File size: 24,165 Bytes
f70f6c9 |
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 |
2023-10-19 10:38:20,282 ----------------------------------------------------------------------------------------------------
2023-10-19 10:38:20,282 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(32001, 128)
(position_embeddings): Embedding(512, 128)
(token_type_embeddings): Embedding(2, 128)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): BertEncoder(
(layer): ModuleList(
(0-1): 2 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=128, out_features=128, bias=True)
(key): Linear(in_features=128, out_features=128, bias=True)
(value): Linear(in_features=128, out_features=128, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=128, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=128, out_features=512, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=512, out_features=128, bias=True)
(LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=128, out_features=128, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=128, out_features=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-19 10:38:20,282 ----------------------------------------------------------------------------------------------------
2023-10-19 10:38:20,282 MultiCorpus: 20847 train + 1123 dev + 3350 test sentences
- NER_HIPE_2022 Corpus: 20847 train + 1123 dev + 3350 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/de/with_doc_seperator
2023-10-19 10:38:20,282 ----------------------------------------------------------------------------------------------------
2023-10-19 10:38:20,282 Train: 20847 sentences
2023-10-19 10:38:20,282 (train_with_dev=False, train_with_test=False)
2023-10-19 10:38:20,282 ----------------------------------------------------------------------------------------------------
2023-10-19 10:38:20,282 Training Params:
2023-10-19 10:38:20,282 - learning_rate: "3e-05"
2023-10-19 10:38:20,283 - mini_batch_size: "4"
2023-10-19 10:38:20,283 - max_epochs: "10"
2023-10-19 10:38:20,283 - shuffle: "True"
2023-10-19 10:38:20,283 ----------------------------------------------------------------------------------------------------
2023-10-19 10:38:20,283 Plugins:
2023-10-19 10:38:20,283 - TensorboardLogger
2023-10-19 10:38:20,283 - LinearScheduler | warmup_fraction: '0.1'
2023-10-19 10:38:20,283 ----------------------------------------------------------------------------------------------------
2023-10-19 10:38:20,283 Final evaluation on model from best epoch (best-model.pt)
2023-10-19 10:38:20,283 - metric: "('micro avg', 'f1-score')"
2023-10-19 10:38:20,283 ----------------------------------------------------------------------------------------------------
2023-10-19 10:38:20,283 Computation:
2023-10-19 10:38:20,283 - compute on device: cuda:0
2023-10-19 10:38:20,283 - embedding storage: none
2023-10-19 10:38:20,283 ----------------------------------------------------------------------------------------------------
2023-10-19 10:38:20,283 Model training base path: "hmbench-newseye/de-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-19 10:38:20,283 ----------------------------------------------------------------------------------------------------
2023-10-19 10:38:20,283 ----------------------------------------------------------------------------------------------------
2023-10-19 10:38:20,283 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-19 10:38:29,211 epoch 1 - iter 521/5212 - loss 2.72371222 - time (sec): 8.93 - samples/sec: 4096.81 - lr: 0.000003 - momentum: 0.000000
2023-10-19 10:38:37,636 epoch 1 - iter 1042/5212 - loss 2.05502782 - time (sec): 17.35 - samples/sec: 4193.38 - lr: 0.000006 - momentum: 0.000000
2023-10-19 10:38:45,775 epoch 1 - iter 1563/5212 - loss 1.58883158 - time (sec): 25.49 - samples/sec: 4304.23 - lr: 0.000009 - momentum: 0.000000
2023-10-19 10:38:54,422 epoch 1 - iter 2084/5212 - loss 1.32512439 - time (sec): 34.14 - samples/sec: 4326.46 - lr: 0.000012 - momentum: 0.000000
2023-10-19 10:39:02,797 epoch 1 - iter 2605/5212 - loss 1.17407642 - time (sec): 42.51 - samples/sec: 4404.91 - lr: 0.000015 - momentum: 0.000000
2023-10-19 10:39:11,106 epoch 1 - iter 3126/5212 - loss 1.08043972 - time (sec): 50.82 - samples/sec: 4390.07 - lr: 0.000018 - momentum: 0.000000
2023-10-19 10:39:19,406 epoch 1 - iter 3647/5212 - loss 1.01154260 - time (sec): 59.12 - samples/sec: 4394.12 - lr: 0.000021 - momentum: 0.000000
2023-10-19 10:39:27,380 epoch 1 - iter 4168/5212 - loss 0.94868657 - time (sec): 67.10 - samples/sec: 4397.60 - lr: 0.000024 - momentum: 0.000000
2023-10-19 10:39:35,900 epoch 1 - iter 4689/5212 - loss 0.88980347 - time (sec): 75.62 - samples/sec: 4380.26 - lr: 0.000027 - momentum: 0.000000
2023-10-19 10:39:44,202 epoch 1 - iter 5210/5212 - loss 0.84260434 - time (sec): 83.92 - samples/sec: 4377.91 - lr: 0.000030 - momentum: 0.000000
2023-10-19 10:39:44,234 ----------------------------------------------------------------------------------------------------
2023-10-19 10:39:44,234 EPOCH 1 done: loss 0.8426 - lr: 0.000030
2023-10-19 10:39:46,475 DEV : loss 0.1433752328157425 - f1-score (micro avg) 0.0291
2023-10-19 10:39:46,497 saving best model
2023-10-19 10:39:46,526 ----------------------------------------------------------------------------------------------------
2023-10-19 10:39:54,611 epoch 2 - iter 521/5212 - loss 0.40879602 - time (sec): 8.09 - samples/sec: 4230.25 - lr: 0.000030 - momentum: 0.000000
2023-10-19 10:40:02,957 epoch 2 - iter 1042/5212 - loss 0.41373491 - time (sec): 16.43 - samples/sec: 4320.03 - lr: 0.000029 - momentum: 0.000000
2023-10-19 10:40:11,068 epoch 2 - iter 1563/5212 - loss 0.41110618 - time (sec): 24.54 - samples/sec: 4287.77 - lr: 0.000029 - momentum: 0.000000
2023-10-19 10:40:19,346 epoch 2 - iter 2084/5212 - loss 0.39790099 - time (sec): 32.82 - samples/sec: 4380.84 - lr: 0.000029 - momentum: 0.000000
2023-10-19 10:40:27,777 epoch 2 - iter 2605/5212 - loss 0.39498142 - time (sec): 41.25 - samples/sec: 4415.85 - lr: 0.000028 - momentum: 0.000000
2023-10-19 10:40:36,056 epoch 2 - iter 3126/5212 - loss 0.38607733 - time (sec): 49.53 - samples/sec: 4425.77 - lr: 0.000028 - momentum: 0.000000
2023-10-19 10:40:44,415 epoch 2 - iter 3647/5212 - loss 0.37832413 - time (sec): 57.89 - samples/sec: 4438.26 - lr: 0.000028 - momentum: 0.000000
2023-10-19 10:40:53,047 epoch 2 - iter 4168/5212 - loss 0.37306540 - time (sec): 66.52 - samples/sec: 4423.81 - lr: 0.000027 - momentum: 0.000000
2023-10-19 10:41:01,276 epoch 2 - iter 4689/5212 - loss 0.36784622 - time (sec): 74.75 - samples/sec: 4421.81 - lr: 0.000027 - momentum: 0.000000
2023-10-19 10:41:09,649 epoch 2 - iter 5210/5212 - loss 0.36564089 - time (sec): 83.12 - samples/sec: 4419.15 - lr: 0.000027 - momentum: 0.000000
2023-10-19 10:41:09,684 ----------------------------------------------------------------------------------------------------
2023-10-19 10:41:09,684 EPOCH 2 done: loss 0.3656 - lr: 0.000027
2023-10-19 10:41:14,780 DEV : loss 0.1391393542289734 - f1-score (micro avg) 0.2928
2023-10-19 10:41:14,804 saving best model
2023-10-19 10:41:14,840 ----------------------------------------------------------------------------------------------------
2023-10-19 10:41:23,076 epoch 3 - iter 521/5212 - loss 0.32466550 - time (sec): 8.24 - samples/sec: 4448.41 - lr: 0.000026 - momentum: 0.000000
2023-10-19 10:41:31,416 epoch 3 - iter 1042/5212 - loss 0.31821062 - time (sec): 16.58 - samples/sec: 4451.66 - lr: 0.000026 - momentum: 0.000000
2023-10-19 10:41:39,462 epoch 3 - iter 1563/5212 - loss 0.32728696 - time (sec): 24.62 - samples/sec: 4451.18 - lr: 0.000026 - momentum: 0.000000
2023-10-19 10:41:47,786 epoch 3 - iter 2084/5212 - loss 0.31607836 - time (sec): 32.95 - samples/sec: 4487.01 - lr: 0.000025 - momentum: 0.000000
2023-10-19 10:41:56,154 epoch 3 - iter 2605/5212 - loss 0.31480478 - time (sec): 41.31 - samples/sec: 4494.25 - lr: 0.000025 - momentum: 0.000000
2023-10-19 10:42:04,737 epoch 3 - iter 3126/5212 - loss 0.31038250 - time (sec): 49.90 - samples/sec: 4473.56 - lr: 0.000025 - momentum: 0.000000
2023-10-19 10:42:12,965 epoch 3 - iter 3647/5212 - loss 0.31026995 - time (sec): 58.12 - samples/sec: 4453.95 - lr: 0.000024 - momentum: 0.000000
2023-10-19 10:42:21,291 epoch 3 - iter 4168/5212 - loss 0.31015311 - time (sec): 66.45 - samples/sec: 4444.22 - lr: 0.000024 - momentum: 0.000000
2023-10-19 10:42:29,486 epoch 3 - iter 4689/5212 - loss 0.31207638 - time (sec): 74.65 - samples/sec: 4429.21 - lr: 0.000024 - momentum: 0.000000
2023-10-19 10:42:37,697 epoch 3 - iter 5210/5212 - loss 0.31193129 - time (sec): 82.86 - samples/sec: 4433.61 - lr: 0.000023 - momentum: 0.000000
2023-10-19 10:42:37,733 ----------------------------------------------------------------------------------------------------
2023-10-19 10:42:37,733 EPOCH 3 done: loss 0.3119 - lr: 0.000023
2023-10-19 10:42:42,852 DEV : loss 0.13720223307609558 - f1-score (micro avg) 0.311
2023-10-19 10:42:42,876 saving best model
2023-10-19 10:42:42,916 ----------------------------------------------------------------------------------------------------
2023-10-19 10:42:51,333 epoch 4 - iter 521/5212 - loss 0.25709768 - time (sec): 8.42 - samples/sec: 4514.44 - lr: 0.000023 - momentum: 0.000000
2023-10-19 10:42:59,518 epoch 4 - iter 1042/5212 - loss 0.26137660 - time (sec): 16.60 - samples/sec: 4367.48 - lr: 0.000023 - momentum: 0.000000
2023-10-19 10:43:07,636 epoch 4 - iter 1563/5212 - loss 0.27629460 - time (sec): 24.72 - samples/sec: 4311.54 - lr: 0.000022 - momentum: 0.000000
2023-10-19 10:43:15,953 epoch 4 - iter 2084/5212 - loss 0.27504062 - time (sec): 33.04 - samples/sec: 4377.33 - lr: 0.000022 - momentum: 0.000000
2023-10-19 10:43:24,401 epoch 4 - iter 2605/5212 - loss 0.27571513 - time (sec): 41.48 - samples/sec: 4448.97 - lr: 0.000022 - momentum: 0.000000
2023-10-19 10:43:32,741 epoch 4 - iter 3126/5212 - loss 0.27824346 - time (sec): 49.82 - samples/sec: 4441.02 - lr: 0.000021 - momentum: 0.000000
2023-10-19 10:43:41,037 epoch 4 - iter 3647/5212 - loss 0.27812047 - time (sec): 58.12 - samples/sec: 4436.80 - lr: 0.000021 - momentum: 0.000000
2023-10-19 10:43:49,174 epoch 4 - iter 4168/5212 - loss 0.28260826 - time (sec): 66.26 - samples/sec: 4408.59 - lr: 0.000021 - momentum: 0.000000
2023-10-19 10:43:57,417 epoch 4 - iter 4689/5212 - loss 0.28076208 - time (sec): 74.50 - samples/sec: 4423.38 - lr: 0.000020 - momentum: 0.000000
2023-10-19 10:44:05,823 epoch 4 - iter 5210/5212 - loss 0.27822833 - time (sec): 82.91 - samples/sec: 4430.02 - lr: 0.000020 - momentum: 0.000000
2023-10-19 10:44:05,856 ----------------------------------------------------------------------------------------------------
2023-10-19 10:44:05,856 EPOCH 4 done: loss 0.2782 - lr: 0.000020
2023-10-19 10:44:11,009 DEV : loss 0.14805111289024353 - f1-score (micro avg) 0.2656
2023-10-19 10:44:11,033 ----------------------------------------------------------------------------------------------------
2023-10-19 10:44:19,265 epoch 5 - iter 521/5212 - loss 0.25458992 - time (sec): 8.23 - samples/sec: 4663.96 - lr: 0.000020 - momentum: 0.000000
2023-10-19 10:44:27,618 epoch 5 - iter 1042/5212 - loss 0.23812930 - time (sec): 16.58 - samples/sec: 4645.30 - lr: 0.000019 - momentum: 0.000000
2023-10-19 10:44:35,825 epoch 5 - iter 1563/5212 - loss 0.23909797 - time (sec): 24.79 - samples/sec: 4518.74 - lr: 0.000019 - momentum: 0.000000
2023-10-19 10:44:44,069 epoch 5 - iter 2084/5212 - loss 0.24733360 - time (sec): 33.04 - samples/sec: 4498.14 - lr: 0.000019 - momentum: 0.000000
2023-10-19 10:44:52,390 epoch 5 - iter 2605/5212 - loss 0.24607385 - time (sec): 41.36 - samples/sec: 4479.65 - lr: 0.000018 - momentum: 0.000000
2023-10-19 10:45:00,692 epoch 5 - iter 3126/5212 - loss 0.25227478 - time (sec): 49.66 - samples/sec: 4455.06 - lr: 0.000018 - momentum: 0.000000
2023-10-19 10:45:08,913 epoch 5 - iter 3647/5212 - loss 0.25331088 - time (sec): 57.88 - samples/sec: 4439.80 - lr: 0.000018 - momentum: 0.000000
2023-10-19 10:45:17,372 epoch 5 - iter 4168/5212 - loss 0.25226388 - time (sec): 66.34 - samples/sec: 4444.21 - lr: 0.000017 - momentum: 0.000000
2023-10-19 10:45:25,748 epoch 5 - iter 4689/5212 - loss 0.25437720 - time (sec): 74.71 - samples/sec: 4442.17 - lr: 0.000017 - momentum: 0.000000
2023-10-19 10:45:34,004 epoch 5 - iter 5210/5212 - loss 0.25346983 - time (sec): 82.97 - samples/sec: 4428.05 - lr: 0.000017 - momentum: 0.000000
2023-10-19 10:45:34,030 ----------------------------------------------------------------------------------------------------
2023-10-19 10:45:34,030 EPOCH 5 done: loss 0.2535 - lr: 0.000017
2023-10-19 10:45:39,162 DEV : loss 0.1490069180727005 - f1-score (micro avg) 0.2855
2023-10-19 10:45:39,197 ----------------------------------------------------------------------------------------------------
2023-10-19 10:45:47,655 epoch 6 - iter 521/5212 - loss 0.26624853 - time (sec): 8.46 - samples/sec: 3991.54 - lr: 0.000016 - momentum: 0.000000
2023-10-19 10:45:56,023 epoch 6 - iter 1042/5212 - loss 0.25453009 - time (sec): 16.82 - samples/sec: 4277.48 - lr: 0.000016 - momentum: 0.000000
2023-10-19 10:46:04,383 epoch 6 - iter 1563/5212 - loss 0.24457171 - time (sec): 25.18 - samples/sec: 4370.58 - lr: 0.000016 - momentum: 0.000000
2023-10-19 10:46:12,858 epoch 6 - iter 2084/5212 - loss 0.23560982 - time (sec): 33.66 - samples/sec: 4419.97 - lr: 0.000015 - momentum: 0.000000
2023-10-19 10:46:21,179 epoch 6 - iter 2605/5212 - loss 0.23329802 - time (sec): 41.98 - samples/sec: 4434.39 - lr: 0.000015 - momentum: 0.000000
2023-10-19 10:46:29,142 epoch 6 - iter 3126/5212 - loss 0.23211029 - time (sec): 49.94 - samples/sec: 4480.71 - lr: 0.000015 - momentum: 0.000000
2023-10-19 10:46:37,489 epoch 6 - iter 3647/5212 - loss 0.23718572 - time (sec): 58.29 - samples/sec: 4445.74 - lr: 0.000014 - momentum: 0.000000
2023-10-19 10:46:45,775 epoch 6 - iter 4168/5212 - loss 0.23673427 - time (sec): 66.58 - samples/sec: 4422.59 - lr: 0.000014 - momentum: 0.000000
2023-10-19 10:46:53,971 epoch 6 - iter 4689/5212 - loss 0.23213792 - time (sec): 74.77 - samples/sec: 4423.57 - lr: 0.000014 - momentum: 0.000000
2023-10-19 10:47:02,844 epoch 6 - iter 5210/5212 - loss 0.23638178 - time (sec): 83.65 - samples/sec: 4391.69 - lr: 0.000013 - momentum: 0.000000
2023-10-19 10:47:02,878 ----------------------------------------------------------------------------------------------------
2023-10-19 10:47:02,879 EPOCH 6 done: loss 0.2364 - lr: 0.000013
2023-10-19 10:47:07,421 DEV : loss 0.1647791564464569 - f1-score (micro avg) 0.2693
2023-10-19 10:47:07,444 ----------------------------------------------------------------------------------------------------
2023-10-19 10:47:15,620 epoch 7 - iter 521/5212 - loss 0.23817423 - time (sec): 8.18 - samples/sec: 4509.93 - lr: 0.000013 - momentum: 0.000000
2023-10-19 10:47:23,915 epoch 7 - iter 1042/5212 - loss 0.22435065 - time (sec): 16.47 - samples/sec: 4529.78 - lr: 0.000013 - momentum: 0.000000
2023-10-19 10:47:32,101 epoch 7 - iter 1563/5212 - loss 0.22347997 - time (sec): 24.66 - samples/sec: 4503.56 - lr: 0.000012 - momentum: 0.000000
2023-10-19 10:47:40,302 epoch 7 - iter 2084/5212 - loss 0.22299657 - time (sec): 32.86 - samples/sec: 4508.96 - lr: 0.000012 - momentum: 0.000000
2023-10-19 10:47:49,223 epoch 7 - iter 2605/5212 - loss 0.21747021 - time (sec): 41.78 - samples/sec: 4483.66 - lr: 0.000012 - momentum: 0.000000
2023-10-19 10:47:57,646 epoch 7 - iter 3126/5212 - loss 0.21824678 - time (sec): 50.20 - samples/sec: 4446.54 - lr: 0.000011 - momentum: 0.000000
2023-10-19 10:48:05,716 epoch 7 - iter 3647/5212 - loss 0.22182997 - time (sec): 58.27 - samples/sec: 4432.18 - lr: 0.000011 - momentum: 0.000000
2023-10-19 10:48:14,327 epoch 7 - iter 4168/5212 - loss 0.22075907 - time (sec): 66.88 - samples/sec: 4401.35 - lr: 0.000011 - momentum: 0.000000
2023-10-19 10:48:22,791 epoch 7 - iter 4689/5212 - loss 0.22067660 - time (sec): 75.35 - samples/sec: 4403.88 - lr: 0.000010 - momentum: 0.000000
2023-10-19 10:48:31,053 epoch 7 - iter 5210/5212 - loss 0.22219262 - time (sec): 83.61 - samples/sec: 4391.66 - lr: 0.000010 - momentum: 0.000000
2023-10-19 10:48:31,096 ----------------------------------------------------------------------------------------------------
2023-10-19 10:48:31,096 EPOCH 7 done: loss 0.2220 - lr: 0.000010
2023-10-19 10:48:35,618 DEV : loss 0.16794627904891968 - f1-score (micro avg) 0.2714
2023-10-19 10:48:35,641 ----------------------------------------------------------------------------------------------------
2023-10-19 10:48:44,018 epoch 8 - iter 521/5212 - loss 0.24024885 - time (sec): 8.38 - samples/sec: 4188.92 - lr: 0.000010 - momentum: 0.000000
2023-10-19 10:48:52,244 epoch 8 - iter 1042/5212 - loss 0.23380355 - time (sec): 16.60 - samples/sec: 4239.48 - lr: 0.000009 - momentum: 0.000000
2023-10-19 10:49:00,438 epoch 8 - iter 1563/5212 - loss 0.22497629 - time (sec): 24.80 - samples/sec: 4319.11 - lr: 0.000009 - momentum: 0.000000
2023-10-19 10:49:08,803 epoch 8 - iter 2084/5212 - loss 0.23065970 - time (sec): 33.16 - samples/sec: 4388.00 - lr: 0.000009 - momentum: 0.000000
2023-10-19 10:49:17,008 epoch 8 - iter 2605/5212 - loss 0.22269701 - time (sec): 41.37 - samples/sec: 4419.51 - lr: 0.000008 - momentum: 0.000000
2023-10-19 10:49:25,407 epoch 8 - iter 3126/5212 - loss 0.21867285 - time (sec): 49.77 - samples/sec: 4421.06 - lr: 0.000008 - momentum: 0.000000
2023-10-19 10:49:33,743 epoch 8 - iter 3647/5212 - loss 0.21580111 - time (sec): 58.10 - samples/sec: 4448.90 - lr: 0.000008 - momentum: 0.000000
2023-10-19 10:49:42,116 epoch 8 - iter 4168/5212 - loss 0.21326174 - time (sec): 66.47 - samples/sec: 4465.53 - lr: 0.000007 - momentum: 0.000000
2023-10-19 10:49:50,650 epoch 8 - iter 4689/5212 - loss 0.21523254 - time (sec): 75.01 - samples/sec: 4435.87 - lr: 0.000007 - momentum: 0.000000
2023-10-19 10:49:58,898 epoch 8 - iter 5210/5212 - loss 0.21610124 - time (sec): 83.26 - samples/sec: 4412.87 - lr: 0.000007 - momentum: 0.000000
2023-10-19 10:49:58,930 ----------------------------------------------------------------------------------------------------
2023-10-19 10:49:58,930 EPOCH 8 done: loss 0.2161 - lr: 0.000007
2023-10-19 10:50:04,152 DEV : loss 0.17187514901161194 - f1-score (micro avg) 0.266
2023-10-19 10:50:04,179 ----------------------------------------------------------------------------------------------------
2023-10-19 10:50:12,398 epoch 9 - iter 521/5212 - loss 0.21708792 - time (sec): 8.22 - samples/sec: 4102.45 - lr: 0.000006 - momentum: 0.000000
2023-10-19 10:50:20,747 epoch 9 - iter 1042/5212 - loss 0.19390674 - time (sec): 16.57 - samples/sec: 4312.08 - lr: 0.000006 - momentum: 0.000000
2023-10-19 10:50:29,009 epoch 9 - iter 1563/5212 - loss 0.20054262 - time (sec): 24.83 - samples/sec: 4341.33 - lr: 0.000006 - momentum: 0.000000
2023-10-19 10:50:37,304 epoch 9 - iter 2084/5212 - loss 0.21128490 - time (sec): 33.12 - samples/sec: 4338.21 - lr: 0.000005 - momentum: 0.000000
2023-10-19 10:50:45,779 epoch 9 - iter 2605/5212 - loss 0.21144254 - time (sec): 41.60 - samples/sec: 4433.69 - lr: 0.000005 - momentum: 0.000000
2023-10-19 10:50:54,001 epoch 9 - iter 3126/5212 - loss 0.20955816 - time (sec): 49.82 - samples/sec: 4423.37 - lr: 0.000005 - momentum: 0.000000
2023-10-19 10:51:02,390 epoch 9 - iter 3647/5212 - loss 0.21385600 - time (sec): 58.21 - samples/sec: 4438.87 - lr: 0.000004 - momentum: 0.000000
2023-10-19 10:51:10,644 epoch 9 - iter 4168/5212 - loss 0.21026236 - time (sec): 66.46 - samples/sec: 4440.93 - lr: 0.000004 - momentum: 0.000000
2023-10-19 10:51:19,014 epoch 9 - iter 4689/5212 - loss 0.20931466 - time (sec): 74.83 - samples/sec: 4411.30 - lr: 0.000004 - momentum: 0.000000
2023-10-19 10:51:27,385 epoch 9 - iter 5210/5212 - loss 0.20947495 - time (sec): 83.20 - samples/sec: 4414.82 - lr: 0.000003 - momentum: 0.000000
2023-10-19 10:51:27,418 ----------------------------------------------------------------------------------------------------
2023-10-19 10:51:27,418 EPOCH 9 done: loss 0.2094 - lr: 0.000003
2023-10-19 10:51:32,597 DEV : loss 0.1816394329071045 - f1-score (micro avg) 0.272
2023-10-19 10:51:32,621 ----------------------------------------------------------------------------------------------------
2023-10-19 10:51:41,163 epoch 10 - iter 521/5212 - loss 0.20993504 - time (sec): 8.54 - samples/sec: 4199.80 - lr: 0.000003 - momentum: 0.000000
2023-10-19 10:51:49,336 epoch 10 - iter 1042/5212 - loss 0.19721566 - time (sec): 16.71 - samples/sec: 4408.14 - lr: 0.000003 - momentum: 0.000000
2023-10-19 10:51:57,596 epoch 10 - iter 1563/5212 - loss 0.19886392 - time (sec): 24.97 - samples/sec: 4399.09 - lr: 0.000002 - momentum: 0.000000
2023-10-19 10:52:06,090 epoch 10 - iter 2084/5212 - loss 0.20201347 - time (sec): 33.47 - samples/sec: 4376.06 - lr: 0.000002 - momentum: 0.000000
2023-10-19 10:52:14,589 epoch 10 - iter 2605/5212 - loss 0.20351706 - time (sec): 41.97 - samples/sec: 4365.97 - lr: 0.000002 - momentum: 0.000000
2023-10-19 10:52:23,017 epoch 10 - iter 3126/5212 - loss 0.20748962 - time (sec): 50.40 - samples/sec: 4427.54 - lr: 0.000001 - momentum: 0.000000
2023-10-19 10:52:31,397 epoch 10 - iter 3647/5212 - loss 0.20967361 - time (sec): 58.78 - samples/sec: 4432.91 - lr: 0.000001 - momentum: 0.000000
2023-10-19 10:52:39,724 epoch 10 - iter 4168/5212 - loss 0.20825590 - time (sec): 67.10 - samples/sec: 4426.18 - lr: 0.000001 - momentum: 0.000000
2023-10-19 10:52:48,049 epoch 10 - iter 4689/5212 - loss 0.20604718 - time (sec): 75.43 - samples/sec: 4423.79 - lr: 0.000000 - momentum: 0.000000
2023-10-19 10:52:56,230 epoch 10 - iter 5210/5212 - loss 0.20716771 - time (sec): 83.61 - samples/sec: 4391.22 - lr: 0.000000 - momentum: 0.000000
2023-10-19 10:52:56,272 ----------------------------------------------------------------------------------------------------
2023-10-19 10:52:56,272 EPOCH 10 done: loss 0.2072 - lr: 0.000000
2023-10-19 10:53:01,417 DEV : loss 0.17811782658100128 - f1-score (micro avg) 0.2638
2023-10-19 10:53:01,469 ----------------------------------------------------------------------------------------------------
2023-10-19 10:53:01,470 Loading model from best epoch ...
2023-10-19 10:53:01,547 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-19 10:53:07,781
Results:
- F-score (micro) 0.299
- F-score (macro) 0.1455
- Accuracy 0.177
By class:
precision recall f1-score support
LOC 0.4586 0.4605 0.4595 1214
PER 0.1394 0.0718 0.0948 808
ORG 0.0470 0.0198 0.0279 353
HumanProd 0.0000 0.0000 0.0000 15
micro avg 0.3498 0.2611 0.2990 2390
macro avg 0.1612 0.1380 0.1455 2390
weighted avg 0.2870 0.2611 0.2696 2390
2023-10-19 10:53:07,782 ----------------------------------------------------------------------------------------------------
|