File size: 23,965 Bytes
b591e1d |
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 |
2023-10-14 20:44:51,075 ----------------------------------------------------------------------------------------------------
2023-10-14 20:44:51,076 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=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-14 20:44:51,076 ----------------------------------------------------------------------------------------------------
2023-10-14 20:44:51,076 MultiCorpus: 14465 train + 1392 dev + 2432 test sentences
- NER_HIPE_2022 Corpus: 14465 train + 1392 dev + 2432 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/letemps/fr/with_doc_seperator
2023-10-14 20:44:51,076 ----------------------------------------------------------------------------------------------------
2023-10-14 20:44:51,076 Train: 14465 sentences
2023-10-14 20:44:51,076 (train_with_dev=False, train_with_test=False)
2023-10-14 20:44:51,076 ----------------------------------------------------------------------------------------------------
2023-10-14 20:44:51,077 Training Params:
2023-10-14 20:44:51,077 - learning_rate: "5e-05"
2023-10-14 20:44:51,077 - mini_batch_size: "4"
2023-10-14 20:44:51,077 - max_epochs: "10"
2023-10-14 20:44:51,077 - shuffle: "True"
2023-10-14 20:44:51,077 ----------------------------------------------------------------------------------------------------
2023-10-14 20:44:51,077 Plugins:
2023-10-14 20:44:51,077 - LinearScheduler | warmup_fraction: '0.1'
2023-10-14 20:44:51,077 ----------------------------------------------------------------------------------------------------
2023-10-14 20:44:51,077 Final evaluation on model from best epoch (best-model.pt)
2023-10-14 20:44:51,077 - metric: "('micro avg', 'f1-score')"
2023-10-14 20:44:51,077 ----------------------------------------------------------------------------------------------------
2023-10-14 20:44:51,077 Computation:
2023-10-14 20:44:51,077 - compute on device: cuda:0
2023-10-14 20:44:51,077 - embedding storage: none
2023-10-14 20:44:51,077 ----------------------------------------------------------------------------------------------------
2023-10-14 20:44:51,077 Model training base path: "hmbench-letemps/fr-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-14 20:44:51,077 ----------------------------------------------------------------------------------------------------
2023-10-14 20:44:51,077 ----------------------------------------------------------------------------------------------------
2023-10-14 20:45:08,016 epoch 1 - iter 361/3617 - loss 1.21049643 - time (sec): 16.94 - samples/sec: 2222.06 - lr: 0.000005 - momentum: 0.000000
2023-10-14 20:45:26,819 epoch 1 - iter 722/3617 - loss 0.69887718 - time (sec): 35.74 - samples/sec: 2111.51 - lr: 0.000010 - momentum: 0.000000
2023-10-14 20:45:43,412 epoch 1 - iter 1083/3617 - loss 0.52197079 - time (sec): 52.33 - samples/sec: 2129.26 - lr: 0.000015 - momentum: 0.000000
2023-10-14 20:45:59,782 epoch 1 - iter 1444/3617 - loss 0.42681095 - time (sec): 68.70 - samples/sec: 2161.97 - lr: 0.000020 - momentum: 0.000000
2023-10-14 20:46:16,622 epoch 1 - iter 1805/3617 - loss 0.36380391 - time (sec): 85.54 - samples/sec: 2201.08 - lr: 0.000025 - momentum: 0.000000
2023-10-14 20:46:32,999 epoch 1 - iter 2166/3617 - loss 0.32380187 - time (sec): 101.92 - samples/sec: 2217.20 - lr: 0.000030 - momentum: 0.000000
2023-10-14 20:46:49,761 epoch 1 - iter 2527/3617 - loss 0.29392513 - time (sec): 118.68 - samples/sec: 2240.07 - lr: 0.000035 - momentum: 0.000000
2023-10-14 20:47:05,904 epoch 1 - iter 2888/3617 - loss 0.27259528 - time (sec): 134.83 - samples/sec: 2251.82 - lr: 0.000040 - momentum: 0.000000
2023-10-14 20:47:22,100 epoch 1 - iter 3249/3617 - loss 0.25444813 - time (sec): 151.02 - samples/sec: 2258.31 - lr: 0.000045 - momentum: 0.000000
2023-10-14 20:47:38,309 epoch 1 - iter 3610/3617 - loss 0.24111479 - time (sec): 167.23 - samples/sec: 2267.41 - lr: 0.000050 - momentum: 0.000000
2023-10-14 20:47:38,617 ----------------------------------------------------------------------------------------------------
2023-10-14 20:47:38,617 EPOCH 1 done: loss 0.2409 - lr: 0.000050
2023-10-14 20:47:43,951 DEV : loss 0.12323478609323502 - f1-score (micro avg) 0.5865
2023-10-14 20:47:43,990 saving best model
2023-10-14 20:47:44,376 ----------------------------------------------------------------------------------------------------
2023-10-14 20:48:00,735 epoch 2 - iter 361/3617 - loss 0.10632803 - time (sec): 16.36 - samples/sec: 2344.09 - lr: 0.000049 - momentum: 0.000000
2023-10-14 20:48:16,934 epoch 2 - iter 722/3617 - loss 0.10322207 - time (sec): 32.56 - samples/sec: 2320.21 - lr: 0.000049 - momentum: 0.000000
2023-10-14 20:48:33,240 epoch 2 - iter 1083/3617 - loss 0.10593850 - time (sec): 48.86 - samples/sec: 2307.09 - lr: 0.000048 - momentum: 0.000000
2023-10-14 20:48:50,075 epoch 2 - iter 1444/3617 - loss 0.10664188 - time (sec): 65.70 - samples/sec: 2323.73 - lr: 0.000048 - momentum: 0.000000
2023-10-14 20:49:06,509 epoch 2 - iter 1805/3617 - loss 0.10605572 - time (sec): 82.13 - samples/sec: 2311.63 - lr: 0.000047 - momentum: 0.000000
2023-10-14 20:49:22,832 epoch 2 - iter 2166/3617 - loss 0.10448677 - time (sec): 98.45 - samples/sec: 2324.11 - lr: 0.000047 - momentum: 0.000000
2023-10-14 20:49:39,417 epoch 2 - iter 2527/3617 - loss 0.10397977 - time (sec): 115.04 - samples/sec: 2330.95 - lr: 0.000046 - momentum: 0.000000
2023-10-14 20:49:55,725 epoch 2 - iter 2888/3617 - loss 0.10404474 - time (sec): 131.35 - samples/sec: 2325.67 - lr: 0.000046 - momentum: 0.000000
2023-10-14 20:50:11,903 epoch 2 - iter 3249/3617 - loss 0.10371473 - time (sec): 147.52 - samples/sec: 2313.23 - lr: 0.000045 - momentum: 0.000000
2023-10-14 20:50:28,258 epoch 2 - iter 3610/3617 - loss 0.10388679 - time (sec): 163.88 - samples/sec: 2314.48 - lr: 0.000044 - momentum: 0.000000
2023-10-14 20:50:28,578 ----------------------------------------------------------------------------------------------------
2023-10-14 20:50:28,578 EPOCH 2 done: loss 0.1038 - lr: 0.000044
2023-10-14 20:50:34,850 DEV : loss 0.13535423576831818 - f1-score (micro avg) 0.6246
2023-10-14 20:50:34,883 saving best model
2023-10-14 20:50:35,338 ----------------------------------------------------------------------------------------------------
2023-10-14 20:50:52,031 epoch 3 - iter 361/3617 - loss 0.07083650 - time (sec): 16.69 - samples/sec: 2214.49 - lr: 0.000044 - momentum: 0.000000
2023-10-14 20:51:08,168 epoch 3 - iter 722/3617 - loss 0.07700446 - time (sec): 32.83 - samples/sec: 2276.17 - lr: 0.000043 - momentum: 0.000000
2023-10-14 20:51:24,969 epoch 3 - iter 1083/3617 - loss 0.08564511 - time (sec): 49.63 - samples/sec: 2271.86 - lr: 0.000043 - momentum: 0.000000
2023-10-14 20:51:41,074 epoch 3 - iter 1444/3617 - loss 0.08752690 - time (sec): 65.73 - samples/sec: 2285.24 - lr: 0.000042 - momentum: 0.000000
2023-10-14 20:51:57,079 epoch 3 - iter 1805/3617 - loss 0.08547374 - time (sec): 81.74 - samples/sec: 2299.19 - lr: 0.000042 - momentum: 0.000000
2023-10-14 20:52:13,234 epoch 3 - iter 2166/3617 - loss 0.08336422 - time (sec): 97.89 - samples/sec: 2311.65 - lr: 0.000041 - momentum: 0.000000
2023-10-14 20:52:29,645 epoch 3 - iter 2527/3617 - loss 0.08433411 - time (sec): 114.31 - samples/sec: 2322.61 - lr: 0.000041 - momentum: 0.000000
2023-10-14 20:52:46,055 epoch 3 - iter 2888/3617 - loss 0.08461852 - time (sec): 130.72 - samples/sec: 2322.03 - lr: 0.000040 - momentum: 0.000000
2023-10-14 20:53:04,567 epoch 3 - iter 3249/3617 - loss 0.08613219 - time (sec): 149.23 - samples/sec: 2288.28 - lr: 0.000039 - momentum: 0.000000
2023-10-14 20:53:23,351 epoch 3 - iter 3610/3617 - loss 0.08630119 - time (sec): 168.01 - samples/sec: 2257.30 - lr: 0.000039 - momentum: 0.000000
2023-10-14 20:53:23,704 ----------------------------------------------------------------------------------------------------
2023-10-14 20:53:23,705 EPOCH 3 done: loss 0.0863 - lr: 0.000039
2023-10-14 20:53:30,042 DEV : loss 0.20893967151641846 - f1-score (micro avg) 0.6222
2023-10-14 20:53:30,072 ----------------------------------------------------------------------------------------------------
2023-10-14 20:53:46,674 epoch 4 - iter 361/3617 - loss 0.05714291 - time (sec): 16.60 - samples/sec: 2213.44 - lr: 0.000038 - momentum: 0.000000
2023-10-14 20:54:02,975 epoch 4 - iter 722/3617 - loss 0.06279731 - time (sec): 32.90 - samples/sec: 2283.66 - lr: 0.000038 - momentum: 0.000000
2023-10-14 20:54:19,182 epoch 4 - iter 1083/3617 - loss 0.06156439 - time (sec): 49.11 - samples/sec: 2298.82 - lr: 0.000037 - momentum: 0.000000
2023-10-14 20:54:35,242 epoch 4 - iter 1444/3617 - loss 0.06231476 - time (sec): 65.17 - samples/sec: 2310.81 - lr: 0.000037 - momentum: 0.000000
2023-10-14 20:54:51,551 epoch 4 - iter 1805/3617 - loss 0.06294749 - time (sec): 81.48 - samples/sec: 2330.34 - lr: 0.000036 - momentum: 0.000000
2023-10-14 20:55:07,654 epoch 4 - iter 2166/3617 - loss 0.06283706 - time (sec): 97.58 - samples/sec: 2342.77 - lr: 0.000036 - momentum: 0.000000
2023-10-14 20:55:23,712 epoch 4 - iter 2527/3617 - loss 0.06347340 - time (sec): 113.64 - samples/sec: 2342.74 - lr: 0.000035 - momentum: 0.000000
2023-10-14 20:55:39,816 epoch 4 - iter 2888/3617 - loss 0.06460083 - time (sec): 129.74 - samples/sec: 2348.91 - lr: 0.000034 - momentum: 0.000000
2023-10-14 20:55:56,184 epoch 4 - iter 3249/3617 - loss 0.06589441 - time (sec): 146.11 - samples/sec: 2343.03 - lr: 0.000034 - momentum: 0.000000
2023-10-14 20:56:12,534 epoch 4 - iter 3610/3617 - loss 0.06532681 - time (sec): 162.46 - samples/sec: 2333.94 - lr: 0.000033 - momentum: 0.000000
2023-10-14 20:56:12,840 ----------------------------------------------------------------------------------------------------
2023-10-14 20:56:12,840 EPOCH 4 done: loss 0.0652 - lr: 0.000033
2023-10-14 20:56:20,148 DEV : loss 0.2746644914150238 - f1-score (micro avg) 0.6113
2023-10-14 20:56:20,191 ----------------------------------------------------------------------------------------------------
2023-10-14 20:56:37,963 epoch 5 - iter 361/3617 - loss 0.05107162 - time (sec): 17.77 - samples/sec: 2078.16 - lr: 0.000033 - momentum: 0.000000
2023-10-14 20:56:55,752 epoch 5 - iter 722/3617 - loss 0.04825107 - time (sec): 35.56 - samples/sec: 2145.48 - lr: 0.000032 - momentum: 0.000000
2023-10-14 20:57:12,168 epoch 5 - iter 1083/3617 - loss 0.04805194 - time (sec): 51.97 - samples/sec: 2198.24 - lr: 0.000032 - momentum: 0.000000
2023-10-14 20:57:28,412 epoch 5 - iter 1444/3617 - loss 0.04873195 - time (sec): 68.22 - samples/sec: 2220.55 - lr: 0.000031 - momentum: 0.000000
2023-10-14 20:57:44,708 epoch 5 - iter 1805/3617 - loss 0.04639778 - time (sec): 84.52 - samples/sec: 2230.82 - lr: 0.000031 - momentum: 0.000000
2023-10-14 20:58:01,016 epoch 5 - iter 2166/3617 - loss 0.04761930 - time (sec): 100.82 - samples/sec: 2265.67 - lr: 0.000030 - momentum: 0.000000
2023-10-14 20:58:17,191 epoch 5 - iter 2527/3617 - loss 0.04831656 - time (sec): 117.00 - samples/sec: 2277.51 - lr: 0.000029 - momentum: 0.000000
2023-10-14 20:58:33,296 epoch 5 - iter 2888/3617 - loss 0.04792017 - time (sec): 133.10 - samples/sec: 2288.04 - lr: 0.000029 - momentum: 0.000000
2023-10-14 20:58:49,211 epoch 5 - iter 3249/3617 - loss 0.04879276 - time (sec): 149.02 - samples/sec: 2294.34 - lr: 0.000028 - momentum: 0.000000
2023-10-14 20:59:05,207 epoch 5 - iter 3610/3617 - loss 0.04751972 - time (sec): 165.01 - samples/sec: 2297.36 - lr: 0.000028 - momentum: 0.000000
2023-10-14 20:59:05,509 ----------------------------------------------------------------------------------------------------
2023-10-14 20:59:05,509 EPOCH 5 done: loss 0.0475 - lr: 0.000028
2023-10-14 20:59:11,197 DEV : loss 0.2522399425506592 - f1-score (micro avg) 0.615
2023-10-14 20:59:11,235 ----------------------------------------------------------------------------------------------------
2023-10-14 20:59:27,227 epoch 6 - iter 361/3617 - loss 0.04060810 - time (sec): 15.99 - samples/sec: 2417.59 - lr: 0.000027 - momentum: 0.000000
2023-10-14 20:59:43,109 epoch 6 - iter 722/3617 - loss 0.03470326 - time (sec): 31.87 - samples/sec: 2383.55 - lr: 0.000027 - momentum: 0.000000
2023-10-14 20:59:59,354 epoch 6 - iter 1083/3617 - loss 0.03448545 - time (sec): 48.12 - samples/sec: 2373.29 - lr: 0.000026 - momentum: 0.000000
2023-10-14 21:00:15,500 epoch 6 - iter 1444/3617 - loss 0.03572893 - time (sec): 64.26 - samples/sec: 2345.33 - lr: 0.000026 - momentum: 0.000000
2023-10-14 21:00:32,080 epoch 6 - iter 1805/3617 - loss 0.03790498 - time (sec): 80.84 - samples/sec: 2323.64 - lr: 0.000025 - momentum: 0.000000
2023-10-14 21:00:48,399 epoch 6 - iter 2166/3617 - loss 0.03756533 - time (sec): 97.16 - samples/sec: 2318.62 - lr: 0.000024 - momentum: 0.000000
2023-10-14 21:01:04,962 epoch 6 - iter 2527/3617 - loss 0.03637689 - time (sec): 113.73 - samples/sec: 2311.06 - lr: 0.000024 - momentum: 0.000000
2023-10-14 21:01:21,260 epoch 6 - iter 2888/3617 - loss 0.03628099 - time (sec): 130.02 - samples/sec: 2318.72 - lr: 0.000023 - momentum: 0.000000
2023-10-14 21:01:39,237 epoch 6 - iter 3249/3617 - loss 0.03563678 - time (sec): 148.00 - samples/sec: 2296.92 - lr: 0.000023 - momentum: 0.000000
2023-10-14 21:01:55,520 epoch 6 - iter 3610/3617 - loss 0.03644492 - time (sec): 164.28 - samples/sec: 2308.78 - lr: 0.000022 - momentum: 0.000000
2023-10-14 21:01:55,832 ----------------------------------------------------------------------------------------------------
2023-10-14 21:01:55,832 EPOCH 6 done: loss 0.0364 - lr: 0.000022
2023-10-14 21:02:01,379 DEV : loss 0.29494383931159973 - f1-score (micro avg) 0.6292
2023-10-14 21:02:01,413 saving best model
2023-10-14 21:02:01,869 ----------------------------------------------------------------------------------------------------
2023-10-14 21:02:18,077 epoch 7 - iter 361/3617 - loss 0.03617598 - time (sec): 16.20 - samples/sec: 2387.11 - lr: 0.000022 - momentum: 0.000000
2023-10-14 21:02:34,376 epoch 7 - iter 722/3617 - loss 0.03091675 - time (sec): 32.50 - samples/sec: 2354.05 - lr: 0.000021 - momentum: 0.000000
2023-10-14 21:02:50,448 epoch 7 - iter 1083/3617 - loss 0.02948926 - time (sec): 48.58 - samples/sec: 2350.56 - lr: 0.000021 - momentum: 0.000000
2023-10-14 21:03:07,558 epoch 7 - iter 1444/3617 - loss 0.02797629 - time (sec): 65.69 - samples/sec: 2320.25 - lr: 0.000020 - momentum: 0.000000
2023-10-14 21:03:24,994 epoch 7 - iter 1805/3617 - loss 0.02853652 - time (sec): 83.12 - samples/sec: 2290.94 - lr: 0.000019 - momentum: 0.000000
2023-10-14 21:03:41,182 epoch 7 - iter 2166/3617 - loss 0.02802450 - time (sec): 99.31 - samples/sec: 2291.42 - lr: 0.000019 - momentum: 0.000000
2023-10-14 21:03:57,540 epoch 7 - iter 2527/3617 - loss 0.02675150 - time (sec): 115.67 - samples/sec: 2291.73 - lr: 0.000018 - momentum: 0.000000
2023-10-14 21:04:14,018 epoch 7 - iter 2888/3617 - loss 0.02637225 - time (sec): 132.15 - samples/sec: 2306.72 - lr: 0.000018 - momentum: 0.000000
2023-10-14 21:04:30,290 epoch 7 - iter 3249/3617 - loss 0.02610657 - time (sec): 148.42 - samples/sec: 2304.06 - lr: 0.000017 - momentum: 0.000000
2023-10-14 21:04:46,584 epoch 7 - iter 3610/3617 - loss 0.02552403 - time (sec): 164.71 - samples/sec: 2303.82 - lr: 0.000017 - momentum: 0.000000
2023-10-14 21:04:46,886 ----------------------------------------------------------------------------------------------------
2023-10-14 21:04:46,886 EPOCH 7 done: loss 0.0255 - lr: 0.000017
2023-10-14 21:04:53,190 DEV : loss 0.3446139395236969 - f1-score (micro avg) 0.615
2023-10-14 21:04:53,220 ----------------------------------------------------------------------------------------------------
2023-10-14 21:05:09,555 epoch 8 - iter 361/3617 - loss 0.01650971 - time (sec): 16.33 - samples/sec: 2325.96 - lr: 0.000016 - momentum: 0.000000
2023-10-14 21:05:26,581 epoch 8 - iter 722/3617 - loss 0.01768386 - time (sec): 33.36 - samples/sec: 2297.34 - lr: 0.000016 - momentum: 0.000000
2023-10-14 21:05:42,941 epoch 8 - iter 1083/3617 - loss 0.01765876 - time (sec): 49.72 - samples/sec: 2279.93 - lr: 0.000015 - momentum: 0.000000
2023-10-14 21:05:59,688 epoch 8 - iter 1444/3617 - loss 0.01706874 - time (sec): 66.47 - samples/sec: 2283.33 - lr: 0.000014 - momentum: 0.000000
2023-10-14 21:06:15,920 epoch 8 - iter 1805/3617 - loss 0.01715211 - time (sec): 82.70 - samples/sec: 2302.95 - lr: 0.000014 - momentum: 0.000000
2023-10-14 21:06:32,165 epoch 8 - iter 2166/3617 - loss 0.01796261 - time (sec): 98.94 - samples/sec: 2304.19 - lr: 0.000013 - momentum: 0.000000
2023-10-14 21:06:48,452 epoch 8 - iter 2527/3617 - loss 0.01775176 - time (sec): 115.23 - samples/sec: 2306.13 - lr: 0.000013 - momentum: 0.000000
2023-10-14 21:07:05,449 epoch 8 - iter 2888/3617 - loss 0.01711009 - time (sec): 132.23 - samples/sec: 2300.93 - lr: 0.000012 - momentum: 0.000000
2023-10-14 21:07:22,138 epoch 8 - iter 3249/3617 - loss 0.01724597 - time (sec): 148.92 - samples/sec: 2294.00 - lr: 0.000012 - momentum: 0.000000
2023-10-14 21:07:38,424 epoch 8 - iter 3610/3617 - loss 0.01721940 - time (sec): 165.20 - samples/sec: 2296.29 - lr: 0.000011 - momentum: 0.000000
2023-10-14 21:07:38,727 ----------------------------------------------------------------------------------------------------
2023-10-14 21:07:38,727 EPOCH 8 done: loss 0.0172 - lr: 0.000011
2023-10-14 21:07:46,067 DEV : loss 0.35566461086273193 - f1-score (micro avg) 0.6203
2023-10-14 21:07:46,100 ----------------------------------------------------------------------------------------------------
2023-10-14 21:08:02,982 epoch 9 - iter 361/3617 - loss 0.01334246 - time (sec): 16.88 - samples/sec: 2253.22 - lr: 0.000011 - momentum: 0.000000
2023-10-14 21:08:19,223 epoch 9 - iter 722/3617 - loss 0.01262507 - time (sec): 33.12 - samples/sec: 2305.55 - lr: 0.000010 - momentum: 0.000000
2023-10-14 21:08:35,886 epoch 9 - iter 1083/3617 - loss 0.01123698 - time (sec): 49.78 - samples/sec: 2327.81 - lr: 0.000009 - momentum: 0.000000
2023-10-14 21:08:52,230 epoch 9 - iter 1444/3617 - loss 0.01016665 - time (sec): 66.13 - samples/sec: 2311.20 - lr: 0.000009 - momentum: 0.000000
2023-10-14 21:09:08,463 epoch 9 - iter 1805/3617 - loss 0.01156230 - time (sec): 82.36 - samples/sec: 2304.33 - lr: 0.000008 - momentum: 0.000000
2023-10-14 21:09:24,577 epoch 9 - iter 2166/3617 - loss 0.01168601 - time (sec): 98.48 - samples/sec: 2306.12 - lr: 0.000008 - momentum: 0.000000
2023-10-14 21:09:40,951 epoch 9 - iter 2527/3617 - loss 0.01191755 - time (sec): 114.85 - samples/sec: 2306.41 - lr: 0.000007 - momentum: 0.000000
2023-10-14 21:09:57,560 epoch 9 - iter 2888/3617 - loss 0.01167735 - time (sec): 131.46 - samples/sec: 2303.86 - lr: 0.000007 - momentum: 0.000000
2023-10-14 21:10:14,092 epoch 9 - iter 3249/3617 - loss 0.01159086 - time (sec): 147.99 - samples/sec: 2304.37 - lr: 0.000006 - momentum: 0.000000
2023-10-14 21:10:30,396 epoch 9 - iter 3610/3617 - loss 0.01173161 - time (sec): 164.29 - samples/sec: 2308.88 - lr: 0.000006 - momentum: 0.000000
2023-10-14 21:10:30,693 ----------------------------------------------------------------------------------------------------
2023-10-14 21:10:30,693 EPOCH 9 done: loss 0.0117 - lr: 0.000006
2023-10-14 21:10:37,244 DEV : loss 0.36872944235801697 - f1-score (micro avg) 0.6175
2023-10-14 21:10:37,276 ----------------------------------------------------------------------------------------------------
2023-10-14 21:10:54,918 epoch 10 - iter 361/3617 - loss 0.00910245 - time (sec): 17.64 - samples/sec: 2148.65 - lr: 0.000005 - momentum: 0.000000
2023-10-14 21:11:11,348 epoch 10 - iter 722/3617 - loss 0.00776251 - time (sec): 34.07 - samples/sec: 2213.30 - lr: 0.000004 - momentum: 0.000000
2023-10-14 21:11:28,723 epoch 10 - iter 1083/3617 - loss 0.00798966 - time (sec): 51.45 - samples/sec: 2214.46 - lr: 0.000004 - momentum: 0.000000
2023-10-14 21:11:45,035 epoch 10 - iter 1444/3617 - loss 0.00742183 - time (sec): 67.76 - samples/sec: 2228.67 - lr: 0.000003 - momentum: 0.000000
2023-10-14 21:12:01,459 epoch 10 - iter 1805/3617 - loss 0.00832240 - time (sec): 84.18 - samples/sec: 2252.30 - lr: 0.000003 - momentum: 0.000000
2023-10-14 21:12:17,443 epoch 10 - iter 2166/3617 - loss 0.00906769 - time (sec): 100.17 - samples/sec: 2258.24 - lr: 0.000002 - momentum: 0.000000
2023-10-14 21:12:33,753 epoch 10 - iter 2527/3617 - loss 0.00897717 - time (sec): 116.48 - samples/sec: 2266.17 - lr: 0.000002 - momentum: 0.000000
2023-10-14 21:12:50,112 epoch 10 - iter 2888/3617 - loss 0.00846428 - time (sec): 132.83 - samples/sec: 2283.85 - lr: 0.000001 - momentum: 0.000000
2023-10-14 21:13:06,556 epoch 10 - iter 3249/3617 - loss 0.00812779 - time (sec): 149.28 - samples/sec: 2286.02 - lr: 0.000001 - momentum: 0.000000
2023-10-14 21:13:22,906 epoch 10 - iter 3610/3617 - loss 0.00783079 - time (sec): 165.63 - samples/sec: 2289.65 - lr: 0.000000 - momentum: 0.000000
2023-10-14 21:13:23,217 ----------------------------------------------------------------------------------------------------
2023-10-14 21:13:23,218 EPOCH 10 done: loss 0.0078 - lr: 0.000000
2023-10-14 21:13:30,387 DEV : loss 0.3962736129760742 - f1-score (micro avg) 0.628
2023-10-14 21:13:30,829 ----------------------------------------------------------------------------------------------------
2023-10-14 21:13:30,830 Loading model from best epoch ...
2023-10-14 21:13:32,581 SequenceTagger predicts: Dictionary with 13 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org
2023-10-14 21:13:40,099
Results:
- F-score (micro) 0.6398
- F-score (macro) 0.4724
- Accuracy 0.4905
By class:
precision recall f1-score support
loc 0.6204 0.7631 0.6844 591
pers 0.5745 0.7451 0.6488 357
org 0.1250 0.0633 0.0840 79
micro avg 0.5870 0.7030 0.6398 1027
macro avg 0.4400 0.5238 0.4724 1027
weighted avg 0.5663 0.7030 0.6258 1027
2023-10-14 21:13:40,099 ----------------------------------------------------------------------------------------------------
|