File size: 23,883 Bytes
bea00be |
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 |
2023-10-17 20:12:25,534 ----------------------------------------------------------------------------------------------------
2023-10-17 20:12:25,535 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): ElectraModel(
(embeddings): ElectraEmbeddings(
(word_embeddings): Embedding(32001, 768)
(position_embeddings): Embedding(512, 768)
(token_type_embeddings): Embedding(2, 768)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): ElectraEncoder(
(layer): ModuleList(
(0-11): 12 x ElectraLayer(
(attention): ElectraAttention(
(self): ElectraSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): ElectraSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): ElectraIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ElectraOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 20:12:25,535 ----------------------------------------------------------------------------------------------------
2023-10-17 20:12:25,535 MultiCorpus: 1085 train + 148 dev + 364 test sentences
- NER_HIPE_2022 Corpus: 1085 train + 148 dev + 364 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/sv/with_doc_seperator
2023-10-17 20:12:25,535 ----------------------------------------------------------------------------------------------------
2023-10-17 20:12:25,535 Train: 1085 sentences
2023-10-17 20:12:25,536 (train_with_dev=False, train_with_test=False)
2023-10-17 20:12:25,536 ----------------------------------------------------------------------------------------------------
2023-10-17 20:12:25,536 Training Params:
2023-10-17 20:12:25,536 - learning_rate: "3e-05"
2023-10-17 20:12:25,536 - mini_batch_size: "8"
2023-10-17 20:12:25,536 - max_epochs: "10"
2023-10-17 20:12:25,536 - shuffle: "True"
2023-10-17 20:12:25,536 ----------------------------------------------------------------------------------------------------
2023-10-17 20:12:25,536 Plugins:
2023-10-17 20:12:25,536 - TensorboardLogger
2023-10-17 20:12:25,536 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 20:12:25,536 ----------------------------------------------------------------------------------------------------
2023-10-17 20:12:25,536 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 20:12:25,536 - metric: "('micro avg', 'f1-score')"
2023-10-17 20:12:25,536 ----------------------------------------------------------------------------------------------------
2023-10-17 20:12:25,536 Computation:
2023-10-17 20:12:25,536 - compute on device: cuda:0
2023-10-17 20:12:25,536 - embedding storage: none
2023-10-17 20:12:25,536 ----------------------------------------------------------------------------------------------------
2023-10-17 20:12:25,536 Model training base path: "hmbench-newseye/sv-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-17 20:12:25,536 ----------------------------------------------------------------------------------------------------
2023-10-17 20:12:25,536 ----------------------------------------------------------------------------------------------------
2023-10-17 20:12:25,536 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 20:12:26,980 epoch 1 - iter 13/136 - loss 3.56469652 - time (sec): 1.44 - samples/sec: 3585.96 - lr: 0.000003 - momentum: 0.000000
2023-10-17 20:12:28,536 epoch 1 - iter 26/136 - loss 3.33018314 - time (sec): 3.00 - samples/sec: 3781.08 - lr: 0.000006 - momentum: 0.000000
2023-10-17 20:12:30,021 epoch 1 - iter 39/136 - loss 2.95543144 - time (sec): 4.48 - samples/sec: 3735.60 - lr: 0.000008 - momentum: 0.000000
2023-10-17 20:12:31,254 epoch 1 - iter 52/136 - loss 2.55098163 - time (sec): 5.72 - samples/sec: 3716.28 - lr: 0.000011 - momentum: 0.000000
2023-10-17 20:12:32,731 epoch 1 - iter 65/136 - loss 2.16561603 - time (sec): 7.19 - samples/sec: 3665.22 - lr: 0.000014 - momentum: 0.000000
2023-10-17 20:12:33,786 epoch 1 - iter 78/136 - loss 1.95630135 - time (sec): 8.25 - samples/sec: 3676.32 - lr: 0.000017 - momentum: 0.000000
2023-10-17 20:12:35,197 epoch 1 - iter 91/136 - loss 1.74967521 - time (sec): 9.66 - samples/sec: 3616.63 - lr: 0.000020 - momentum: 0.000000
2023-10-17 20:12:36,561 epoch 1 - iter 104/136 - loss 1.55037843 - time (sec): 11.02 - samples/sec: 3669.59 - lr: 0.000023 - momentum: 0.000000
2023-10-17 20:12:37,858 epoch 1 - iter 117/136 - loss 1.41500927 - time (sec): 12.32 - samples/sec: 3698.94 - lr: 0.000026 - momentum: 0.000000
2023-10-17 20:12:39,033 epoch 1 - iter 130/136 - loss 1.31279338 - time (sec): 13.50 - samples/sec: 3697.45 - lr: 0.000028 - momentum: 0.000000
2023-10-17 20:12:39,636 ----------------------------------------------------------------------------------------------------
2023-10-17 20:12:39,636 EPOCH 1 done: loss 1.2734 - lr: 0.000028
2023-10-17 20:12:40,860 DEV : loss 0.17920981347560883 - f1-score (micro avg) 0.6068
2023-10-17 20:12:40,865 saving best model
2023-10-17 20:12:41,222 ----------------------------------------------------------------------------------------------------
2023-10-17 20:12:42,459 epoch 2 - iter 13/136 - loss 0.23296681 - time (sec): 1.24 - samples/sec: 3765.35 - lr: 0.000030 - momentum: 0.000000
2023-10-17 20:12:43,820 epoch 2 - iter 26/136 - loss 0.24799180 - time (sec): 2.60 - samples/sec: 3732.19 - lr: 0.000029 - momentum: 0.000000
2023-10-17 20:12:44,999 epoch 2 - iter 39/136 - loss 0.22292966 - time (sec): 3.78 - samples/sec: 3845.54 - lr: 0.000029 - momentum: 0.000000
2023-10-17 20:12:46,415 epoch 2 - iter 52/136 - loss 0.21149373 - time (sec): 5.19 - samples/sec: 3724.35 - lr: 0.000029 - momentum: 0.000000
2023-10-17 20:12:47,777 epoch 2 - iter 65/136 - loss 0.20240517 - time (sec): 6.55 - samples/sec: 3700.29 - lr: 0.000028 - momentum: 0.000000
2023-10-17 20:12:49,399 epoch 2 - iter 78/136 - loss 0.19342818 - time (sec): 8.18 - samples/sec: 3600.69 - lr: 0.000028 - momentum: 0.000000
2023-10-17 20:12:50,745 epoch 2 - iter 91/136 - loss 0.18557542 - time (sec): 9.52 - samples/sec: 3600.10 - lr: 0.000028 - momentum: 0.000000
2023-10-17 20:12:52,371 epoch 2 - iter 104/136 - loss 0.18162443 - time (sec): 11.15 - samples/sec: 3578.27 - lr: 0.000027 - momentum: 0.000000
2023-10-17 20:12:53,841 epoch 2 - iter 117/136 - loss 0.17641577 - time (sec): 12.62 - samples/sec: 3601.85 - lr: 0.000027 - momentum: 0.000000
2023-10-17 20:12:55,198 epoch 2 - iter 130/136 - loss 0.16835610 - time (sec): 13.97 - samples/sec: 3576.06 - lr: 0.000027 - momentum: 0.000000
2023-10-17 20:12:55,859 ----------------------------------------------------------------------------------------------------
2023-10-17 20:12:55,860 EPOCH 2 done: loss 0.1664 - lr: 0.000027
2023-10-17 20:12:57,380 DEV : loss 0.12096702307462692 - f1-score (micro avg) 0.7178
2023-10-17 20:12:57,385 saving best model
2023-10-17 20:12:57,885 ----------------------------------------------------------------------------------------------------
2023-10-17 20:12:59,565 epoch 3 - iter 13/136 - loss 0.10688261 - time (sec): 1.68 - samples/sec: 2818.79 - lr: 0.000026 - momentum: 0.000000
2023-10-17 20:13:00,706 epoch 3 - iter 26/136 - loss 0.11348635 - time (sec): 2.82 - samples/sec: 3122.93 - lr: 0.000026 - momentum: 0.000000
2023-10-17 20:13:02,144 epoch 3 - iter 39/136 - loss 0.10946288 - time (sec): 4.26 - samples/sec: 3337.16 - lr: 0.000026 - momentum: 0.000000
2023-10-17 20:13:03,448 epoch 3 - iter 52/136 - loss 0.09875128 - time (sec): 5.56 - samples/sec: 3419.92 - lr: 0.000025 - momentum: 0.000000
2023-10-17 20:13:04,795 epoch 3 - iter 65/136 - loss 0.10238528 - time (sec): 6.91 - samples/sec: 3496.81 - lr: 0.000025 - momentum: 0.000000
2023-10-17 20:13:06,069 epoch 3 - iter 78/136 - loss 0.09890742 - time (sec): 8.18 - samples/sec: 3529.37 - lr: 0.000025 - momentum: 0.000000
2023-10-17 20:13:07,400 epoch 3 - iter 91/136 - loss 0.09927605 - time (sec): 9.51 - samples/sec: 3489.99 - lr: 0.000024 - momentum: 0.000000
2023-10-17 20:13:09,059 epoch 3 - iter 104/136 - loss 0.10213739 - time (sec): 11.17 - samples/sec: 3465.90 - lr: 0.000024 - momentum: 0.000000
2023-10-17 20:13:10,483 epoch 3 - iter 117/136 - loss 0.10125728 - time (sec): 12.60 - samples/sec: 3489.58 - lr: 0.000024 - momentum: 0.000000
2023-10-17 20:13:12,174 epoch 3 - iter 130/136 - loss 0.09993701 - time (sec): 14.29 - samples/sec: 3467.14 - lr: 0.000024 - momentum: 0.000000
2023-10-17 20:13:12,782 ----------------------------------------------------------------------------------------------------
2023-10-17 20:13:12,782 EPOCH 3 done: loss 0.0986 - lr: 0.000024
2023-10-17 20:13:14,255 DEV : loss 0.08591549098491669 - f1-score (micro avg) 0.7942
2023-10-17 20:13:14,260 saving best model
2023-10-17 20:13:14,727 ----------------------------------------------------------------------------------------------------
2023-10-17 20:13:15,913 epoch 4 - iter 13/136 - loss 0.06018926 - time (sec): 1.18 - samples/sec: 3558.55 - lr: 0.000023 - momentum: 0.000000
2023-10-17 20:13:17,244 epoch 4 - iter 26/136 - loss 0.06085278 - time (sec): 2.51 - samples/sec: 3578.74 - lr: 0.000023 - momentum: 0.000000
2023-10-17 20:13:18,689 epoch 4 - iter 39/136 - loss 0.06043327 - time (sec): 3.96 - samples/sec: 3479.26 - lr: 0.000022 - momentum: 0.000000
2023-10-17 20:13:20,225 epoch 4 - iter 52/136 - loss 0.06185287 - time (sec): 5.49 - samples/sec: 3426.97 - lr: 0.000022 - momentum: 0.000000
2023-10-17 20:13:21,401 epoch 4 - iter 65/136 - loss 0.05755276 - time (sec): 6.67 - samples/sec: 3537.79 - lr: 0.000022 - momentum: 0.000000
2023-10-17 20:13:23,133 epoch 4 - iter 78/136 - loss 0.06276857 - time (sec): 8.40 - samples/sec: 3485.61 - lr: 0.000021 - momentum: 0.000000
2023-10-17 20:13:24,317 epoch 4 - iter 91/136 - loss 0.06484688 - time (sec): 9.59 - samples/sec: 3540.53 - lr: 0.000021 - momentum: 0.000000
2023-10-17 20:13:25,962 epoch 4 - iter 104/136 - loss 0.06068370 - time (sec): 11.23 - samples/sec: 3532.08 - lr: 0.000021 - momentum: 0.000000
2023-10-17 20:13:27,379 epoch 4 - iter 117/136 - loss 0.06208566 - time (sec): 12.65 - samples/sec: 3578.80 - lr: 0.000021 - momentum: 0.000000
2023-10-17 20:13:28,583 epoch 4 - iter 130/136 - loss 0.06298638 - time (sec): 13.85 - samples/sec: 3603.24 - lr: 0.000020 - momentum: 0.000000
2023-10-17 20:13:29,189 ----------------------------------------------------------------------------------------------------
2023-10-17 20:13:29,189 EPOCH 4 done: loss 0.0615 - lr: 0.000020
2023-10-17 20:13:30,692 DEV : loss 0.10039487481117249 - f1-score (micro avg) 0.7971
2023-10-17 20:13:30,697 saving best model
2023-10-17 20:13:31,165 ----------------------------------------------------------------------------------------------------
2023-10-17 20:13:32,955 epoch 5 - iter 13/136 - loss 0.03987252 - time (sec): 1.79 - samples/sec: 3074.23 - lr: 0.000020 - momentum: 0.000000
2023-10-17 20:13:34,130 epoch 5 - iter 26/136 - loss 0.03912310 - time (sec): 2.96 - samples/sec: 3323.99 - lr: 0.000019 - momentum: 0.000000
2023-10-17 20:13:35,468 epoch 5 - iter 39/136 - loss 0.03863979 - time (sec): 4.30 - samples/sec: 3262.17 - lr: 0.000019 - momentum: 0.000000
2023-10-17 20:13:36,801 epoch 5 - iter 52/136 - loss 0.03622420 - time (sec): 5.63 - samples/sec: 3478.53 - lr: 0.000019 - momentum: 0.000000
2023-10-17 20:13:38,388 epoch 5 - iter 65/136 - loss 0.03387900 - time (sec): 7.22 - samples/sec: 3483.48 - lr: 0.000018 - momentum: 0.000000
2023-10-17 20:13:39,949 epoch 5 - iter 78/136 - loss 0.03469888 - time (sec): 8.78 - samples/sec: 3504.32 - lr: 0.000018 - momentum: 0.000000
2023-10-17 20:13:41,157 epoch 5 - iter 91/136 - loss 0.03515893 - time (sec): 9.99 - samples/sec: 3502.52 - lr: 0.000018 - momentum: 0.000000
2023-10-17 20:13:42,657 epoch 5 - iter 104/136 - loss 0.03745754 - time (sec): 11.49 - samples/sec: 3501.30 - lr: 0.000018 - momentum: 0.000000
2023-10-17 20:13:44,109 epoch 5 - iter 117/136 - loss 0.03876292 - time (sec): 12.94 - samples/sec: 3462.10 - lr: 0.000017 - momentum: 0.000000
2023-10-17 20:13:45,464 epoch 5 - iter 130/136 - loss 0.03993650 - time (sec): 14.30 - samples/sec: 3492.18 - lr: 0.000017 - momentum: 0.000000
2023-10-17 20:13:46,046 ----------------------------------------------------------------------------------------------------
2023-10-17 20:13:46,047 EPOCH 5 done: loss 0.0401 - lr: 0.000017
2023-10-17 20:13:47,576 DEV : loss 0.11180326342582703 - f1-score (micro avg) 0.7904
2023-10-17 20:13:47,582 ----------------------------------------------------------------------------------------------------
2023-10-17 20:13:49,187 epoch 6 - iter 13/136 - loss 0.02256178 - time (sec): 1.60 - samples/sec: 3203.98 - lr: 0.000016 - momentum: 0.000000
2023-10-17 20:13:50,516 epoch 6 - iter 26/136 - loss 0.02483432 - time (sec): 2.93 - samples/sec: 3436.08 - lr: 0.000016 - momentum: 0.000000
2023-10-17 20:13:51,729 epoch 6 - iter 39/136 - loss 0.02830385 - time (sec): 4.15 - samples/sec: 3482.11 - lr: 0.000016 - momentum: 0.000000
2023-10-17 20:13:53,135 epoch 6 - iter 52/136 - loss 0.02444290 - time (sec): 5.55 - samples/sec: 3479.67 - lr: 0.000015 - momentum: 0.000000
2023-10-17 20:13:54,664 epoch 6 - iter 65/136 - loss 0.02307868 - time (sec): 7.08 - samples/sec: 3515.74 - lr: 0.000015 - momentum: 0.000000
2023-10-17 20:13:55,765 epoch 6 - iter 78/136 - loss 0.02339547 - time (sec): 8.18 - samples/sec: 3541.86 - lr: 0.000015 - momentum: 0.000000
2023-10-17 20:13:57,053 epoch 6 - iter 91/136 - loss 0.02421288 - time (sec): 9.47 - samples/sec: 3540.14 - lr: 0.000015 - momentum: 0.000000
2023-10-17 20:13:58,360 epoch 6 - iter 104/136 - loss 0.02596671 - time (sec): 10.78 - samples/sec: 3563.87 - lr: 0.000014 - momentum: 0.000000
2023-10-17 20:13:59,837 epoch 6 - iter 117/136 - loss 0.02545975 - time (sec): 12.25 - samples/sec: 3577.26 - lr: 0.000014 - momentum: 0.000000
2023-10-17 20:14:01,410 epoch 6 - iter 130/136 - loss 0.02737841 - time (sec): 13.83 - samples/sec: 3566.42 - lr: 0.000014 - momentum: 0.000000
2023-10-17 20:14:02,079 ----------------------------------------------------------------------------------------------------
2023-10-17 20:14:02,080 EPOCH 6 done: loss 0.0269 - lr: 0.000014
2023-10-17 20:14:03,594 DEV : loss 0.11931055039167404 - f1-score (micro avg) 0.7906
2023-10-17 20:14:03,599 ----------------------------------------------------------------------------------------------------
2023-10-17 20:14:05,036 epoch 7 - iter 13/136 - loss 0.02728369 - time (sec): 1.44 - samples/sec: 3647.55 - lr: 0.000013 - momentum: 0.000000
2023-10-17 20:14:06,754 epoch 7 - iter 26/136 - loss 0.02508212 - time (sec): 3.15 - samples/sec: 3243.52 - lr: 0.000013 - momentum: 0.000000
2023-10-17 20:14:07,977 epoch 7 - iter 39/136 - loss 0.01944223 - time (sec): 4.38 - samples/sec: 3359.95 - lr: 0.000012 - momentum: 0.000000
2023-10-17 20:14:09,367 epoch 7 - iter 52/136 - loss 0.01891054 - time (sec): 5.77 - samples/sec: 3445.16 - lr: 0.000012 - momentum: 0.000000
2023-10-17 20:14:10,931 epoch 7 - iter 65/136 - loss 0.01886752 - time (sec): 7.33 - samples/sec: 3414.88 - lr: 0.000012 - momentum: 0.000000
2023-10-17 20:14:12,273 epoch 7 - iter 78/136 - loss 0.02600379 - time (sec): 8.67 - samples/sec: 3438.12 - lr: 0.000012 - momentum: 0.000000
2023-10-17 20:14:13,711 epoch 7 - iter 91/136 - loss 0.02405010 - time (sec): 10.11 - samples/sec: 3474.80 - lr: 0.000011 - momentum: 0.000000
2023-10-17 20:14:15,069 epoch 7 - iter 104/136 - loss 0.02314826 - time (sec): 11.47 - samples/sec: 3458.49 - lr: 0.000011 - momentum: 0.000000
2023-10-17 20:14:16,353 epoch 7 - iter 117/136 - loss 0.02220090 - time (sec): 12.75 - samples/sec: 3495.04 - lr: 0.000011 - momentum: 0.000000
2023-10-17 20:14:17,726 epoch 7 - iter 130/136 - loss 0.02185557 - time (sec): 14.13 - samples/sec: 3499.49 - lr: 0.000010 - momentum: 0.000000
2023-10-17 20:14:18,387 ----------------------------------------------------------------------------------------------------
2023-10-17 20:14:18,387 EPOCH 7 done: loss 0.0216 - lr: 0.000010
2023-10-17 20:14:19,971 DEV : loss 0.1356595754623413 - f1-score (micro avg) 0.7842
2023-10-17 20:14:19,978 ----------------------------------------------------------------------------------------------------
2023-10-17 20:14:21,419 epoch 8 - iter 13/136 - loss 0.01459122 - time (sec): 1.44 - samples/sec: 3223.63 - lr: 0.000010 - momentum: 0.000000
2023-10-17 20:14:22,884 epoch 8 - iter 26/136 - loss 0.01524438 - time (sec): 2.91 - samples/sec: 3345.87 - lr: 0.000009 - momentum: 0.000000
2023-10-17 20:14:24,288 epoch 8 - iter 39/136 - loss 0.01681237 - time (sec): 4.31 - samples/sec: 3506.61 - lr: 0.000009 - momentum: 0.000000
2023-10-17 20:14:25,632 epoch 8 - iter 52/136 - loss 0.01579062 - time (sec): 5.65 - samples/sec: 3488.23 - lr: 0.000009 - momentum: 0.000000
2023-10-17 20:14:27,289 epoch 8 - iter 65/136 - loss 0.01850053 - time (sec): 7.31 - samples/sec: 3476.24 - lr: 0.000009 - momentum: 0.000000
2023-10-17 20:14:28,856 epoch 8 - iter 78/136 - loss 0.01750239 - time (sec): 8.88 - samples/sec: 3400.75 - lr: 0.000008 - momentum: 0.000000
2023-10-17 20:14:30,174 epoch 8 - iter 91/136 - loss 0.01679539 - time (sec): 10.20 - samples/sec: 3458.71 - lr: 0.000008 - momentum: 0.000000
2023-10-17 20:14:31,621 epoch 8 - iter 104/136 - loss 0.01636561 - time (sec): 11.64 - samples/sec: 3460.03 - lr: 0.000008 - momentum: 0.000000
2023-10-17 20:14:32,839 epoch 8 - iter 117/136 - loss 0.01625463 - time (sec): 12.86 - samples/sec: 3488.55 - lr: 0.000007 - momentum: 0.000000
2023-10-17 20:14:34,305 epoch 8 - iter 130/136 - loss 0.01599673 - time (sec): 14.33 - samples/sec: 3475.87 - lr: 0.000007 - momentum: 0.000000
2023-10-17 20:14:34,871 ----------------------------------------------------------------------------------------------------
2023-10-17 20:14:34,872 EPOCH 8 done: loss 0.0155 - lr: 0.000007
2023-10-17 20:14:36,374 DEV : loss 0.13436543941497803 - f1-score (micro avg) 0.8051
2023-10-17 20:14:36,378 saving best model
2023-10-17 20:14:36,851 ----------------------------------------------------------------------------------------------------
2023-10-17 20:14:38,467 epoch 9 - iter 13/136 - loss 0.01017808 - time (sec): 1.61 - samples/sec: 3125.28 - lr: 0.000006 - momentum: 0.000000
2023-10-17 20:14:39,861 epoch 9 - iter 26/136 - loss 0.00781613 - time (sec): 3.01 - samples/sec: 3384.75 - lr: 0.000006 - momentum: 0.000000
2023-10-17 20:14:41,513 epoch 9 - iter 39/136 - loss 0.00860655 - time (sec): 4.66 - samples/sec: 3275.36 - lr: 0.000006 - momentum: 0.000000
2023-10-17 20:14:42,651 epoch 9 - iter 52/136 - loss 0.00830057 - time (sec): 5.80 - samples/sec: 3299.77 - lr: 0.000006 - momentum: 0.000000
2023-10-17 20:14:43,851 epoch 9 - iter 65/136 - loss 0.00782478 - time (sec): 7.00 - samples/sec: 3391.09 - lr: 0.000005 - momentum: 0.000000
2023-10-17 20:14:45,388 epoch 9 - iter 78/136 - loss 0.00919217 - time (sec): 8.53 - samples/sec: 3393.01 - lr: 0.000005 - momentum: 0.000000
2023-10-17 20:14:46,901 epoch 9 - iter 91/136 - loss 0.01018609 - time (sec): 10.05 - samples/sec: 3385.34 - lr: 0.000005 - momentum: 0.000000
2023-10-17 20:14:48,204 epoch 9 - iter 104/136 - loss 0.01072845 - time (sec): 11.35 - samples/sec: 3442.69 - lr: 0.000004 - momentum: 0.000000
2023-10-17 20:14:49,825 epoch 9 - iter 117/136 - loss 0.01148823 - time (sec): 12.97 - samples/sec: 3450.66 - lr: 0.000004 - momentum: 0.000000
2023-10-17 20:14:51,005 epoch 9 - iter 130/136 - loss 0.01159600 - time (sec): 14.15 - samples/sec: 3492.50 - lr: 0.000004 - momentum: 0.000000
2023-10-17 20:14:51,676 ----------------------------------------------------------------------------------------------------
2023-10-17 20:14:51,676 EPOCH 9 done: loss 0.0114 - lr: 0.000004
2023-10-17 20:14:53,187 DEV : loss 0.14552520215511322 - f1-score (micro avg) 0.8183
2023-10-17 20:14:53,191 saving best model
2023-10-17 20:14:53,662 ----------------------------------------------------------------------------------------------------
2023-10-17 20:14:54,950 epoch 10 - iter 13/136 - loss 0.01469245 - time (sec): 1.29 - samples/sec: 3230.38 - lr: 0.000003 - momentum: 0.000000
2023-10-17 20:14:56,089 epoch 10 - iter 26/136 - loss 0.02040672 - time (sec): 2.42 - samples/sec: 3368.12 - lr: 0.000003 - momentum: 0.000000
2023-10-17 20:14:57,468 epoch 10 - iter 39/136 - loss 0.01427909 - time (sec): 3.80 - samples/sec: 3452.11 - lr: 0.000003 - momentum: 0.000000
2023-10-17 20:14:58,855 epoch 10 - iter 52/136 - loss 0.01331876 - time (sec): 5.19 - samples/sec: 3514.99 - lr: 0.000002 - momentum: 0.000000
2023-10-17 20:15:00,113 epoch 10 - iter 65/136 - loss 0.01438903 - time (sec): 6.45 - samples/sec: 3491.61 - lr: 0.000002 - momentum: 0.000000
2023-10-17 20:15:01,548 epoch 10 - iter 78/136 - loss 0.01258656 - time (sec): 7.88 - samples/sec: 3591.51 - lr: 0.000002 - momentum: 0.000000
2023-10-17 20:15:03,037 epoch 10 - iter 91/136 - loss 0.01300637 - time (sec): 9.37 - samples/sec: 3576.70 - lr: 0.000001 - momentum: 0.000000
2023-10-17 20:15:04,424 epoch 10 - iter 104/136 - loss 0.01261802 - time (sec): 10.76 - samples/sec: 3624.70 - lr: 0.000001 - momentum: 0.000000
2023-10-17 20:15:05,828 epoch 10 - iter 117/136 - loss 0.01157527 - time (sec): 12.16 - samples/sec: 3630.39 - lr: 0.000001 - momentum: 0.000000
2023-10-17 20:15:07,223 epoch 10 - iter 130/136 - loss 0.01084597 - time (sec): 13.56 - samples/sec: 3632.39 - lr: 0.000000 - momentum: 0.000000
2023-10-17 20:15:08,084 ----------------------------------------------------------------------------------------------------
2023-10-17 20:15:08,084 EPOCH 10 done: loss 0.0107 - lr: 0.000000
2023-10-17 20:15:09,555 DEV : loss 0.15006117522716522 - f1-score (micro avg) 0.8132
2023-10-17 20:15:09,926 ----------------------------------------------------------------------------------------------------
2023-10-17 20:15:09,928 Loading model from best epoch ...
2023-10-17 20:15:11,517 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd, S-ORG, B-ORG, E-ORG, I-ORG
2023-10-17 20:15:13,740
Results:
- F-score (micro) 0.7926
- F-score (macro) 0.7586
- Accuracy 0.6716
By class:
precision recall f1-score support
LOC 0.8479 0.8397 0.8438 312
PER 0.7121 0.8798 0.7871 208
ORG 0.4667 0.5091 0.4870 55
HumanProd 0.8462 1.0000 0.9167 22
micro avg 0.7592 0.8291 0.7926 597
macro avg 0.7182 0.8072 0.7586 597
weighted avg 0.7654 0.8291 0.7939 597
2023-10-17 20:15:13,740 ----------------------------------------------------------------------------------------------------
|