File size: 23,794 Bytes
6bbf4ce |
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 |
2023-10-16 18:44:14,992 ----------------------------------------------------------------------------------------------------
2023-10-16 18:44:14,993 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 18:44:14,993 ----------------------------------------------------------------------------------------------------
2023-10-16 18:44:14,993 MultiCorpus: 1166 train + 165 dev + 415 test sentences
- NER_HIPE_2022 Corpus: 1166 train + 165 dev + 415 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fi/with_doc_seperator
2023-10-16 18:44:14,993 ----------------------------------------------------------------------------------------------------
2023-10-16 18:44:14,993 Train: 1166 sentences
2023-10-16 18:44:14,993 (train_with_dev=False, train_with_test=False)
2023-10-16 18:44:14,993 ----------------------------------------------------------------------------------------------------
2023-10-16 18:44:14,993 Training Params:
2023-10-16 18:44:14,993 - learning_rate: "5e-05"
2023-10-16 18:44:14,993 - mini_batch_size: "8"
2023-10-16 18:44:14,993 - max_epochs: "10"
2023-10-16 18:44:14,993 - shuffle: "True"
2023-10-16 18:44:14,994 ----------------------------------------------------------------------------------------------------
2023-10-16 18:44:14,994 Plugins:
2023-10-16 18:44:14,994 - LinearScheduler | warmup_fraction: '0.1'
2023-10-16 18:44:14,994 ----------------------------------------------------------------------------------------------------
2023-10-16 18:44:14,994 Final evaluation on model from best epoch (best-model.pt)
2023-10-16 18:44:14,994 - metric: "('micro avg', 'f1-score')"
2023-10-16 18:44:14,994 ----------------------------------------------------------------------------------------------------
2023-10-16 18:44:14,994 Computation:
2023-10-16 18:44:14,994 - compute on device: cuda:0
2023-10-16 18:44:14,994 - embedding storage: none
2023-10-16 18:44:14,994 ----------------------------------------------------------------------------------------------------
2023-10-16 18:44:14,994 Model training base path: "hmbench-newseye/fi-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-16 18:44:14,994 ----------------------------------------------------------------------------------------------------
2023-10-16 18:44:14,994 ----------------------------------------------------------------------------------------------------
2023-10-16 18:44:16,450 epoch 1 - iter 14/146 - loss 2.93554078 - time (sec): 1.46 - samples/sec: 3031.39 - lr: 0.000004 - momentum: 0.000000
2023-10-16 18:44:17,598 epoch 1 - iter 28/146 - loss 2.59641910 - time (sec): 2.60 - samples/sec: 3093.50 - lr: 0.000009 - momentum: 0.000000
2023-10-16 18:44:19,197 epoch 1 - iter 42/146 - loss 1.98647565 - time (sec): 4.20 - samples/sec: 3046.18 - lr: 0.000014 - momentum: 0.000000
2023-10-16 18:44:20,958 epoch 1 - iter 56/146 - loss 1.65493251 - time (sec): 5.96 - samples/sec: 2908.07 - lr: 0.000019 - momentum: 0.000000
2023-10-16 18:44:22,265 epoch 1 - iter 70/146 - loss 1.48549199 - time (sec): 7.27 - samples/sec: 2906.30 - lr: 0.000024 - momentum: 0.000000
2023-10-16 18:44:23,957 epoch 1 - iter 84/146 - loss 1.33028113 - time (sec): 8.96 - samples/sec: 2878.44 - lr: 0.000028 - momentum: 0.000000
2023-10-16 18:44:25,337 epoch 1 - iter 98/146 - loss 1.19387586 - time (sec): 10.34 - samples/sec: 2907.41 - lr: 0.000033 - momentum: 0.000000
2023-10-16 18:44:26,724 epoch 1 - iter 112/146 - loss 1.07702304 - time (sec): 11.73 - samples/sec: 2939.16 - lr: 0.000038 - momentum: 0.000000
2023-10-16 18:44:28,150 epoch 1 - iter 126/146 - loss 0.98680341 - time (sec): 13.16 - samples/sec: 2960.74 - lr: 0.000043 - momentum: 0.000000
2023-10-16 18:44:29,393 epoch 1 - iter 140/146 - loss 0.91930004 - time (sec): 14.40 - samples/sec: 2986.90 - lr: 0.000048 - momentum: 0.000000
2023-10-16 18:44:29,886 ----------------------------------------------------------------------------------------------------
2023-10-16 18:44:29,887 EPOCH 1 done: loss 0.8989 - lr: 0.000048
2023-10-16 18:44:30,915 DEV : loss 0.19385209679603577 - f1-score (micro avg) 0.4388
2023-10-16 18:44:30,919 saving best model
2023-10-16 18:44:31,329 ----------------------------------------------------------------------------------------------------
2023-10-16 18:44:32,656 epoch 2 - iter 14/146 - loss 0.26646816 - time (sec): 1.33 - samples/sec: 3202.19 - lr: 0.000050 - momentum: 0.000000
2023-10-16 18:44:34,281 epoch 2 - iter 28/146 - loss 0.27245130 - time (sec): 2.95 - samples/sec: 3188.14 - lr: 0.000049 - momentum: 0.000000
2023-10-16 18:44:35,932 epoch 2 - iter 42/146 - loss 0.28481191 - time (sec): 4.60 - samples/sec: 2962.86 - lr: 0.000048 - momentum: 0.000000
2023-10-16 18:44:37,632 epoch 2 - iter 56/146 - loss 0.25721835 - time (sec): 6.30 - samples/sec: 2913.56 - lr: 0.000048 - momentum: 0.000000
2023-10-16 18:44:38,924 epoch 2 - iter 70/146 - loss 0.24376985 - time (sec): 7.59 - samples/sec: 2929.54 - lr: 0.000047 - momentum: 0.000000
2023-10-16 18:44:40,297 epoch 2 - iter 84/146 - loss 0.23798723 - time (sec): 8.97 - samples/sec: 2952.15 - lr: 0.000047 - momentum: 0.000000
2023-10-16 18:44:41,482 epoch 2 - iter 98/146 - loss 0.23172878 - time (sec): 10.15 - samples/sec: 2977.39 - lr: 0.000046 - momentum: 0.000000
2023-10-16 18:44:42,768 epoch 2 - iter 112/146 - loss 0.21818927 - time (sec): 11.44 - samples/sec: 3000.61 - lr: 0.000046 - momentum: 0.000000
2023-10-16 18:44:44,349 epoch 2 - iter 126/146 - loss 0.21149100 - time (sec): 13.02 - samples/sec: 2974.30 - lr: 0.000045 - momentum: 0.000000
2023-10-16 18:44:45,592 epoch 2 - iter 140/146 - loss 0.20682752 - time (sec): 14.26 - samples/sec: 2983.62 - lr: 0.000045 - momentum: 0.000000
2023-10-16 18:44:46,258 ----------------------------------------------------------------------------------------------------
2023-10-16 18:44:46,259 EPOCH 2 done: loss 0.2017 - lr: 0.000045
2023-10-16 18:44:47,492 DEV : loss 0.11957067251205444 - f1-score (micro avg) 0.6291
2023-10-16 18:44:47,503 saving best model
2023-10-16 18:44:47,979 ----------------------------------------------------------------------------------------------------
2023-10-16 18:44:49,571 epoch 3 - iter 14/146 - loss 0.11191478 - time (sec): 1.58 - samples/sec: 3023.30 - lr: 0.000044 - momentum: 0.000000
2023-10-16 18:44:51,442 epoch 3 - iter 28/146 - loss 0.11158050 - time (sec): 3.45 - samples/sec: 2823.35 - lr: 0.000043 - momentum: 0.000000
2023-10-16 18:44:52,817 epoch 3 - iter 42/146 - loss 0.11844144 - time (sec): 4.83 - samples/sec: 2769.61 - lr: 0.000043 - momentum: 0.000000
2023-10-16 18:44:53,989 epoch 3 - iter 56/146 - loss 0.11296688 - time (sec): 6.00 - samples/sec: 2807.39 - lr: 0.000042 - momentum: 0.000000
2023-10-16 18:44:55,247 epoch 3 - iter 70/146 - loss 0.11117130 - time (sec): 7.26 - samples/sec: 2862.75 - lr: 0.000042 - momentum: 0.000000
2023-10-16 18:44:56,780 epoch 3 - iter 84/146 - loss 0.10769328 - time (sec): 8.79 - samples/sec: 2918.14 - lr: 0.000041 - momentum: 0.000000
2023-10-16 18:44:58,413 epoch 3 - iter 98/146 - loss 0.10822287 - time (sec): 10.42 - samples/sec: 2956.84 - lr: 0.000041 - momentum: 0.000000
2023-10-16 18:44:59,893 epoch 3 - iter 112/146 - loss 0.10974325 - time (sec): 11.90 - samples/sec: 2946.24 - lr: 0.000040 - momentum: 0.000000
2023-10-16 18:45:01,233 epoch 3 - iter 126/146 - loss 0.10961191 - time (sec): 13.24 - samples/sec: 2948.19 - lr: 0.000040 - momentum: 0.000000
2023-10-16 18:45:02,762 epoch 3 - iter 140/146 - loss 0.11159692 - time (sec): 14.77 - samples/sec: 2908.76 - lr: 0.000039 - momentum: 0.000000
2023-10-16 18:45:03,277 ----------------------------------------------------------------------------------------------------
2023-10-16 18:45:03,277 EPOCH 3 done: loss 0.1116 - lr: 0.000039
2023-10-16 18:45:04,525 DEV : loss 0.10484768450260162 - f1-score (micro avg) 0.7265
2023-10-16 18:45:04,529 saving best model
2023-10-16 18:45:05,074 ----------------------------------------------------------------------------------------------------
2023-10-16 18:45:06,751 epoch 4 - iter 14/146 - loss 0.06773359 - time (sec): 1.67 - samples/sec: 3019.84 - lr: 0.000038 - momentum: 0.000000
2023-10-16 18:45:08,400 epoch 4 - iter 28/146 - loss 0.08808391 - time (sec): 3.32 - samples/sec: 2840.89 - lr: 0.000038 - momentum: 0.000000
2023-10-16 18:45:09,895 epoch 4 - iter 42/146 - loss 0.07321162 - time (sec): 4.82 - samples/sec: 2896.89 - lr: 0.000037 - momentum: 0.000000
2023-10-16 18:45:11,044 epoch 4 - iter 56/146 - loss 0.07713816 - time (sec): 5.97 - samples/sec: 2905.45 - lr: 0.000037 - momentum: 0.000000
2023-10-16 18:45:12,599 epoch 4 - iter 70/146 - loss 0.07754016 - time (sec): 7.52 - samples/sec: 2913.73 - lr: 0.000036 - momentum: 0.000000
2023-10-16 18:45:14,247 epoch 4 - iter 84/146 - loss 0.07736573 - time (sec): 9.17 - samples/sec: 2861.76 - lr: 0.000036 - momentum: 0.000000
2023-10-16 18:45:15,521 epoch 4 - iter 98/146 - loss 0.07596628 - time (sec): 10.44 - samples/sec: 2860.25 - lr: 0.000035 - momentum: 0.000000
2023-10-16 18:45:17,303 epoch 4 - iter 112/146 - loss 0.07335282 - time (sec): 12.23 - samples/sec: 2866.41 - lr: 0.000035 - momentum: 0.000000
2023-10-16 18:45:18,600 epoch 4 - iter 126/146 - loss 0.07440373 - time (sec): 13.52 - samples/sec: 2899.44 - lr: 0.000034 - momentum: 0.000000
2023-10-16 18:45:19,917 epoch 4 - iter 140/146 - loss 0.07551967 - time (sec): 14.84 - samples/sec: 2900.64 - lr: 0.000034 - momentum: 0.000000
2023-10-16 18:45:20,359 ----------------------------------------------------------------------------------------------------
2023-10-16 18:45:20,359 EPOCH 4 done: loss 0.0754 - lr: 0.000034
2023-10-16 18:45:21,637 DEV : loss 0.106049545109272 - f1-score (micro avg) 0.7489
2023-10-16 18:45:21,642 saving best model
2023-10-16 18:45:22,190 ----------------------------------------------------------------------------------------------------
2023-10-16 18:45:23,719 epoch 5 - iter 14/146 - loss 0.04848651 - time (sec): 1.53 - samples/sec: 2833.76 - lr: 0.000033 - momentum: 0.000000
2023-10-16 18:45:25,226 epoch 5 - iter 28/146 - loss 0.04053376 - time (sec): 3.03 - samples/sec: 2740.38 - lr: 0.000032 - momentum: 0.000000
2023-10-16 18:45:26,737 epoch 5 - iter 42/146 - loss 0.04042929 - time (sec): 4.54 - samples/sec: 2797.98 - lr: 0.000032 - momentum: 0.000000
2023-10-16 18:45:28,427 epoch 5 - iter 56/146 - loss 0.04755111 - time (sec): 6.23 - samples/sec: 2885.92 - lr: 0.000031 - momentum: 0.000000
2023-10-16 18:45:30,011 epoch 5 - iter 70/146 - loss 0.04653289 - time (sec): 7.82 - samples/sec: 2907.59 - lr: 0.000031 - momentum: 0.000000
2023-10-16 18:45:31,331 epoch 5 - iter 84/146 - loss 0.05035831 - time (sec): 9.14 - samples/sec: 2918.02 - lr: 0.000030 - momentum: 0.000000
2023-10-16 18:45:32,949 epoch 5 - iter 98/146 - loss 0.04699916 - time (sec): 10.76 - samples/sec: 2946.14 - lr: 0.000030 - momentum: 0.000000
2023-10-16 18:45:34,106 epoch 5 - iter 112/146 - loss 0.04895681 - time (sec): 11.91 - samples/sec: 2962.47 - lr: 0.000029 - momentum: 0.000000
2023-10-16 18:45:35,479 epoch 5 - iter 126/146 - loss 0.05041908 - time (sec): 13.29 - samples/sec: 2959.37 - lr: 0.000029 - momentum: 0.000000
2023-10-16 18:45:36,721 epoch 5 - iter 140/146 - loss 0.05285443 - time (sec): 14.53 - samples/sec: 2974.98 - lr: 0.000028 - momentum: 0.000000
2023-10-16 18:45:37,198 ----------------------------------------------------------------------------------------------------
2023-10-16 18:45:37,198 EPOCH 5 done: loss 0.0529 - lr: 0.000028
2023-10-16 18:45:38,518 DEV : loss 0.11109450459480286 - f1-score (micro avg) 0.7431
2023-10-16 18:45:38,525 ----------------------------------------------------------------------------------------------------
2023-10-16 18:45:39,890 epoch 6 - iter 14/146 - loss 0.05005263 - time (sec): 1.36 - samples/sec: 3099.52 - lr: 0.000027 - momentum: 0.000000
2023-10-16 18:45:41,470 epoch 6 - iter 28/146 - loss 0.04617285 - time (sec): 2.94 - samples/sec: 3109.63 - lr: 0.000027 - momentum: 0.000000
2023-10-16 18:45:42,781 epoch 6 - iter 42/146 - loss 0.03762395 - time (sec): 4.25 - samples/sec: 3087.76 - lr: 0.000026 - momentum: 0.000000
2023-10-16 18:45:44,214 epoch 6 - iter 56/146 - loss 0.03821664 - time (sec): 5.69 - samples/sec: 3009.65 - lr: 0.000026 - momentum: 0.000000
2023-10-16 18:45:45,976 epoch 6 - iter 70/146 - loss 0.03792029 - time (sec): 7.45 - samples/sec: 2884.01 - lr: 0.000025 - momentum: 0.000000
2023-10-16 18:45:47,431 epoch 6 - iter 84/146 - loss 0.03947725 - time (sec): 8.90 - samples/sec: 2851.63 - lr: 0.000025 - momentum: 0.000000
2023-10-16 18:45:48,916 epoch 6 - iter 98/146 - loss 0.03763630 - time (sec): 10.39 - samples/sec: 2885.48 - lr: 0.000024 - momentum: 0.000000
2023-10-16 18:45:50,323 epoch 6 - iter 112/146 - loss 0.03716618 - time (sec): 11.80 - samples/sec: 2911.43 - lr: 0.000024 - momentum: 0.000000
2023-10-16 18:45:51,783 epoch 6 - iter 126/146 - loss 0.03715143 - time (sec): 13.26 - samples/sec: 2934.88 - lr: 0.000023 - momentum: 0.000000
2023-10-16 18:45:53,097 epoch 6 - iter 140/146 - loss 0.03776802 - time (sec): 14.57 - samples/sec: 2928.64 - lr: 0.000023 - momentum: 0.000000
2023-10-16 18:45:53,735 ----------------------------------------------------------------------------------------------------
2023-10-16 18:45:53,735 EPOCH 6 done: loss 0.0377 - lr: 0.000023
2023-10-16 18:45:55,064 DEV : loss 0.10371122509241104 - f1-score (micro avg) 0.7702
2023-10-16 18:45:55,070 saving best model
2023-10-16 18:45:55,594 ----------------------------------------------------------------------------------------------------
2023-10-16 18:45:56,996 epoch 7 - iter 14/146 - loss 0.02717694 - time (sec): 1.40 - samples/sec: 3059.75 - lr: 0.000022 - momentum: 0.000000
2023-10-16 18:45:58,279 epoch 7 - iter 28/146 - loss 0.02170902 - time (sec): 2.68 - samples/sec: 3094.36 - lr: 0.000021 - momentum: 0.000000
2023-10-16 18:45:59,710 epoch 7 - iter 42/146 - loss 0.01935375 - time (sec): 4.11 - samples/sec: 3129.38 - lr: 0.000021 - momentum: 0.000000
2023-10-16 18:46:01,234 epoch 7 - iter 56/146 - loss 0.02181448 - time (sec): 5.64 - samples/sec: 3090.41 - lr: 0.000020 - momentum: 0.000000
2023-10-16 18:46:02,563 epoch 7 - iter 70/146 - loss 0.02092834 - time (sec): 6.97 - samples/sec: 3056.05 - lr: 0.000020 - momentum: 0.000000
2023-10-16 18:46:04,330 epoch 7 - iter 84/146 - loss 0.02507712 - time (sec): 8.73 - samples/sec: 2999.06 - lr: 0.000019 - momentum: 0.000000
2023-10-16 18:46:05,695 epoch 7 - iter 98/146 - loss 0.02423133 - time (sec): 10.10 - samples/sec: 3005.29 - lr: 0.000019 - momentum: 0.000000
2023-10-16 18:46:07,233 epoch 7 - iter 112/146 - loss 0.02532614 - time (sec): 11.64 - samples/sec: 2964.77 - lr: 0.000018 - momentum: 0.000000
2023-10-16 18:46:08,741 epoch 7 - iter 126/146 - loss 0.02581130 - time (sec): 13.15 - samples/sec: 2945.18 - lr: 0.000018 - momentum: 0.000000
2023-10-16 18:46:10,281 epoch 7 - iter 140/146 - loss 0.02646009 - time (sec): 14.69 - samples/sec: 2926.02 - lr: 0.000017 - momentum: 0.000000
2023-10-16 18:46:10,770 ----------------------------------------------------------------------------------------------------
2023-10-16 18:46:10,770 EPOCH 7 done: loss 0.0263 - lr: 0.000017
2023-10-16 18:46:12,029 DEV : loss 0.12524546682834625 - f1-score (micro avg) 0.7613
2023-10-16 18:46:12,034 ----------------------------------------------------------------------------------------------------
2023-10-16 18:46:13,421 epoch 8 - iter 14/146 - loss 0.01540096 - time (sec): 1.39 - samples/sec: 3120.75 - lr: 0.000016 - momentum: 0.000000
2023-10-16 18:46:14,944 epoch 8 - iter 28/146 - loss 0.01738761 - time (sec): 2.91 - samples/sec: 3063.83 - lr: 0.000016 - momentum: 0.000000
2023-10-16 18:46:16,589 epoch 8 - iter 42/146 - loss 0.02301455 - time (sec): 4.55 - samples/sec: 3009.18 - lr: 0.000015 - momentum: 0.000000
2023-10-16 18:46:17,768 epoch 8 - iter 56/146 - loss 0.02368094 - time (sec): 5.73 - samples/sec: 2956.04 - lr: 0.000015 - momentum: 0.000000
2023-10-16 18:46:19,225 epoch 8 - iter 70/146 - loss 0.02328872 - time (sec): 7.19 - samples/sec: 2990.68 - lr: 0.000014 - momentum: 0.000000
2023-10-16 18:46:20,897 epoch 8 - iter 84/146 - loss 0.02214130 - time (sec): 8.86 - samples/sec: 2969.74 - lr: 0.000014 - momentum: 0.000000
2023-10-16 18:46:22,215 epoch 8 - iter 98/146 - loss 0.02037658 - time (sec): 10.18 - samples/sec: 3002.75 - lr: 0.000013 - momentum: 0.000000
2023-10-16 18:46:23,724 epoch 8 - iter 112/146 - loss 0.02107436 - time (sec): 11.69 - samples/sec: 2975.20 - lr: 0.000013 - momentum: 0.000000
2023-10-16 18:46:25,210 epoch 8 - iter 126/146 - loss 0.02013553 - time (sec): 13.18 - samples/sec: 2976.87 - lr: 0.000012 - momentum: 0.000000
2023-10-16 18:46:26,439 epoch 8 - iter 140/146 - loss 0.01954804 - time (sec): 14.40 - samples/sec: 2984.73 - lr: 0.000012 - momentum: 0.000000
2023-10-16 18:46:26,930 ----------------------------------------------------------------------------------------------------
2023-10-16 18:46:26,930 EPOCH 8 done: loss 0.0194 - lr: 0.000012
2023-10-16 18:46:28,209 DEV : loss 0.1471555233001709 - f1-score (micro avg) 0.7468
2023-10-16 18:46:28,214 ----------------------------------------------------------------------------------------------------
2023-10-16 18:46:29,632 epoch 9 - iter 14/146 - loss 0.02032134 - time (sec): 1.42 - samples/sec: 3324.63 - lr: 0.000011 - momentum: 0.000000
2023-10-16 18:46:31,349 epoch 9 - iter 28/146 - loss 0.01946398 - time (sec): 3.13 - samples/sec: 2988.38 - lr: 0.000010 - momentum: 0.000000
2023-10-16 18:46:32,688 epoch 9 - iter 42/146 - loss 0.01654113 - time (sec): 4.47 - samples/sec: 2913.92 - lr: 0.000010 - momentum: 0.000000
2023-10-16 18:46:34,080 epoch 9 - iter 56/146 - loss 0.01628118 - time (sec): 5.86 - samples/sec: 2922.13 - lr: 0.000009 - momentum: 0.000000
2023-10-16 18:46:35,284 epoch 9 - iter 70/146 - loss 0.01650357 - time (sec): 7.07 - samples/sec: 2956.01 - lr: 0.000009 - momentum: 0.000000
2023-10-16 18:46:36,626 epoch 9 - iter 84/146 - loss 0.01631769 - time (sec): 8.41 - samples/sec: 2971.67 - lr: 0.000008 - momentum: 0.000000
2023-10-16 18:46:38,373 epoch 9 - iter 98/146 - loss 0.01409268 - time (sec): 10.16 - samples/sec: 2910.90 - lr: 0.000008 - momentum: 0.000000
2023-10-16 18:46:39,652 epoch 9 - iter 112/146 - loss 0.01357809 - time (sec): 11.44 - samples/sec: 2907.10 - lr: 0.000007 - momentum: 0.000000
2023-10-16 18:46:41,271 epoch 9 - iter 126/146 - loss 0.01361320 - time (sec): 13.06 - samples/sec: 2887.92 - lr: 0.000007 - momentum: 0.000000
2023-10-16 18:46:42,819 epoch 9 - iter 140/146 - loss 0.01401163 - time (sec): 14.60 - samples/sec: 2903.52 - lr: 0.000006 - momentum: 0.000000
2023-10-16 18:46:43,393 ----------------------------------------------------------------------------------------------------
2023-10-16 18:46:43,393 EPOCH 9 done: loss 0.0137 - lr: 0.000006
2023-10-16 18:46:44,724 DEV : loss 0.14237351715564728 - f1-score (micro avg) 0.7425
2023-10-16 18:46:44,729 ----------------------------------------------------------------------------------------------------
2023-10-16 18:46:46,002 epoch 10 - iter 14/146 - loss 0.01164471 - time (sec): 1.27 - samples/sec: 2898.78 - lr: 0.000005 - momentum: 0.000000
2023-10-16 18:46:47,276 epoch 10 - iter 28/146 - loss 0.00769121 - time (sec): 2.55 - samples/sec: 3100.67 - lr: 0.000005 - momentum: 0.000000
2023-10-16 18:46:48,584 epoch 10 - iter 42/146 - loss 0.01041592 - time (sec): 3.85 - samples/sec: 3128.38 - lr: 0.000004 - momentum: 0.000000
2023-10-16 18:46:50,364 epoch 10 - iter 56/146 - loss 0.01063731 - time (sec): 5.63 - samples/sec: 3008.74 - lr: 0.000004 - momentum: 0.000000
2023-10-16 18:46:51,888 epoch 10 - iter 70/146 - loss 0.01116153 - time (sec): 7.16 - samples/sec: 3038.46 - lr: 0.000003 - momentum: 0.000000
2023-10-16 18:46:53,303 epoch 10 - iter 84/146 - loss 0.01395303 - time (sec): 8.57 - samples/sec: 3047.73 - lr: 0.000003 - momentum: 0.000000
2023-10-16 18:46:54,633 epoch 10 - iter 98/146 - loss 0.01344263 - time (sec): 9.90 - samples/sec: 3046.71 - lr: 0.000002 - momentum: 0.000000
2023-10-16 18:46:56,080 epoch 10 - iter 112/146 - loss 0.01227917 - time (sec): 11.35 - samples/sec: 3019.23 - lr: 0.000002 - momentum: 0.000000
2023-10-16 18:46:57,603 epoch 10 - iter 126/146 - loss 0.01125604 - time (sec): 12.87 - samples/sec: 3017.04 - lr: 0.000001 - momentum: 0.000000
2023-10-16 18:46:59,081 epoch 10 - iter 140/146 - loss 0.01038127 - time (sec): 14.35 - samples/sec: 3020.01 - lr: 0.000000 - momentum: 0.000000
2023-10-16 18:46:59,533 ----------------------------------------------------------------------------------------------------
2023-10-16 18:46:59,533 EPOCH 10 done: loss 0.0104 - lr: 0.000000
2023-10-16 18:47:00,784 DEV : loss 0.14098528027534485 - f1-score (micro avg) 0.7532
2023-10-16 18:47:01,250 ----------------------------------------------------------------------------------------------------
2023-10-16 18:47:01,252 Loading model from best epoch ...
2023-10-16 18:47:02,920 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
2023-10-16 18:47:05,389
Results:
- F-score (micro) 0.7535
- F-score (macro) 0.6922
- Accuracy 0.6267
By class:
precision recall f1-score support
PER 0.7902 0.8333 0.8112 348
LOC 0.6719 0.8161 0.7370 261
ORG 0.4340 0.4423 0.4381 52
HumanProd 0.7500 0.8182 0.7826 22
micro avg 0.7148 0.7965 0.7535 683
macro avg 0.6615 0.7275 0.6922 683
weighted avg 0.7166 0.7965 0.7535 683
2023-10-16 18:47:05,390 ----------------------------------------------------------------------------------------------------
|