File size: 23,996 Bytes
25a4fc1 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 |
2023-10-13 08:57:44,355 ----------------------------------------------------------------------------------------------------
2023-10-13 08:57:44,356 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=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-13 08:57:44,356 ----------------------------------------------------------------------------------------------------
2023-10-13 08:57:44,356 MultiCorpus: 1100 train + 206 dev + 240 test sentences
- NER_HIPE_2022 Corpus: 1100 train + 206 dev + 240 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/de/with_doc_seperator
2023-10-13 08:57:44,356 ----------------------------------------------------------------------------------------------------
2023-10-13 08:57:44,356 Train: 1100 sentences
2023-10-13 08:57:44,356 (train_with_dev=False, train_with_test=False)
2023-10-13 08:57:44,356 ----------------------------------------------------------------------------------------------------
2023-10-13 08:57:44,356 Training Params:
2023-10-13 08:57:44,356 - learning_rate: "5e-05"
2023-10-13 08:57:44,356 - mini_batch_size: "8"
2023-10-13 08:57:44,356 - max_epochs: "10"
2023-10-13 08:57:44,356 - shuffle: "True"
2023-10-13 08:57:44,356 ----------------------------------------------------------------------------------------------------
2023-10-13 08:57:44,356 Plugins:
2023-10-13 08:57:44,356 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 08:57:44,356 ----------------------------------------------------------------------------------------------------
2023-10-13 08:57:44,356 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 08:57:44,356 - metric: "('micro avg', 'f1-score')"
2023-10-13 08:57:44,356 ----------------------------------------------------------------------------------------------------
2023-10-13 08:57:44,356 Computation:
2023-10-13 08:57:44,356 - compute on device: cuda:0
2023-10-13 08:57:44,356 - embedding storage: none
2023-10-13 08:57:44,356 ----------------------------------------------------------------------------------------------------
2023-10-13 08:57:44,357 Model training base path: "hmbench-ajmc/de-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-13 08:57:44,357 ----------------------------------------------------------------------------------------------------
2023-10-13 08:57:44,357 ----------------------------------------------------------------------------------------------------
2023-10-13 08:57:45,066 epoch 1 - iter 13/138 - loss 3.19871477 - time (sec): 0.71 - samples/sec: 2838.21 - lr: 0.000004 - momentum: 0.000000
2023-10-13 08:57:45,802 epoch 1 - iter 26/138 - loss 2.89610673 - time (sec): 1.44 - samples/sec: 2760.43 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:57:46,524 epoch 1 - iter 39/138 - loss 2.36049543 - time (sec): 2.17 - samples/sec: 2768.07 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:57:47,246 epoch 1 - iter 52/138 - loss 1.95008445 - time (sec): 2.89 - samples/sec: 2883.52 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:57:47,976 epoch 1 - iter 65/138 - loss 1.69219973 - time (sec): 3.62 - samples/sec: 2930.93 - lr: 0.000023 - momentum: 0.000000
2023-10-13 08:57:48,723 epoch 1 - iter 78/138 - loss 1.51635053 - time (sec): 4.37 - samples/sec: 2930.56 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:57:49,458 epoch 1 - iter 91/138 - loss 1.37181089 - time (sec): 5.10 - samples/sec: 2922.55 - lr: 0.000033 - momentum: 0.000000
2023-10-13 08:57:50,183 epoch 1 - iter 104/138 - loss 1.24447856 - time (sec): 5.82 - samples/sec: 2955.21 - lr: 0.000037 - momentum: 0.000000
2023-10-13 08:57:50,865 epoch 1 - iter 117/138 - loss 1.14921726 - time (sec): 6.51 - samples/sec: 2980.86 - lr: 0.000042 - momentum: 0.000000
2023-10-13 08:57:51,559 epoch 1 - iter 130/138 - loss 1.07565563 - time (sec): 7.20 - samples/sec: 2962.77 - lr: 0.000047 - momentum: 0.000000
2023-10-13 08:57:52,012 ----------------------------------------------------------------------------------------------------
2023-10-13 08:57:52,012 EPOCH 1 done: loss 1.0337 - lr: 0.000047
2023-10-13 08:57:52,671 DEV : loss 0.22752070426940918 - f1-score (micro avg) 0.6682
2023-10-13 08:57:52,676 saving best model
2023-10-13 08:57:53,057 ----------------------------------------------------------------------------------------------------
2023-10-13 08:57:53,755 epoch 2 - iter 13/138 - loss 0.26158939 - time (sec): 0.70 - samples/sec: 3006.97 - lr: 0.000050 - momentum: 0.000000
2023-10-13 08:57:54,475 epoch 2 - iter 26/138 - loss 0.21700014 - time (sec): 1.42 - samples/sec: 3025.83 - lr: 0.000049 - momentum: 0.000000
2023-10-13 08:57:55,232 epoch 2 - iter 39/138 - loss 0.19436700 - time (sec): 2.17 - samples/sec: 3005.60 - lr: 0.000048 - momentum: 0.000000
2023-10-13 08:57:56,027 epoch 2 - iter 52/138 - loss 0.18948058 - time (sec): 2.97 - samples/sec: 3018.43 - lr: 0.000048 - momentum: 0.000000
2023-10-13 08:57:56,783 epoch 2 - iter 65/138 - loss 0.19422866 - time (sec): 3.72 - samples/sec: 3067.06 - lr: 0.000047 - momentum: 0.000000
2023-10-13 08:57:57,497 epoch 2 - iter 78/138 - loss 0.19168065 - time (sec): 4.44 - samples/sec: 2994.06 - lr: 0.000047 - momentum: 0.000000
2023-10-13 08:57:58,232 epoch 2 - iter 91/138 - loss 0.18872444 - time (sec): 5.17 - samples/sec: 2950.10 - lr: 0.000046 - momentum: 0.000000
2023-10-13 08:57:58,983 epoch 2 - iter 104/138 - loss 0.18461520 - time (sec): 5.92 - samples/sec: 2945.84 - lr: 0.000046 - momentum: 0.000000
2023-10-13 08:57:59,746 epoch 2 - iter 117/138 - loss 0.18182663 - time (sec): 6.69 - samples/sec: 2927.06 - lr: 0.000045 - momentum: 0.000000
2023-10-13 08:58:00,423 epoch 2 - iter 130/138 - loss 0.18006433 - time (sec): 7.36 - samples/sec: 2936.15 - lr: 0.000045 - momentum: 0.000000
2023-10-13 08:58:00,866 ----------------------------------------------------------------------------------------------------
2023-10-13 08:58:00,867 EPOCH 2 done: loss 0.1785 - lr: 0.000045
2023-10-13 08:58:01,517 DEV : loss 0.13236640393733978 - f1-score (micro avg) 0.8118
2023-10-13 08:58:01,522 saving best model
2023-10-13 08:58:01,977 ----------------------------------------------------------------------------------------------------
2023-10-13 08:58:02,737 epoch 3 - iter 13/138 - loss 0.08803266 - time (sec): 0.75 - samples/sec: 2807.32 - lr: 0.000044 - momentum: 0.000000
2023-10-13 08:58:03,522 epoch 3 - iter 26/138 - loss 0.07423184 - time (sec): 1.54 - samples/sec: 2913.30 - lr: 0.000043 - momentum: 0.000000
2023-10-13 08:58:04,214 epoch 3 - iter 39/138 - loss 0.08403636 - time (sec): 2.23 - samples/sec: 2912.80 - lr: 0.000043 - momentum: 0.000000
2023-10-13 08:58:04,912 epoch 3 - iter 52/138 - loss 0.08542265 - time (sec): 2.93 - samples/sec: 2852.10 - lr: 0.000042 - momentum: 0.000000
2023-10-13 08:58:05,637 epoch 3 - iter 65/138 - loss 0.08944773 - time (sec): 3.65 - samples/sec: 2912.28 - lr: 0.000042 - momentum: 0.000000
2023-10-13 08:58:06,362 epoch 3 - iter 78/138 - loss 0.08798574 - time (sec): 4.37 - samples/sec: 2907.54 - lr: 0.000041 - momentum: 0.000000
2023-10-13 08:58:07,089 epoch 3 - iter 91/138 - loss 0.09869091 - time (sec): 5.10 - samples/sec: 2961.96 - lr: 0.000041 - momentum: 0.000000
2023-10-13 08:58:07,806 epoch 3 - iter 104/138 - loss 0.09713533 - time (sec): 5.82 - samples/sec: 2968.89 - lr: 0.000040 - momentum: 0.000000
2023-10-13 08:58:08,550 epoch 3 - iter 117/138 - loss 0.09854157 - time (sec): 6.56 - samples/sec: 2973.31 - lr: 0.000040 - momentum: 0.000000
2023-10-13 08:58:09,276 epoch 3 - iter 130/138 - loss 0.09889222 - time (sec): 7.29 - samples/sec: 2957.02 - lr: 0.000039 - momentum: 0.000000
2023-10-13 08:58:09,718 ----------------------------------------------------------------------------------------------------
2023-10-13 08:58:09,718 EPOCH 3 done: loss 0.0998 - lr: 0.000039
2023-10-13 08:58:10,354 DEV : loss 0.13262909650802612 - f1-score (micro avg) 0.8483
2023-10-13 08:58:10,359 saving best model
2023-10-13 08:58:10,838 ----------------------------------------------------------------------------------------------------
2023-10-13 08:58:11,553 epoch 4 - iter 13/138 - loss 0.05094794 - time (sec): 0.71 - samples/sec: 3200.90 - lr: 0.000038 - momentum: 0.000000
2023-10-13 08:58:12,253 epoch 4 - iter 26/138 - loss 0.06347611 - time (sec): 1.41 - samples/sec: 3223.38 - lr: 0.000038 - momentum: 0.000000
2023-10-13 08:58:13,030 epoch 4 - iter 39/138 - loss 0.05700293 - time (sec): 2.19 - samples/sec: 3076.49 - lr: 0.000037 - momentum: 0.000000
2023-10-13 08:58:13,746 epoch 4 - iter 52/138 - loss 0.05711960 - time (sec): 2.91 - samples/sec: 2936.69 - lr: 0.000037 - momentum: 0.000000
2023-10-13 08:58:14,527 epoch 4 - iter 65/138 - loss 0.06612821 - time (sec): 3.69 - samples/sec: 2865.25 - lr: 0.000036 - momentum: 0.000000
2023-10-13 08:58:15,250 epoch 4 - iter 78/138 - loss 0.06073313 - time (sec): 4.41 - samples/sec: 2876.44 - lr: 0.000036 - momentum: 0.000000
2023-10-13 08:58:16,001 epoch 4 - iter 91/138 - loss 0.06310000 - time (sec): 5.16 - samples/sec: 2853.53 - lr: 0.000035 - momentum: 0.000000
2023-10-13 08:58:16,828 epoch 4 - iter 104/138 - loss 0.06924452 - time (sec): 5.99 - samples/sec: 2868.76 - lr: 0.000035 - momentum: 0.000000
2023-10-13 08:58:17,601 epoch 4 - iter 117/138 - loss 0.06882500 - time (sec): 6.76 - samples/sec: 2851.58 - lr: 0.000034 - momentum: 0.000000
2023-10-13 08:58:18,346 epoch 4 - iter 130/138 - loss 0.06960305 - time (sec): 7.51 - samples/sec: 2850.83 - lr: 0.000034 - momentum: 0.000000
2023-10-13 08:58:18,824 ----------------------------------------------------------------------------------------------------
2023-10-13 08:58:18,825 EPOCH 4 done: loss 0.0712 - lr: 0.000034
2023-10-13 08:58:19,475 DEV : loss 0.14707542955875397 - f1-score (micro avg) 0.8272
2023-10-13 08:58:19,480 ----------------------------------------------------------------------------------------------------
2023-10-13 08:58:20,274 epoch 5 - iter 13/138 - loss 0.03543947 - time (sec): 0.79 - samples/sec: 2901.84 - lr: 0.000033 - momentum: 0.000000
2023-10-13 08:58:20,985 epoch 5 - iter 26/138 - loss 0.04085515 - time (sec): 1.50 - samples/sec: 2903.87 - lr: 0.000032 - momentum: 0.000000
2023-10-13 08:58:21,691 epoch 5 - iter 39/138 - loss 0.04992583 - time (sec): 2.21 - samples/sec: 2912.06 - lr: 0.000032 - momentum: 0.000000
2023-10-13 08:58:22,429 epoch 5 - iter 52/138 - loss 0.05418000 - time (sec): 2.95 - samples/sec: 2959.47 - lr: 0.000031 - momentum: 0.000000
2023-10-13 08:58:23,160 epoch 5 - iter 65/138 - loss 0.05193546 - time (sec): 3.68 - samples/sec: 2998.23 - lr: 0.000031 - momentum: 0.000000
2023-10-13 08:58:23,863 epoch 5 - iter 78/138 - loss 0.04849240 - time (sec): 4.38 - samples/sec: 2956.36 - lr: 0.000030 - momentum: 0.000000
2023-10-13 08:58:24,546 epoch 5 - iter 91/138 - loss 0.04798557 - time (sec): 5.06 - samples/sec: 2980.23 - lr: 0.000030 - momentum: 0.000000
2023-10-13 08:58:25,265 epoch 5 - iter 104/138 - loss 0.04662516 - time (sec): 5.78 - samples/sec: 2976.44 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:58:26,015 epoch 5 - iter 117/138 - loss 0.05203494 - time (sec): 6.53 - samples/sec: 2982.25 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:58:26,707 epoch 5 - iter 130/138 - loss 0.04967259 - time (sec): 7.23 - samples/sec: 2961.77 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:58:27,173 ----------------------------------------------------------------------------------------------------
2023-10-13 08:58:27,173 EPOCH 5 done: loss 0.0490 - lr: 0.000028
2023-10-13 08:58:27,843 DEV : loss 0.15005655586719513 - f1-score (micro avg) 0.8501
2023-10-13 08:58:27,848 saving best model
2023-10-13 08:58:28,310 ----------------------------------------------------------------------------------------------------
2023-10-13 08:58:28,982 epoch 6 - iter 13/138 - loss 0.04557888 - time (sec): 0.67 - samples/sec: 3168.84 - lr: 0.000027 - momentum: 0.000000
2023-10-13 08:58:29,643 epoch 6 - iter 26/138 - loss 0.04511493 - time (sec): 1.33 - samples/sec: 3099.11 - lr: 0.000027 - momentum: 0.000000
2023-10-13 08:58:30,315 epoch 6 - iter 39/138 - loss 0.05003977 - time (sec): 2.00 - samples/sec: 3037.48 - lr: 0.000026 - momentum: 0.000000
2023-10-13 08:58:31,078 epoch 6 - iter 52/138 - loss 0.04184200 - time (sec): 2.77 - samples/sec: 2996.86 - lr: 0.000026 - momentum: 0.000000
2023-10-13 08:58:31,784 epoch 6 - iter 65/138 - loss 0.04444330 - time (sec): 3.47 - samples/sec: 2978.46 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:58:32,532 epoch 6 - iter 78/138 - loss 0.04511894 - time (sec): 4.22 - samples/sec: 2950.34 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:58:33,263 epoch 6 - iter 91/138 - loss 0.04235503 - time (sec): 4.95 - samples/sec: 2943.00 - lr: 0.000024 - momentum: 0.000000
2023-10-13 08:58:33,970 epoch 6 - iter 104/138 - loss 0.04162867 - time (sec): 5.66 - samples/sec: 2968.78 - lr: 0.000024 - momentum: 0.000000
2023-10-13 08:58:34,744 epoch 6 - iter 117/138 - loss 0.03722116 - time (sec): 6.43 - samples/sec: 2991.01 - lr: 0.000023 - momentum: 0.000000
2023-10-13 08:58:35,488 epoch 6 - iter 130/138 - loss 0.03409834 - time (sec): 7.18 - samples/sec: 3001.30 - lr: 0.000023 - momentum: 0.000000
2023-10-13 08:58:35,975 ----------------------------------------------------------------------------------------------------
2023-10-13 08:58:35,975 EPOCH 6 done: loss 0.0354 - lr: 0.000023
2023-10-13 08:58:36,637 DEV : loss 0.16685351729393005 - f1-score (micro avg) 0.8639
2023-10-13 08:58:36,643 saving best model
2023-10-13 08:58:37,114 ----------------------------------------------------------------------------------------------------
2023-10-13 08:58:37,828 epoch 7 - iter 13/138 - loss 0.01977267 - time (sec): 0.71 - samples/sec: 2980.66 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:58:38,560 epoch 7 - iter 26/138 - loss 0.02754021 - time (sec): 1.44 - samples/sec: 3030.57 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:58:39,268 epoch 7 - iter 39/138 - loss 0.02491542 - time (sec): 2.15 - samples/sec: 2953.52 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:58:40,039 epoch 7 - iter 52/138 - loss 0.03342961 - time (sec): 2.92 - samples/sec: 2965.94 - lr: 0.000020 - momentum: 0.000000
2023-10-13 08:58:40,766 epoch 7 - iter 65/138 - loss 0.03137856 - time (sec): 3.65 - samples/sec: 2949.20 - lr: 0.000020 - momentum: 0.000000
2023-10-13 08:58:41,475 epoch 7 - iter 78/138 - loss 0.02967382 - time (sec): 4.36 - samples/sec: 2913.44 - lr: 0.000019 - momentum: 0.000000
2023-10-13 08:58:42,160 epoch 7 - iter 91/138 - loss 0.02685385 - time (sec): 5.04 - samples/sec: 2929.12 - lr: 0.000019 - momentum: 0.000000
2023-10-13 08:58:42,855 epoch 7 - iter 104/138 - loss 0.02721307 - time (sec): 5.74 - samples/sec: 2932.13 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:58:43,606 epoch 7 - iter 117/138 - loss 0.03119356 - time (sec): 6.49 - samples/sec: 2911.22 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:58:44,317 epoch 7 - iter 130/138 - loss 0.02889865 - time (sec): 7.20 - samples/sec: 2961.63 - lr: 0.000017 - momentum: 0.000000
2023-10-13 08:58:44,797 ----------------------------------------------------------------------------------------------------
2023-10-13 08:58:44,797 EPOCH 7 done: loss 0.0298 - lr: 0.000017
2023-10-13 08:58:45,432 DEV : loss 0.17813566327095032 - f1-score (micro avg) 0.8636
2023-10-13 08:58:45,437 ----------------------------------------------------------------------------------------------------
2023-10-13 08:58:46,123 epoch 8 - iter 13/138 - loss 0.03541151 - time (sec): 0.69 - samples/sec: 3178.88 - lr: 0.000016 - momentum: 0.000000
2023-10-13 08:58:46,863 epoch 8 - iter 26/138 - loss 0.02622506 - time (sec): 1.42 - samples/sec: 3159.75 - lr: 0.000016 - momentum: 0.000000
2023-10-13 08:58:47,585 epoch 8 - iter 39/138 - loss 0.02389930 - time (sec): 2.15 - samples/sec: 3137.49 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:58:48,296 epoch 8 - iter 52/138 - loss 0.02205261 - time (sec): 2.86 - samples/sec: 3075.91 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:58:48,973 epoch 8 - iter 65/138 - loss 0.02665679 - time (sec): 3.54 - samples/sec: 3036.59 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:58:49,687 epoch 8 - iter 78/138 - loss 0.02287701 - time (sec): 4.25 - samples/sec: 3038.33 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:58:50,421 epoch 8 - iter 91/138 - loss 0.02221582 - time (sec): 4.98 - samples/sec: 2970.52 - lr: 0.000013 - momentum: 0.000000
2023-10-13 08:58:51,180 epoch 8 - iter 104/138 - loss 0.02401353 - time (sec): 5.74 - samples/sec: 2970.76 - lr: 0.000013 - momentum: 0.000000
2023-10-13 08:58:51,844 epoch 8 - iter 117/138 - loss 0.02385499 - time (sec): 6.41 - samples/sec: 2979.66 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:58:52,605 epoch 8 - iter 130/138 - loss 0.02237982 - time (sec): 7.17 - samples/sec: 2990.92 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:58:53,044 ----------------------------------------------------------------------------------------------------
2023-10-13 08:58:53,044 EPOCH 8 done: loss 0.0215 - lr: 0.000012
2023-10-13 08:58:53,691 DEV : loss 0.16658078134059906 - f1-score (micro avg) 0.8729
2023-10-13 08:58:53,696 saving best model
2023-10-13 08:58:54,142 ----------------------------------------------------------------------------------------------------
2023-10-13 08:58:54,908 epoch 9 - iter 13/138 - loss 0.00332000 - time (sec): 0.76 - samples/sec: 3032.77 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:58:55,676 epoch 9 - iter 26/138 - loss 0.01018189 - time (sec): 1.53 - samples/sec: 2897.38 - lr: 0.000010 - momentum: 0.000000
2023-10-13 08:58:56,374 epoch 9 - iter 39/138 - loss 0.00711881 - time (sec): 2.23 - samples/sec: 2901.51 - lr: 0.000010 - momentum: 0.000000
2023-10-13 08:58:57,094 epoch 9 - iter 52/138 - loss 0.01086945 - time (sec): 2.95 - samples/sec: 2965.20 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:58:57,835 epoch 9 - iter 65/138 - loss 0.01560931 - time (sec): 3.69 - samples/sec: 2984.23 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:58:58,563 epoch 9 - iter 78/138 - loss 0.01446554 - time (sec): 4.42 - samples/sec: 2998.82 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:58:59,278 epoch 9 - iter 91/138 - loss 0.01548107 - time (sec): 5.13 - samples/sec: 2990.47 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:59:00,018 epoch 9 - iter 104/138 - loss 0.01455742 - time (sec): 5.87 - samples/sec: 2978.72 - lr: 0.000007 - momentum: 0.000000
2023-10-13 08:59:00,764 epoch 9 - iter 117/138 - loss 0.01346173 - time (sec): 6.62 - samples/sec: 2974.79 - lr: 0.000007 - momentum: 0.000000
2023-10-13 08:59:01,440 epoch 9 - iter 130/138 - loss 0.01265528 - time (sec): 7.30 - samples/sec: 2964.38 - lr: 0.000006 - momentum: 0.000000
2023-10-13 08:59:01,877 ----------------------------------------------------------------------------------------------------
2023-10-13 08:59:01,878 EPOCH 9 done: loss 0.0140 - lr: 0.000006
2023-10-13 08:59:02,509 DEV : loss 0.16519632935523987 - f1-score (micro avg) 0.8851
2023-10-13 08:59:02,514 saving best model
2023-10-13 08:59:02,966 ----------------------------------------------------------------------------------------------------
2023-10-13 08:59:03,747 epoch 10 - iter 13/138 - loss 0.02129498 - time (sec): 0.78 - samples/sec: 2722.32 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:59:04,445 epoch 10 - iter 26/138 - loss 0.02113946 - time (sec): 1.48 - samples/sec: 2948.47 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:59:05,193 epoch 10 - iter 39/138 - loss 0.01666252 - time (sec): 2.23 - samples/sec: 2936.44 - lr: 0.000004 - momentum: 0.000000
2023-10-13 08:59:05,934 epoch 10 - iter 52/138 - loss 0.01387327 - time (sec): 2.97 - samples/sec: 2946.51 - lr: 0.000004 - momentum: 0.000000
2023-10-13 08:59:06,683 epoch 10 - iter 65/138 - loss 0.01320655 - time (sec): 3.72 - samples/sec: 2953.57 - lr: 0.000003 - momentum: 0.000000
2023-10-13 08:59:07,479 epoch 10 - iter 78/138 - loss 0.01282655 - time (sec): 4.51 - samples/sec: 2922.48 - lr: 0.000003 - momentum: 0.000000
2023-10-13 08:59:08,195 epoch 10 - iter 91/138 - loss 0.01283300 - time (sec): 5.23 - samples/sec: 2916.12 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:59:08,950 epoch 10 - iter 104/138 - loss 0.01181631 - time (sec): 5.98 - samples/sec: 2926.04 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:59:09,726 epoch 10 - iter 117/138 - loss 0.01058247 - time (sec): 6.76 - samples/sec: 2903.36 - lr: 0.000001 - momentum: 0.000000
2023-10-13 08:59:10,433 epoch 10 - iter 130/138 - loss 0.01126915 - time (sec): 7.47 - samples/sec: 2904.91 - lr: 0.000000 - momentum: 0.000000
2023-10-13 08:59:10,865 ----------------------------------------------------------------------------------------------------
2023-10-13 08:59:10,865 EPOCH 10 done: loss 0.0114 - lr: 0.000000
2023-10-13 08:59:11,520 DEV : loss 0.16343694925308228 - f1-score (micro avg) 0.8875
2023-10-13 08:59:11,525 saving best model
2023-10-13 08:59:12,404 ----------------------------------------------------------------------------------------------------
2023-10-13 08:59:12,406 Loading model from best epoch ...
2023-10-13 08:59:13,930 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
2023-10-13 08:59:14,693
Results:
- F-score (micro) 0.9108
- F-score (macro) 0.5485
- Accuracy 0.8484
By class:
precision recall f1-score support
scope 0.8743 0.9091 0.8914 176
pers 0.9762 0.9609 0.9685 128
work 0.9014 0.8649 0.8828 74
object 0.0000 0.0000 0.0000 2
loc 0.0000 0.0000 0.0000 2
micro avg 0.9132 0.9084 0.9108 382
macro avg 0.5504 0.5470 0.5485 382
weighted avg 0.9045 0.9084 0.9062 382
2023-10-13 08:59:14,693 ----------------------------------------------------------------------------------------------------
|