File size: 23,958 Bytes
b4166d0 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 |
2023-10-16 20:21:36,765 ----------------------------------------------------------------------------------------------------
2023-10-16 20:21:36,766 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(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): BertEncoder(
(layer): ModuleList(
(0-11): 12 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(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): BertSelfOutput(
(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): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(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)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=768, out_features=768, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-16 20:21:36,766 ----------------------------------------------------------------------------------------------------
2023-10-16 20:21:36,767 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
2023-10-16 20:21:36,767 ----------------------------------------------------------------------------------------------------
2023-10-16 20:21:36,767 Train: 1085 sentences
2023-10-16 20:21:36,767 (train_with_dev=False, train_with_test=False)
2023-10-16 20:21:36,767 ----------------------------------------------------------------------------------------------------
2023-10-16 20:21:36,767 Training Params:
2023-10-16 20:21:36,767 - learning_rate: "3e-05"
2023-10-16 20:21:36,767 - mini_batch_size: "4"
2023-10-16 20:21:36,767 - max_epochs: "10"
2023-10-16 20:21:36,767 - shuffle: "True"
2023-10-16 20:21:36,767 ----------------------------------------------------------------------------------------------------
2023-10-16 20:21:36,767 Plugins:
2023-10-16 20:21:36,767 - LinearScheduler | warmup_fraction: '0.1'
2023-10-16 20:21:36,767 ----------------------------------------------------------------------------------------------------
2023-10-16 20:21:36,767 Final evaluation on model from best epoch (best-model.pt)
2023-10-16 20:21:36,767 - metric: "('micro avg', 'f1-score')"
2023-10-16 20:21:36,767 ----------------------------------------------------------------------------------------------------
2023-10-16 20:21:36,767 Computation:
2023-10-16 20:21:36,767 - compute on device: cuda:0
2023-10-16 20:21:36,767 - embedding storage: none
2023-10-16 20:21:36,767 ----------------------------------------------------------------------------------------------------
2023-10-16 20:21:36,767 Model training base path: "hmbench-newseye/sv-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-16 20:21:36,767 ----------------------------------------------------------------------------------------------------
2023-10-16 20:21:36,767 ----------------------------------------------------------------------------------------------------
2023-10-16 20:21:38,339 epoch 1 - iter 27/272 - loss 2.78403560 - time (sec): 1.57 - samples/sec: 3229.32 - lr: 0.000003 - momentum: 0.000000
2023-10-16 20:21:40,094 epoch 1 - iter 54/272 - loss 2.40057045 - time (sec): 3.33 - samples/sec: 3248.68 - lr: 0.000006 - momentum: 0.000000
2023-10-16 20:21:41,731 epoch 1 - iter 81/272 - loss 1.82624760 - time (sec): 4.96 - samples/sec: 3345.86 - lr: 0.000009 - momentum: 0.000000
2023-10-16 20:21:43,315 epoch 1 - iter 108/272 - loss 1.52858521 - time (sec): 6.55 - samples/sec: 3282.52 - lr: 0.000012 - momentum: 0.000000
2023-10-16 20:21:44,848 epoch 1 - iter 135/272 - loss 1.32380649 - time (sec): 8.08 - samples/sec: 3297.07 - lr: 0.000015 - momentum: 0.000000
2023-10-16 20:21:46,418 epoch 1 - iter 162/272 - loss 1.18197180 - time (sec): 9.65 - samples/sec: 3300.84 - lr: 0.000018 - momentum: 0.000000
2023-10-16 20:21:47,884 epoch 1 - iter 189/272 - loss 1.05227075 - time (sec): 11.12 - samples/sec: 3314.13 - lr: 0.000021 - momentum: 0.000000
2023-10-16 20:21:49,438 epoch 1 - iter 216/272 - loss 0.95821963 - time (sec): 12.67 - samples/sec: 3311.69 - lr: 0.000024 - momentum: 0.000000
2023-10-16 20:21:50,892 epoch 1 - iter 243/272 - loss 0.89248220 - time (sec): 14.12 - samples/sec: 3299.70 - lr: 0.000027 - momentum: 0.000000
2023-10-16 20:21:52,460 epoch 1 - iter 270/272 - loss 0.82752649 - time (sec): 15.69 - samples/sec: 3304.71 - lr: 0.000030 - momentum: 0.000000
2023-10-16 20:21:52,549 ----------------------------------------------------------------------------------------------------
2023-10-16 20:21:52,550 EPOCH 1 done: loss 0.8257 - lr: 0.000030
2023-10-16 20:21:53,648 DEV : loss 0.20883718132972717 - f1-score (micro avg) 0.5635
2023-10-16 20:21:53,653 saving best model
2023-10-16 20:21:54,009 ----------------------------------------------------------------------------------------------------
2023-10-16 20:21:55,510 epoch 2 - iter 27/272 - loss 0.23865211 - time (sec): 1.50 - samples/sec: 3012.03 - lr: 0.000030 - momentum: 0.000000
2023-10-16 20:21:57,145 epoch 2 - iter 54/272 - loss 0.21111499 - time (sec): 3.13 - samples/sec: 2990.38 - lr: 0.000029 - momentum: 0.000000
2023-10-16 20:21:58,855 epoch 2 - iter 81/272 - loss 0.18236592 - time (sec): 4.84 - samples/sec: 3134.51 - lr: 0.000029 - momentum: 0.000000
2023-10-16 20:22:00,503 epoch 2 - iter 108/272 - loss 0.16340264 - time (sec): 6.49 - samples/sec: 3219.69 - lr: 0.000029 - momentum: 0.000000
2023-10-16 20:22:01,918 epoch 2 - iter 135/272 - loss 0.17367697 - time (sec): 7.91 - samples/sec: 3226.50 - lr: 0.000028 - momentum: 0.000000
2023-10-16 20:22:03,448 epoch 2 - iter 162/272 - loss 0.16437164 - time (sec): 9.44 - samples/sec: 3253.57 - lr: 0.000028 - momentum: 0.000000
2023-10-16 20:22:05,013 epoch 2 - iter 189/272 - loss 0.16007430 - time (sec): 11.00 - samples/sec: 3234.67 - lr: 0.000028 - momentum: 0.000000
2023-10-16 20:22:06,521 epoch 2 - iter 216/272 - loss 0.15664843 - time (sec): 12.51 - samples/sec: 3289.42 - lr: 0.000027 - momentum: 0.000000
2023-10-16 20:22:08,169 epoch 2 - iter 243/272 - loss 0.16161312 - time (sec): 14.16 - samples/sec: 3267.43 - lr: 0.000027 - momentum: 0.000000
2023-10-16 20:22:09,760 epoch 2 - iter 270/272 - loss 0.15717506 - time (sec): 15.75 - samples/sec: 3288.18 - lr: 0.000027 - momentum: 0.000000
2023-10-16 20:22:09,865 ----------------------------------------------------------------------------------------------------
2023-10-16 20:22:09,865 EPOCH 2 done: loss 0.1566 - lr: 0.000027
2023-10-16 20:22:11,292 DEV : loss 0.11095353215932846 - f1-score (micro avg) 0.7437
2023-10-16 20:22:11,299 saving best model
2023-10-16 20:22:11,759 ----------------------------------------------------------------------------------------------------
2023-10-16 20:22:13,277 epoch 3 - iter 27/272 - loss 0.11749999 - time (sec): 1.52 - samples/sec: 2858.87 - lr: 0.000026 - momentum: 0.000000
2023-10-16 20:22:14,910 epoch 3 - iter 54/272 - loss 0.11281766 - time (sec): 3.15 - samples/sec: 3016.40 - lr: 0.000026 - momentum: 0.000000
2023-10-16 20:22:16,349 epoch 3 - iter 81/272 - loss 0.10816546 - time (sec): 4.59 - samples/sec: 3086.19 - lr: 0.000026 - momentum: 0.000000
2023-10-16 20:22:17,971 epoch 3 - iter 108/272 - loss 0.09957259 - time (sec): 6.21 - samples/sec: 3190.98 - lr: 0.000025 - momentum: 0.000000
2023-10-16 20:22:19,509 epoch 3 - iter 135/272 - loss 0.09598081 - time (sec): 7.75 - samples/sec: 3248.08 - lr: 0.000025 - momentum: 0.000000
2023-10-16 20:22:21,106 epoch 3 - iter 162/272 - loss 0.09855270 - time (sec): 9.35 - samples/sec: 3319.72 - lr: 0.000025 - momentum: 0.000000
2023-10-16 20:22:22,721 epoch 3 - iter 189/272 - loss 0.09542228 - time (sec): 10.96 - samples/sec: 3281.50 - lr: 0.000024 - momentum: 0.000000
2023-10-16 20:22:24,159 epoch 3 - iter 216/272 - loss 0.09125236 - time (sec): 12.40 - samples/sec: 3244.45 - lr: 0.000024 - momentum: 0.000000
2023-10-16 20:22:25,925 epoch 3 - iter 243/272 - loss 0.09163871 - time (sec): 14.16 - samples/sec: 3249.55 - lr: 0.000024 - momentum: 0.000000
2023-10-16 20:22:27,581 epoch 3 - iter 270/272 - loss 0.09515133 - time (sec): 15.82 - samples/sec: 3269.62 - lr: 0.000023 - momentum: 0.000000
2023-10-16 20:22:27,698 ----------------------------------------------------------------------------------------------------
2023-10-16 20:22:27,698 EPOCH 3 done: loss 0.0950 - lr: 0.000023
2023-10-16 20:22:29,168 DEV : loss 0.1084960475564003 - f1-score (micro avg) 0.7687
2023-10-16 20:22:29,173 saving best model
2023-10-16 20:22:29,660 ----------------------------------------------------------------------------------------------------
2023-10-16 20:22:31,232 epoch 4 - iter 27/272 - loss 0.08673095 - time (sec): 1.57 - samples/sec: 3367.90 - lr: 0.000023 - momentum: 0.000000
2023-10-16 20:22:32,904 epoch 4 - iter 54/272 - loss 0.06096268 - time (sec): 3.24 - samples/sec: 3366.04 - lr: 0.000023 - momentum: 0.000000
2023-10-16 20:22:34,929 epoch 4 - iter 81/272 - loss 0.05277234 - time (sec): 5.27 - samples/sec: 3191.79 - lr: 0.000022 - momentum: 0.000000
2023-10-16 20:22:36,440 epoch 4 - iter 108/272 - loss 0.05214582 - time (sec): 6.78 - samples/sec: 3170.65 - lr: 0.000022 - momentum: 0.000000
2023-10-16 20:22:37,846 epoch 4 - iter 135/272 - loss 0.05275227 - time (sec): 8.18 - samples/sec: 3140.46 - lr: 0.000022 - momentum: 0.000000
2023-10-16 20:22:39,337 epoch 4 - iter 162/272 - loss 0.05294019 - time (sec): 9.68 - samples/sec: 3194.65 - lr: 0.000021 - momentum: 0.000000
2023-10-16 20:22:40,847 epoch 4 - iter 189/272 - loss 0.05256511 - time (sec): 11.19 - samples/sec: 3191.38 - lr: 0.000021 - momentum: 0.000000
2023-10-16 20:22:42,403 epoch 4 - iter 216/272 - loss 0.05388890 - time (sec): 12.74 - samples/sec: 3187.35 - lr: 0.000021 - momentum: 0.000000
2023-10-16 20:22:43,889 epoch 4 - iter 243/272 - loss 0.05397102 - time (sec): 14.23 - samples/sec: 3216.73 - lr: 0.000020 - momentum: 0.000000
2023-10-16 20:22:45,507 epoch 4 - iter 270/272 - loss 0.05116201 - time (sec): 15.85 - samples/sec: 3274.71 - lr: 0.000020 - momentum: 0.000000
2023-10-16 20:22:45,590 ----------------------------------------------------------------------------------------------------
2023-10-16 20:22:45,590 EPOCH 4 done: loss 0.0511 - lr: 0.000020
2023-10-16 20:22:47,042 DEV : loss 0.11970058083534241 - f1-score (micro avg) 0.7956
2023-10-16 20:22:47,047 saving best model
2023-10-16 20:22:47,506 ----------------------------------------------------------------------------------------------------
2023-10-16 20:22:49,124 epoch 5 - iter 27/272 - loss 0.02705900 - time (sec): 1.62 - samples/sec: 3193.66 - lr: 0.000020 - momentum: 0.000000
2023-10-16 20:22:50,547 epoch 5 - iter 54/272 - loss 0.02506314 - time (sec): 3.04 - samples/sec: 3127.97 - lr: 0.000019 - momentum: 0.000000
2023-10-16 20:22:52,155 epoch 5 - iter 81/272 - loss 0.03031274 - time (sec): 4.65 - samples/sec: 3124.88 - lr: 0.000019 - momentum: 0.000000
2023-10-16 20:22:53,730 epoch 5 - iter 108/272 - loss 0.03097207 - time (sec): 6.22 - samples/sec: 3159.63 - lr: 0.000019 - momentum: 0.000000
2023-10-16 20:22:55,225 epoch 5 - iter 135/272 - loss 0.02985994 - time (sec): 7.72 - samples/sec: 3242.28 - lr: 0.000018 - momentum: 0.000000
2023-10-16 20:22:56,755 epoch 5 - iter 162/272 - loss 0.03208358 - time (sec): 9.25 - samples/sec: 3243.43 - lr: 0.000018 - momentum: 0.000000
2023-10-16 20:22:58,587 epoch 5 - iter 189/272 - loss 0.03465943 - time (sec): 11.08 - samples/sec: 3296.53 - lr: 0.000018 - momentum: 0.000000
2023-10-16 20:23:00,310 epoch 5 - iter 216/272 - loss 0.03639915 - time (sec): 12.80 - samples/sec: 3304.91 - lr: 0.000017 - momentum: 0.000000
2023-10-16 20:23:01,828 epoch 5 - iter 243/272 - loss 0.03542833 - time (sec): 14.32 - samples/sec: 3266.68 - lr: 0.000017 - momentum: 0.000000
2023-10-16 20:23:03,373 epoch 5 - iter 270/272 - loss 0.03755505 - time (sec): 15.87 - samples/sec: 3259.66 - lr: 0.000017 - momentum: 0.000000
2023-10-16 20:23:03,466 ----------------------------------------------------------------------------------------------------
2023-10-16 20:23:03,466 EPOCH 5 done: loss 0.0374 - lr: 0.000017
2023-10-16 20:23:04,927 DEV : loss 0.1268547922372818 - f1-score (micro avg) 0.7971
2023-10-16 20:23:04,931 saving best model
2023-10-16 20:23:05,417 ----------------------------------------------------------------------------------------------------
2023-10-16 20:23:07,117 epoch 6 - iter 27/272 - loss 0.00811967 - time (sec): 1.70 - samples/sec: 3380.93 - lr: 0.000016 - momentum: 0.000000
2023-10-16 20:23:08,639 epoch 6 - iter 54/272 - loss 0.01425739 - time (sec): 3.22 - samples/sec: 3222.56 - lr: 0.000016 - momentum: 0.000000
2023-10-16 20:23:10,127 epoch 6 - iter 81/272 - loss 0.01768857 - time (sec): 4.71 - samples/sec: 3212.83 - lr: 0.000016 - momentum: 0.000000
2023-10-16 20:23:11,760 epoch 6 - iter 108/272 - loss 0.02554340 - time (sec): 6.34 - samples/sec: 3228.81 - lr: 0.000015 - momentum: 0.000000
2023-10-16 20:23:13,459 epoch 6 - iter 135/272 - loss 0.02481300 - time (sec): 8.04 - samples/sec: 3216.75 - lr: 0.000015 - momentum: 0.000000
2023-10-16 20:23:14,910 epoch 6 - iter 162/272 - loss 0.02810161 - time (sec): 9.49 - samples/sec: 3189.64 - lr: 0.000015 - momentum: 0.000000
2023-10-16 20:23:16,351 epoch 6 - iter 189/272 - loss 0.02581547 - time (sec): 10.93 - samples/sec: 3166.18 - lr: 0.000014 - momentum: 0.000000
2023-10-16 20:23:17,969 epoch 6 - iter 216/272 - loss 0.02439654 - time (sec): 12.55 - samples/sec: 3191.87 - lr: 0.000014 - momentum: 0.000000
2023-10-16 20:23:19,576 epoch 6 - iter 243/272 - loss 0.02443384 - time (sec): 14.16 - samples/sec: 3219.27 - lr: 0.000014 - momentum: 0.000000
2023-10-16 20:23:21,284 epoch 6 - iter 270/272 - loss 0.02556831 - time (sec): 15.87 - samples/sec: 3253.43 - lr: 0.000013 - momentum: 0.000000
2023-10-16 20:23:21,383 ----------------------------------------------------------------------------------------------------
2023-10-16 20:23:21,383 EPOCH 6 done: loss 0.0254 - lr: 0.000013
2023-10-16 20:23:22,856 DEV : loss 0.15687650442123413 - f1-score (micro avg) 0.8
2023-10-16 20:23:22,860 saving best model
2023-10-16 20:23:23,331 ----------------------------------------------------------------------------------------------------
2023-10-16 20:23:25,063 epoch 7 - iter 27/272 - loss 0.01894596 - time (sec): 1.73 - samples/sec: 3546.75 - lr: 0.000013 - momentum: 0.000000
2023-10-16 20:23:26,630 epoch 7 - iter 54/272 - loss 0.01748641 - time (sec): 3.30 - samples/sec: 3316.07 - lr: 0.000013 - momentum: 0.000000
2023-10-16 20:23:28,014 epoch 7 - iter 81/272 - loss 0.02121757 - time (sec): 4.68 - samples/sec: 3177.52 - lr: 0.000012 - momentum: 0.000000
2023-10-16 20:23:29,587 epoch 7 - iter 108/272 - loss 0.02299563 - time (sec): 6.25 - samples/sec: 3206.07 - lr: 0.000012 - momentum: 0.000000
2023-10-16 20:23:31,102 epoch 7 - iter 135/272 - loss 0.02088311 - time (sec): 7.77 - samples/sec: 3207.48 - lr: 0.000012 - momentum: 0.000000
2023-10-16 20:23:32,693 epoch 7 - iter 162/272 - loss 0.01951963 - time (sec): 9.36 - samples/sec: 3182.35 - lr: 0.000011 - momentum: 0.000000
2023-10-16 20:23:34,322 epoch 7 - iter 189/272 - loss 0.01987080 - time (sec): 10.99 - samples/sec: 3220.45 - lr: 0.000011 - momentum: 0.000000
2023-10-16 20:23:36,323 epoch 7 - iter 216/272 - loss 0.02118412 - time (sec): 12.99 - samples/sec: 3197.06 - lr: 0.000011 - momentum: 0.000000
2023-10-16 20:23:37,912 epoch 7 - iter 243/272 - loss 0.02066351 - time (sec): 14.58 - samples/sec: 3201.54 - lr: 0.000010 - momentum: 0.000000
2023-10-16 20:23:39,484 epoch 7 - iter 270/272 - loss 0.02156439 - time (sec): 16.15 - samples/sec: 3210.50 - lr: 0.000010 - momentum: 0.000000
2023-10-16 20:23:39,569 ----------------------------------------------------------------------------------------------------
2023-10-16 20:23:39,569 EPOCH 7 done: loss 0.0215 - lr: 0.000010
2023-10-16 20:23:41,023 DEV : loss 0.14880990982055664 - f1-score (micro avg) 0.8177
2023-10-16 20:23:41,028 saving best model
2023-10-16 20:23:41,490 ----------------------------------------------------------------------------------------------------
2023-10-16 20:23:43,041 epoch 8 - iter 27/272 - loss 0.02885847 - time (sec): 1.55 - samples/sec: 3406.76 - lr: 0.000010 - momentum: 0.000000
2023-10-16 20:23:44,729 epoch 8 - iter 54/272 - loss 0.02025245 - time (sec): 3.24 - samples/sec: 3261.21 - lr: 0.000009 - momentum: 0.000000
2023-10-16 20:23:46,167 epoch 8 - iter 81/272 - loss 0.01914734 - time (sec): 4.68 - samples/sec: 3213.83 - lr: 0.000009 - momentum: 0.000000
2023-10-16 20:23:47,822 epoch 8 - iter 108/272 - loss 0.01626159 - time (sec): 6.33 - samples/sec: 3259.52 - lr: 0.000009 - momentum: 0.000000
2023-10-16 20:23:49,237 epoch 8 - iter 135/272 - loss 0.01407233 - time (sec): 7.75 - samples/sec: 3303.45 - lr: 0.000008 - momentum: 0.000000
2023-10-16 20:23:50,883 epoch 8 - iter 162/272 - loss 0.01494253 - time (sec): 9.39 - samples/sec: 3278.17 - lr: 0.000008 - momentum: 0.000000
2023-10-16 20:23:52,476 epoch 8 - iter 189/272 - loss 0.01366276 - time (sec): 10.99 - samples/sec: 3228.69 - lr: 0.000008 - momentum: 0.000000
2023-10-16 20:23:54,141 epoch 8 - iter 216/272 - loss 0.01370213 - time (sec): 12.65 - samples/sec: 3285.37 - lr: 0.000007 - momentum: 0.000000
2023-10-16 20:23:55,686 epoch 8 - iter 243/272 - loss 0.01409608 - time (sec): 14.19 - samples/sec: 3286.25 - lr: 0.000007 - momentum: 0.000000
2023-10-16 20:23:57,435 epoch 8 - iter 270/272 - loss 0.01362443 - time (sec): 15.94 - samples/sec: 3254.98 - lr: 0.000007 - momentum: 0.000000
2023-10-16 20:23:57,523 ----------------------------------------------------------------------------------------------------
2023-10-16 20:23:57,524 EPOCH 8 done: loss 0.0136 - lr: 0.000007
2023-10-16 20:23:58,981 DEV : loss 0.16685882210731506 - f1-score (micro avg) 0.8103
2023-10-16 20:23:58,988 ----------------------------------------------------------------------------------------------------
2023-10-16 20:24:00,624 epoch 9 - iter 27/272 - loss 0.00599380 - time (sec): 1.63 - samples/sec: 3348.82 - lr: 0.000006 - momentum: 0.000000
2023-10-16 20:24:02,276 epoch 9 - iter 54/272 - loss 0.01311148 - time (sec): 3.29 - samples/sec: 3429.61 - lr: 0.000006 - momentum: 0.000000
2023-10-16 20:24:03,891 epoch 9 - iter 81/272 - loss 0.01123314 - time (sec): 4.90 - samples/sec: 3443.51 - lr: 0.000006 - momentum: 0.000000
2023-10-16 20:24:05,413 epoch 9 - iter 108/272 - loss 0.01353945 - time (sec): 6.42 - samples/sec: 3323.39 - lr: 0.000005 - momentum: 0.000000
2023-10-16 20:24:07,043 epoch 9 - iter 135/272 - loss 0.01225708 - time (sec): 8.05 - samples/sec: 3282.12 - lr: 0.000005 - momentum: 0.000000
2023-10-16 20:24:08,759 epoch 9 - iter 162/272 - loss 0.01190655 - time (sec): 9.77 - samples/sec: 3230.90 - lr: 0.000005 - momentum: 0.000000
2023-10-16 20:24:10,412 epoch 9 - iter 189/272 - loss 0.01157508 - time (sec): 11.42 - samples/sec: 3213.24 - lr: 0.000004 - momentum: 0.000000
2023-10-16 20:24:11,943 epoch 9 - iter 216/272 - loss 0.01177737 - time (sec): 12.95 - samples/sec: 3230.00 - lr: 0.000004 - momentum: 0.000000
2023-10-16 20:24:13,506 epoch 9 - iter 243/272 - loss 0.01213604 - time (sec): 14.52 - samples/sec: 3218.45 - lr: 0.000004 - momentum: 0.000000
2023-10-16 20:24:14,983 epoch 9 - iter 270/272 - loss 0.01161932 - time (sec): 15.99 - samples/sec: 3226.42 - lr: 0.000003 - momentum: 0.000000
2023-10-16 20:24:15,106 ----------------------------------------------------------------------------------------------------
2023-10-16 20:24:15,106 EPOCH 9 done: loss 0.0115 - lr: 0.000003
2023-10-16 20:24:16,622 DEV : loss 0.16301465034484863 - f1-score (micro avg) 0.8327
2023-10-16 20:24:16,627 saving best model
2023-10-16 20:24:17,102 ----------------------------------------------------------------------------------------------------
2023-10-16 20:24:18,760 epoch 10 - iter 27/272 - loss 0.01320646 - time (sec): 1.66 - samples/sec: 2935.45 - lr: 0.000003 - momentum: 0.000000
2023-10-16 20:24:20,380 epoch 10 - iter 54/272 - loss 0.01195513 - time (sec): 3.28 - samples/sec: 3000.55 - lr: 0.000003 - momentum: 0.000000
2023-10-16 20:24:22,050 epoch 10 - iter 81/272 - loss 0.01072011 - time (sec): 4.95 - samples/sec: 3165.10 - lr: 0.000002 - momentum: 0.000000
2023-10-16 20:24:23,640 epoch 10 - iter 108/272 - loss 0.01026115 - time (sec): 6.54 - samples/sec: 3232.53 - lr: 0.000002 - momentum: 0.000000
2023-10-16 20:24:25,083 epoch 10 - iter 135/272 - loss 0.01064150 - time (sec): 7.98 - samples/sec: 3247.32 - lr: 0.000002 - momentum: 0.000000
2023-10-16 20:24:26,612 epoch 10 - iter 162/272 - loss 0.00997044 - time (sec): 9.51 - samples/sec: 3235.87 - lr: 0.000001 - momentum: 0.000000
2023-10-16 20:24:28,274 epoch 10 - iter 189/272 - loss 0.01040692 - time (sec): 11.17 - samples/sec: 3250.02 - lr: 0.000001 - momentum: 0.000000
2023-10-16 20:24:30,075 epoch 10 - iter 216/272 - loss 0.00966299 - time (sec): 12.97 - samples/sec: 3285.20 - lr: 0.000001 - momentum: 0.000000
2023-10-16 20:24:31,522 epoch 10 - iter 243/272 - loss 0.00976536 - time (sec): 14.42 - samples/sec: 3264.05 - lr: 0.000000 - momentum: 0.000000
2023-10-16 20:24:32,976 epoch 10 - iter 270/272 - loss 0.00978654 - time (sec): 15.87 - samples/sec: 3252.31 - lr: 0.000000 - momentum: 0.000000
2023-10-16 20:24:33,097 ----------------------------------------------------------------------------------------------------
2023-10-16 20:24:33,097 EPOCH 10 done: loss 0.0097 - lr: 0.000000
2023-10-16 20:24:34,795 DEV : loss 0.1661917269229889 - f1-score (micro avg) 0.8155
2023-10-16 20:24:35,171 ----------------------------------------------------------------------------------------------------
2023-10-16 20:24:35,172 Loading model from best epoch ...
2023-10-16 20:24:36,716 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
2023-10-16 20:24:38,779
Results:
- F-score (micro) 0.7838
- F-score (macro) 0.7384
- Accuracy 0.6582
By class:
precision recall f1-score support
LOC 0.8138 0.8686 0.8403 312
PER 0.6914 0.8510 0.7629 208
ORG 0.5217 0.4364 0.4752 55
HumanProd 0.8077 0.9545 0.8750 22
micro avg 0.7458 0.8258 0.7838 597
macro avg 0.7087 0.7776 0.7384 597
weighted avg 0.7440 0.8258 0.7810 597
2023-10-16 20:24:38,779 ----------------------------------------------------------------------------------------------------
|