 0.0168 vocab_size 100053 (7.05s/11.95m) [Shuffling 35] loss 3992278.7500 lr 0.0165 vocab_size 100053 (7.05s/11.95m) [Shuffling 36] loss 3944166.2500 lr 0.0163 vocab_size 100053 (7.03s/11.95m) [Shuffling 37] loss 3880962.7500 lr 0.0160 vocab_size 100053 (7.04s/11.95m) [Shuffling 38] loss 3811588.2500 lr 0.0158 vocab_size 100053 (7.06s/11.95m) [Shuffling 39] loss 3908887.7500 lr 0.0155 vocab_size 100053 (7.08s/11.96m) [Shuffling 40] loss 3898892.5000 lr 0.0153 vocab_size 100053 (7.06s/11.96m) [Shuffling 41] loss 3776024.2500 lr 0.0150 vocab_size 100053 (7.16s/11.96m) [Shuffling 42] loss 4039525.0000 lr 0.0148 vocab_size 100053 (7.00s/11.96m) [Shuffling 43] loss 3876917.5000 lr 0.0145 vocab_size 100053 (6.98s/11.96m) [Shuffling 44] loss 4009974.0000 lr 0.0143 vocab_size 100053 (7.00s/11.96m) [Shuffling 45] loss 3903532.2500 lr 0.0140 vocab_size 100053 (6.98s/11.96m) [Shuffling 46] loss 3862624.7500 lr 0.0138 vocab_size 100053 (6.99s/11.96m) [Shuffling 47] loss 3886221.0000 lr 0.0135 vocab_size 100053 (6.97s/11.95m) [Shuffling 48] loss 3868012.5000 lr 0.0133 vocab_size 100053 (7.02s/11.95m) [Shuffling 49] loss 3846042.7500 lr 0.0130 vocab_size 100053 (7.00s/11.95m) [Shuffling 50] loss 3885358.5000 lr 0.0128 vocab_size 100053 (7.07s/11.96m) [Shuffling 51] loss 3883760.5000 lr 0.0126 vocab_size 100053 (7.01s/11.95m) [Shuffling 52] loss 3761144.2500 lr 0.0123 vocab_size 100053 (7.17s/11.96m) [Shuffling 53] loss 3772250.0000 lr 0.0121 vocab_size 100053 (7.06s/11.96m) [Shuffling 54] loss 3888863.5000 lr 0.0118 vocab_size 100053 (6.99s/11.96m) [Shuffling 55] loss 3902742.7500 lr 0.0116 vocab_size 100053 (7.06s/11.96m) [Shuffling 56] loss 3890277.2500 lr 0.0113 vocab_size 100053 (7.04s/11.96m) [Shuffling 57] loss 3961499.2500 lr 0.0111 vocab_size 100053 (7.05s/11.96m) [Shuffling 58] loss 4009450.5000 lr 0.0108 vocab_size 100053 (7.09s/11.96m) [Shuffling 59] loss 3806982.0000 lr 0.0106 vocab_size 100053 (7.05s/11.96m) [Shuffling 60] loss 3971790.7500 lr 0.0103 vocab_size 100053 (7.05s/11.96m) [Shuffling 61] loss 3865198.0000 lr 0.0101 vocab_size 100053 (7.02s/11.96m) [Shuffling 62] loss 3896561.5000 lr 0.0098 vocab_size 100053 (7.07s/11.96m) [Shuffling 63] loss 3936805.5000 lr 0.0096 vocab_size 100053 (7.04s/11.96m) [Shuffling 64] loss 3876661.0000 lr 0.0093 vocab_size 100053 (7.07s/11.97m) [Shuffling 65] loss 3823289.0000 lr 0.0091 vocab_size 100053 (7.05s/11.97m) [Shuffling 66] loss 3901687.2500 lr 0.0088 vocab_size 100053 (7.01s/11.97m) [Shuffling 67] loss 3953235.5000 lr 0.0086 vocab_size 100053 (7.04s/11.97m) [Shuffling 68] loss 3938194.2500 lr 0.0083 vocab_size 100053 (6.96s/11.96m) [Shuffling 69] loss 3897798.0000 lr 0.0081 vocab_size 100053 (7.04s/11.96m) [Shuffling 70] loss 3815985.7500 lr 0.0078 vocab_size 100053 (7.07s/11.97m) [Shuffling 71] loss 3792497.5000 lr 0.0076 vocab_size 100053 (7.06s/11.97m) [Shuffling 72] loss 3953292.7500 lr 0.0073 vocab_size 100053 (7.07s/11.97m) [Shuffling 73] loss 3934350.0000 lr 0.0071 vocab_size 100053 (7.07s/11.97m) [Shuffling 74] loss 3987114.5000 lr 0.0068 vocab_size 100053 (7.08s/11.97m) [Shuffling 75] loss 3912823.0000 lr 0.0066 vocab_size 100053 (7.11s/11.97m) [Shuffling 76] loss 3959277.5000 lr 0.0063 vocab_size 100053 (7.06s/11.97m) [Shuffling 77] loss 3860598.2500 lr 0.0061 vocab_size 100053 (7.03s/11.97m) [Shuffling 78] loss 3880392.5000 lr 0.0058 vocab_size 100053 (7.01s/11.97m) [Shuffling 79] loss 3991691.2500 lr 0.0056 vocab_size 100053 (7.09s/11.97m) [Shuffling 80] loss 3721745.5000 lr 0.0053 vocab_size 100053 (7.10s/11.97m) [Shuffling 81] loss 3862700.2500 lr 0.0051 vocab_size 100053 (7.02s/11.97m) [Shuffling 82] loss 3949980.0000 lr 0.0048 vocab_size 100053 (7.06s/11.97m) [Shuffling 83] loss 3871618.2500 lr 0.0046 vocab_size 100053 (7.00s/11.97m) [Shuffling 84] loss 3893291.2500 lr 0.0043 vocab_size 100053 (7.00s/11.97m) [Shuffling 85] loss 4042450.5000 lr 0.0041 vocab_size 100053 (7.07s/11.97m) [Shuffling 86] loss 3869954.5000 lr 0.0038 vocab_size 100053 (7.00s/11.97m) [Shuffling 87] loss 3832285.2500 lr 0.0036 vocab_size 100053 (6.99s/11.97m) [Shuffling 88] loss 3892947.0000 lr 0.0033 vocab_size 100053 (7.02s/11.97m) [Shuffling 89] loss 3878264.7500 lr 0.0031 vocab_size 100053 (7.01s/11.97m) [Shuffling 90] loss 3805779.5000 lr 0.0028 vocab_size 100053 (6.95s/11.97m) [Shuffling 91] loss 3934983.7500 lr 0.0026 vocab_size 100053 (7.04s/11.97m) [Shuffling 92] loss 3841357.2500 lr 0.0023 vocab_size 100053 (6.99s/11.97m) [Shuffling 93] loss 3771168.2500 lr 0.0021 vocab_size 100053 (7.00s/11.97m) [Shuffling 94] loss 3695426.0000 lr 0.0018 vocab_size 100053 (7.12s/11.97m) [Shuffling 95] loss 3946551.2500 lr 0.0016 vocab_size 100053 (6.90s/11.97m) [Shuffling 96] loss 3943549.7500 lr 0.0013 vocab_size 100053 (6.96s/11.97m) [Shuffling 97] loss 3945654.2500 lr 0.0011 vocab_size 100053 (9.93s/12.02m) [Shuffling 98] loss 3913285.2500 lr 0.0008 vocab_size 100053 (6.97s/12.01m) [Shuffling 99] loss 3771480.2500 lr 0.0006 vocab_size 100053 (9.98s/12.06m) [Shuffling 100] loss 3888482.7500 lr 0.0003 vocab_size 100053 (7.82s/12.08m) [03/12/24-17:43:35] Training finished, training Time 12.08m