File size: 24,043 Bytes
4f489cf |
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 |
2023-10-14 22:24:23,665 ----------------------------------------------------------------------------------------------------
2023-10-14 22:24:23,681 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 22:24:23,681 ----------------------------------------------------------------------------------------------------
2023-10-14 22:24:23,681 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 22:24:23,681 ----------------------------------------------------------------------------------------------------
2023-10-14 22:24:23,681 Train: 14465 sentences
2023-10-14 22:24:23,681 (train_with_dev=False, train_with_test=False)
2023-10-14 22:24:23,681 ----------------------------------------------------------------------------------------------------
2023-10-14 22:24:23,681 Training Params:
2023-10-14 22:24:23,682 - learning_rate: "5e-05"
2023-10-14 22:24:23,682 - mini_batch_size: "4"
2023-10-14 22:24:23,682 - max_epochs: "10"
2023-10-14 22:24:23,682 - shuffle: "True"
2023-10-14 22:24:23,682 ----------------------------------------------------------------------------------------------------
2023-10-14 22:24:23,682 Plugins:
2023-10-14 22:24:23,682 - LinearScheduler | warmup_fraction: '0.1'
2023-10-14 22:24:23,682 ----------------------------------------------------------------------------------------------------
2023-10-14 22:24:23,682 Final evaluation on model from best epoch (best-model.pt)
2023-10-14 22:24:23,682 - metric: "('micro avg', 'f1-score')"
2023-10-14 22:24:23,682 ----------------------------------------------------------------------------------------------------
2023-10-14 22:24:23,682 Computation:
2023-10-14 22:24:23,682 - compute on device: cuda:0
2023-10-14 22:24:23,682 - embedding storage: none
2023-10-14 22:24:23,682 ----------------------------------------------------------------------------------------------------
2023-10-14 22:24:23,682 Model training base path: "hmbench-letemps/fr-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-14 22:24:23,682 ----------------------------------------------------------------------------------------------------
2023-10-14 22:24:23,682 ----------------------------------------------------------------------------------------------------
2023-10-14 22:24:39,984 epoch 1 - iter 361/3617 - loss 1.14411237 - time (sec): 16.30 - samples/sec: 2350.48 - lr: 0.000005 - momentum: 0.000000
2023-10-14 22:24:56,212 epoch 1 - iter 722/3617 - loss 0.67514192 - time (sec): 32.53 - samples/sec: 2339.73 - lr: 0.000010 - momentum: 0.000000
2023-10-14 22:25:12,297 epoch 1 - iter 1083/3617 - loss 0.50390301 - time (sec): 48.61 - samples/sec: 2328.13 - lr: 0.000015 - momentum: 0.000000
2023-10-14 22:25:28,530 epoch 1 - iter 1444/3617 - loss 0.40601014 - time (sec): 64.85 - samples/sec: 2353.71 - lr: 0.000020 - momentum: 0.000000
2023-10-14 22:25:44,722 epoch 1 - iter 1805/3617 - loss 0.34943623 - time (sec): 81.04 - samples/sec: 2350.57 - lr: 0.000025 - momentum: 0.000000
2023-10-14 22:26:01,008 epoch 1 - iter 2166/3617 - loss 0.31161011 - time (sec): 97.33 - samples/sec: 2361.85 - lr: 0.000030 - momentum: 0.000000
2023-10-14 22:26:17,105 epoch 1 - iter 2527/3617 - loss 0.28529419 - time (sec): 113.42 - samples/sec: 2354.24 - lr: 0.000035 - momentum: 0.000000
2023-10-14 22:26:33,313 epoch 1 - iter 2888/3617 - loss 0.26433511 - time (sec): 129.63 - samples/sec: 2349.50 - lr: 0.000040 - momentum: 0.000000
2023-10-14 22:26:49,604 epoch 1 - iter 3249/3617 - loss 0.24745537 - time (sec): 145.92 - samples/sec: 2345.15 - lr: 0.000045 - momentum: 0.000000
2023-10-14 22:27:05,658 epoch 1 - iter 3610/3617 - loss 0.23480737 - time (sec): 161.98 - samples/sec: 2341.11 - lr: 0.000050 - momentum: 0.000000
2023-10-14 22:27:05,978 ----------------------------------------------------------------------------------------------------
2023-10-14 22:27:05,979 EPOCH 1 done: loss 0.2345 - lr: 0.000050
2023-10-14 22:27:10,539 DEV : loss 0.1204783022403717 - f1-score (micro avg) 0.6165
2023-10-14 22:27:10,568 saving best model
2023-10-14 22:27:11,666 ----------------------------------------------------------------------------------------------------
2023-10-14 22:27:27,870 epoch 2 - iter 361/3617 - loss 0.10944507 - time (sec): 16.20 - samples/sec: 2293.52 - lr: 0.000049 - momentum: 0.000000
2023-10-14 22:27:44,134 epoch 2 - iter 722/3617 - loss 0.11087767 - time (sec): 32.47 - samples/sec: 2324.72 - lr: 0.000049 - momentum: 0.000000
2023-10-14 22:28:00,333 epoch 2 - iter 1083/3617 - loss 0.10916242 - time (sec): 48.67 - samples/sec: 2340.48 - lr: 0.000048 - momentum: 0.000000
2023-10-14 22:28:16,951 epoch 2 - iter 1444/3617 - loss 0.10841737 - time (sec): 65.28 - samples/sec: 2342.27 - lr: 0.000048 - momentum: 0.000000
2023-10-14 22:28:33,376 epoch 2 - iter 1805/3617 - loss 0.10539028 - time (sec): 81.71 - samples/sec: 2348.58 - lr: 0.000047 - momentum: 0.000000
2023-10-14 22:28:49,734 epoch 2 - iter 2166/3617 - loss 0.10557440 - time (sec): 98.07 - samples/sec: 2340.97 - lr: 0.000047 - momentum: 0.000000
2023-10-14 22:29:06,485 epoch 2 - iter 2527/3617 - loss 0.10690549 - time (sec): 114.82 - samples/sec: 2324.15 - lr: 0.000046 - momentum: 0.000000
2023-10-14 22:29:23,693 epoch 2 - iter 2888/3617 - loss 0.10699756 - time (sec): 132.03 - samples/sec: 2293.87 - lr: 0.000046 - momentum: 0.000000
2023-10-14 22:29:40,741 epoch 2 - iter 3249/3617 - loss 0.10562800 - time (sec): 149.07 - samples/sec: 2292.12 - lr: 0.000045 - momentum: 0.000000
2023-10-14 22:29:58,572 epoch 2 - iter 3610/3617 - loss 0.10569211 - time (sec): 166.90 - samples/sec: 2272.40 - lr: 0.000044 - momentum: 0.000000
2023-10-14 22:29:58,905 ----------------------------------------------------------------------------------------------------
2023-10-14 22:29:58,906 EPOCH 2 done: loss 0.1055 - lr: 0.000044
2023-10-14 22:30:04,590 DEV : loss 0.12628547847270966 - f1-score (micro avg) 0.6317
2023-10-14 22:30:04,620 saving best model
2023-10-14 22:30:05,095 ----------------------------------------------------------------------------------------------------
2023-10-14 22:30:22,157 epoch 3 - iter 361/3617 - loss 0.07241240 - time (sec): 17.06 - samples/sec: 2276.69 - lr: 0.000044 - momentum: 0.000000
2023-10-14 22:30:38,462 epoch 3 - iter 722/3617 - loss 0.08210114 - time (sec): 33.36 - samples/sec: 2296.47 - lr: 0.000043 - momentum: 0.000000
2023-10-14 22:30:54,856 epoch 3 - iter 1083/3617 - loss 0.08593425 - time (sec): 49.76 - samples/sec: 2290.48 - lr: 0.000043 - momentum: 0.000000
2023-10-14 22:31:11,133 epoch 3 - iter 1444/3617 - loss 0.08548915 - time (sec): 66.03 - samples/sec: 2313.74 - lr: 0.000042 - momentum: 0.000000
2023-10-14 22:31:27,541 epoch 3 - iter 1805/3617 - loss 0.08461469 - time (sec): 82.44 - samples/sec: 2306.26 - lr: 0.000042 - momentum: 0.000000
2023-10-14 22:31:43,628 epoch 3 - iter 2166/3617 - loss 0.08635041 - time (sec): 98.53 - samples/sec: 2312.85 - lr: 0.000041 - momentum: 0.000000
2023-10-14 22:31:59,971 epoch 3 - iter 2527/3617 - loss 0.08485842 - time (sec): 114.87 - samples/sec: 2312.01 - lr: 0.000041 - momentum: 0.000000
2023-10-14 22:32:16,971 epoch 3 - iter 2888/3617 - loss 0.08549985 - time (sec): 131.87 - samples/sec: 2296.93 - lr: 0.000040 - momentum: 0.000000
2023-10-14 22:32:33,296 epoch 3 - iter 3249/3617 - loss 0.08663415 - time (sec): 148.20 - samples/sec: 2303.61 - lr: 0.000039 - momentum: 0.000000
2023-10-14 22:32:49,490 epoch 3 - iter 3610/3617 - loss 0.08429051 - time (sec): 164.39 - samples/sec: 2306.91 - lr: 0.000039 - momentum: 0.000000
2023-10-14 22:32:49,802 ----------------------------------------------------------------------------------------------------
2023-10-14 22:32:49,802 EPOCH 3 done: loss 0.0842 - lr: 0.000039
2023-10-14 22:32:55,446 DEV : loss 0.20989196002483368 - f1-score (micro avg) 0.6037
2023-10-14 22:32:55,479 ----------------------------------------------------------------------------------------------------
2023-10-14 22:33:12,003 epoch 4 - iter 361/3617 - loss 0.05375131 - time (sec): 16.52 - samples/sec: 2361.28 - lr: 0.000038 - momentum: 0.000000
2023-10-14 22:33:28,293 epoch 4 - iter 722/3617 - loss 0.05179870 - time (sec): 32.81 - samples/sec: 2333.17 - lr: 0.000038 - momentum: 0.000000
2023-10-14 22:33:44,488 epoch 4 - iter 1083/3617 - loss 0.05834655 - time (sec): 49.01 - samples/sec: 2341.86 - lr: 0.000037 - momentum: 0.000000
2023-10-14 22:34:00,581 epoch 4 - iter 1444/3617 - loss 0.06136827 - time (sec): 65.10 - samples/sec: 2328.16 - lr: 0.000037 - momentum: 0.000000
2023-10-14 22:34:16,773 epoch 4 - iter 1805/3617 - loss 0.06036277 - time (sec): 81.29 - samples/sec: 2325.90 - lr: 0.000036 - momentum: 0.000000
2023-10-14 22:34:33,269 epoch 4 - iter 2166/3617 - loss 0.05903167 - time (sec): 97.79 - samples/sec: 2325.15 - lr: 0.000036 - momentum: 0.000000
2023-10-14 22:34:49,619 epoch 4 - iter 2527/3617 - loss 0.06019423 - time (sec): 114.14 - samples/sec: 2326.09 - lr: 0.000035 - momentum: 0.000000
2023-10-14 22:35:05,919 epoch 4 - iter 2888/3617 - loss 0.05917379 - time (sec): 130.44 - samples/sec: 2332.20 - lr: 0.000034 - momentum: 0.000000
2023-10-14 22:35:22,008 epoch 4 - iter 3249/3617 - loss 0.05967316 - time (sec): 146.53 - samples/sec: 2331.84 - lr: 0.000034 - momentum: 0.000000
2023-10-14 22:35:38,187 epoch 4 - iter 3610/3617 - loss 0.06116373 - time (sec): 162.71 - samples/sec: 2330.41 - lr: 0.000033 - momentum: 0.000000
2023-10-14 22:35:38,504 ----------------------------------------------------------------------------------------------------
2023-10-14 22:35:38,504 EPOCH 4 done: loss 0.0612 - lr: 0.000033
2023-10-14 22:35:44,746 DEV : loss 0.1986590474843979 - f1-score (micro avg) 0.5878
2023-10-14 22:35:44,778 ----------------------------------------------------------------------------------------------------
2023-10-14 22:36:01,076 epoch 5 - iter 361/3617 - loss 0.03740724 - time (sec): 16.30 - samples/sec: 2319.97 - lr: 0.000033 - momentum: 0.000000
2023-10-14 22:36:17,138 epoch 5 - iter 722/3617 - loss 0.03966765 - time (sec): 32.36 - samples/sec: 2346.52 - lr: 0.000032 - momentum: 0.000000
2023-10-14 22:36:33,336 epoch 5 - iter 1083/3617 - loss 0.03821815 - time (sec): 48.56 - samples/sec: 2340.28 - lr: 0.000032 - momentum: 0.000000
2023-10-14 22:36:49,708 epoch 5 - iter 1444/3617 - loss 0.04031259 - time (sec): 64.93 - samples/sec: 2319.12 - lr: 0.000031 - momentum: 0.000000
2023-10-14 22:37:05,485 epoch 5 - iter 1805/3617 - loss 0.04174152 - time (sec): 80.71 - samples/sec: 2332.59 - lr: 0.000031 - momentum: 0.000000
2023-10-14 22:37:21,125 epoch 5 - iter 2166/3617 - loss 0.04297498 - time (sec): 96.35 - samples/sec: 2339.87 - lr: 0.000030 - momentum: 0.000000
2023-10-14 22:37:36,708 epoch 5 - iter 2527/3617 - loss 0.04258961 - time (sec): 111.93 - samples/sec: 2344.68 - lr: 0.000029 - momentum: 0.000000
2023-10-14 22:37:52,583 epoch 5 - iter 2888/3617 - loss 0.04372580 - time (sec): 127.80 - samples/sec: 2364.58 - lr: 0.000029 - momentum: 0.000000
2023-10-14 22:38:08,775 epoch 5 - iter 3249/3617 - loss 0.04359994 - time (sec): 144.00 - samples/sec: 2365.75 - lr: 0.000028 - momentum: 0.000000
2023-10-14 22:38:25,264 epoch 5 - iter 3610/3617 - loss 0.04442056 - time (sec): 160.48 - samples/sec: 2364.36 - lr: 0.000028 - momentum: 0.000000
2023-10-14 22:38:25,575 ----------------------------------------------------------------------------------------------------
2023-10-14 22:38:25,575 EPOCH 5 done: loss 0.0444 - lr: 0.000028
2023-10-14 22:38:31,994 DEV : loss 0.3178099989891052 - f1-score (micro avg) 0.6277
2023-10-14 22:38:32,026 ----------------------------------------------------------------------------------------------------
2023-10-14 22:38:48,231 epoch 6 - iter 361/3617 - loss 0.03001037 - time (sec): 16.20 - samples/sec: 2221.52 - lr: 0.000027 - momentum: 0.000000
2023-10-14 22:39:04,620 epoch 6 - iter 722/3617 - loss 0.02850755 - time (sec): 32.59 - samples/sec: 2281.75 - lr: 0.000027 - momentum: 0.000000
2023-10-14 22:39:20,750 epoch 6 - iter 1083/3617 - loss 0.02922986 - time (sec): 48.72 - samples/sec: 2284.28 - lr: 0.000026 - momentum: 0.000000
2023-10-14 22:39:36,936 epoch 6 - iter 1444/3617 - loss 0.02915910 - time (sec): 64.91 - samples/sec: 2298.63 - lr: 0.000026 - momentum: 0.000000
2023-10-14 22:39:53,230 epoch 6 - iter 1805/3617 - loss 0.02933413 - time (sec): 81.20 - samples/sec: 2314.09 - lr: 0.000025 - momentum: 0.000000
2023-10-14 22:40:09,709 epoch 6 - iter 2166/3617 - loss 0.03005081 - time (sec): 97.68 - samples/sec: 2315.74 - lr: 0.000024 - momentum: 0.000000
2023-10-14 22:40:26,048 epoch 6 - iter 2527/3617 - loss 0.02966232 - time (sec): 114.02 - samples/sec: 2326.67 - lr: 0.000024 - momentum: 0.000000
2023-10-14 22:40:42,383 epoch 6 - iter 2888/3617 - loss 0.02981839 - time (sec): 130.36 - samples/sec: 2339.47 - lr: 0.000023 - momentum: 0.000000
2023-10-14 22:40:58,524 epoch 6 - iter 3249/3617 - loss 0.02973966 - time (sec): 146.50 - samples/sec: 2330.85 - lr: 0.000023 - momentum: 0.000000
2023-10-14 22:41:14,900 epoch 6 - iter 3610/3617 - loss 0.02923805 - time (sec): 162.87 - samples/sec: 2328.46 - lr: 0.000022 - momentum: 0.000000
2023-10-14 22:41:15,224 ----------------------------------------------------------------------------------------------------
2023-10-14 22:41:15,224 EPOCH 6 done: loss 0.0292 - lr: 0.000022
2023-10-14 22:41:21,572 DEV : loss 0.33707916736602783 - f1-score (micro avg) 0.6213
2023-10-14 22:41:21,602 ----------------------------------------------------------------------------------------------------
2023-10-14 22:41:37,794 epoch 7 - iter 361/3617 - loss 0.01859802 - time (sec): 16.19 - samples/sec: 2300.51 - lr: 0.000022 - momentum: 0.000000
2023-10-14 22:41:54,033 epoch 7 - iter 722/3617 - loss 0.01903138 - time (sec): 32.43 - samples/sec: 2362.56 - lr: 0.000021 - momentum: 0.000000
2023-10-14 22:42:10,288 epoch 7 - iter 1083/3617 - loss 0.01865096 - time (sec): 48.68 - samples/sec: 2382.51 - lr: 0.000021 - momentum: 0.000000
2023-10-14 22:42:26,441 epoch 7 - iter 1444/3617 - loss 0.02002715 - time (sec): 64.84 - samples/sec: 2354.60 - lr: 0.000020 - momentum: 0.000000
2023-10-14 22:42:42,763 epoch 7 - iter 1805/3617 - loss 0.02137933 - time (sec): 81.16 - samples/sec: 2346.99 - lr: 0.000019 - momentum: 0.000000
2023-10-14 22:42:59,040 epoch 7 - iter 2166/3617 - loss 0.02148678 - time (sec): 97.44 - samples/sec: 2344.17 - lr: 0.000019 - momentum: 0.000000
2023-10-14 22:43:15,412 epoch 7 - iter 2527/3617 - loss 0.02120220 - time (sec): 113.81 - samples/sec: 2350.02 - lr: 0.000018 - momentum: 0.000000
2023-10-14 22:43:31,984 epoch 7 - iter 2888/3617 - loss 0.02087245 - time (sec): 130.38 - samples/sec: 2325.28 - lr: 0.000018 - momentum: 0.000000
2023-10-14 22:43:48,572 epoch 7 - iter 3249/3617 - loss 0.02107711 - time (sec): 146.97 - samples/sec: 2323.76 - lr: 0.000017 - momentum: 0.000000
2023-10-14 22:44:04,971 epoch 7 - iter 3610/3617 - loss 0.02122186 - time (sec): 163.37 - samples/sec: 2320.88 - lr: 0.000017 - momentum: 0.000000
2023-10-14 22:44:05,290 ----------------------------------------------------------------------------------------------------
2023-10-14 22:44:05,290 EPOCH 7 done: loss 0.0212 - lr: 0.000017
2023-10-14 22:44:11,627 DEV : loss 0.361693799495697 - f1-score (micro avg) 0.6369
2023-10-14 22:44:11,657 saving best model
2023-10-14 22:44:12,152 ----------------------------------------------------------------------------------------------------
2023-10-14 22:44:27,823 epoch 8 - iter 361/3617 - loss 0.01351269 - time (sec): 15.67 - samples/sec: 2353.69 - lr: 0.000016 - momentum: 0.000000
2023-10-14 22:44:43,592 epoch 8 - iter 722/3617 - loss 0.01225598 - time (sec): 31.44 - samples/sec: 2406.07 - lr: 0.000016 - momentum: 0.000000
2023-10-14 22:44:59,904 epoch 8 - iter 1083/3617 - loss 0.01341753 - time (sec): 47.75 - samples/sec: 2381.72 - lr: 0.000015 - momentum: 0.000000
2023-10-14 22:45:16,276 epoch 8 - iter 1444/3617 - loss 0.01425253 - time (sec): 64.12 - samples/sec: 2362.31 - lr: 0.000014 - momentum: 0.000000
2023-10-14 22:45:32,508 epoch 8 - iter 1805/3617 - loss 0.01310324 - time (sec): 80.35 - samples/sec: 2346.05 - lr: 0.000014 - momentum: 0.000000
2023-10-14 22:45:48,816 epoch 8 - iter 2166/3617 - loss 0.01431964 - time (sec): 96.66 - samples/sec: 2349.41 - lr: 0.000013 - momentum: 0.000000
2023-10-14 22:46:05,005 epoch 8 - iter 2527/3617 - loss 0.01396577 - time (sec): 112.85 - samples/sec: 2355.77 - lr: 0.000013 - momentum: 0.000000
2023-10-14 22:46:21,476 epoch 8 - iter 2888/3617 - loss 0.01450336 - time (sec): 129.32 - samples/sec: 2344.06 - lr: 0.000012 - momentum: 0.000000
2023-10-14 22:46:37,782 epoch 8 - iter 3249/3617 - loss 0.01434939 - time (sec): 145.63 - samples/sec: 2338.44 - lr: 0.000012 - momentum: 0.000000
2023-10-14 22:46:54,197 epoch 8 - iter 3610/3617 - loss 0.01473032 - time (sec): 162.04 - samples/sec: 2341.62 - lr: 0.000011 - momentum: 0.000000
2023-10-14 22:46:54,511 ----------------------------------------------------------------------------------------------------
2023-10-14 22:46:54,511 EPOCH 8 done: loss 0.0147 - lr: 0.000011
2023-10-14 22:47:00,034 DEV : loss 0.34752506017684937 - f1-score (micro avg) 0.6394
2023-10-14 22:47:00,064 saving best model
2023-10-14 22:47:00,550 ----------------------------------------------------------------------------------------------------
2023-10-14 22:47:17,022 epoch 9 - iter 361/3617 - loss 0.00749889 - time (sec): 16.47 - samples/sec: 2321.67 - lr: 0.000011 - momentum: 0.000000
2023-10-14 22:47:34,147 epoch 9 - iter 722/3617 - loss 0.00782066 - time (sec): 33.59 - samples/sec: 2255.80 - lr: 0.000010 - momentum: 0.000000
2023-10-14 22:47:50,414 epoch 9 - iter 1083/3617 - loss 0.00872382 - time (sec): 49.86 - samples/sec: 2277.97 - lr: 0.000009 - momentum: 0.000000
2023-10-14 22:48:06,823 epoch 9 - iter 1444/3617 - loss 0.00879945 - time (sec): 66.27 - samples/sec: 2300.11 - lr: 0.000009 - momentum: 0.000000
2023-10-14 22:48:23,139 epoch 9 - iter 1805/3617 - loss 0.00900892 - time (sec): 82.58 - samples/sec: 2307.07 - lr: 0.000008 - momentum: 0.000000
2023-10-14 22:48:39,443 epoch 9 - iter 2166/3617 - loss 0.00903195 - time (sec): 98.89 - samples/sec: 2324.67 - lr: 0.000008 - momentum: 0.000000
2023-10-14 22:48:55,699 epoch 9 - iter 2527/3617 - loss 0.00896221 - time (sec): 115.14 - samples/sec: 2324.22 - lr: 0.000007 - momentum: 0.000000
2023-10-14 22:49:11,720 epoch 9 - iter 2888/3617 - loss 0.00879283 - time (sec): 131.16 - samples/sec: 2321.25 - lr: 0.000007 - momentum: 0.000000
2023-10-14 22:49:27,516 epoch 9 - iter 3249/3617 - loss 0.00879301 - time (sec): 146.96 - samples/sec: 2330.96 - lr: 0.000006 - momentum: 0.000000
2023-10-14 22:49:43,105 epoch 9 - iter 3610/3617 - loss 0.00879727 - time (sec): 162.55 - samples/sec: 2333.30 - lr: 0.000006 - momentum: 0.000000
2023-10-14 22:49:43,399 ----------------------------------------------------------------------------------------------------
2023-10-14 22:49:43,399 EPOCH 9 done: loss 0.0088 - lr: 0.000006
2023-10-14 22:49:49,027 DEV : loss 0.3727048337459564 - f1-score (micro avg) 0.642
2023-10-14 22:49:49,058 saving best model
2023-10-14 22:49:49,702 ----------------------------------------------------------------------------------------------------
2023-10-14 22:50:06,227 epoch 10 - iter 361/3617 - loss 0.00310624 - time (sec): 16.52 - samples/sec: 2310.74 - lr: 0.000005 - momentum: 0.000000
2023-10-14 22:50:22,502 epoch 10 - iter 722/3617 - loss 0.00451557 - time (sec): 32.80 - samples/sec: 2328.34 - lr: 0.000004 - momentum: 0.000000
2023-10-14 22:50:38,889 epoch 10 - iter 1083/3617 - loss 0.00444647 - time (sec): 49.19 - samples/sec: 2309.90 - lr: 0.000004 - momentum: 0.000000
2023-10-14 22:50:55,390 epoch 10 - iter 1444/3617 - loss 0.00463820 - time (sec): 65.69 - samples/sec: 2314.50 - lr: 0.000003 - momentum: 0.000000
2023-10-14 22:51:11,711 epoch 10 - iter 1805/3617 - loss 0.00515326 - time (sec): 82.01 - samples/sec: 2320.49 - lr: 0.000003 - momentum: 0.000000
2023-10-14 22:51:28,115 epoch 10 - iter 2166/3617 - loss 0.00547593 - time (sec): 98.41 - samples/sec: 2315.27 - lr: 0.000002 - momentum: 0.000000
2023-10-14 22:51:44,415 epoch 10 - iter 2527/3617 - loss 0.00531709 - time (sec): 114.71 - samples/sec: 2314.93 - lr: 0.000002 - momentum: 0.000000
2023-10-14 22:52:00,847 epoch 10 - iter 2888/3617 - loss 0.00530082 - time (sec): 131.14 - samples/sec: 2319.13 - lr: 0.000001 - momentum: 0.000000
2023-10-14 22:52:16,770 epoch 10 - iter 3249/3617 - loss 0.00559223 - time (sec): 147.07 - samples/sec: 2311.01 - lr: 0.000001 - momentum: 0.000000
2023-10-14 22:52:33,249 epoch 10 - iter 3610/3617 - loss 0.00561927 - time (sec): 163.55 - samples/sec: 2320.26 - lr: 0.000000 - momentum: 0.000000
2023-10-14 22:52:33,559 ----------------------------------------------------------------------------------------------------
2023-10-14 22:52:33,559 EPOCH 10 done: loss 0.0056 - lr: 0.000000
2023-10-14 22:52:39,953 DEV : loss 0.3892402648925781 - f1-score (micro avg) 0.64
2023-10-14 22:52:40,542 ----------------------------------------------------------------------------------------------------
2023-10-14 22:52:40,543 Loading model from best epoch ...
2023-10-14 22:52:42,344 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 22:52:50,738
Results:
- F-score (micro) 0.6467
- F-score (macro) 0.4935
- Accuracy 0.4931
By class:
precision recall f1-score support
loc 0.6436 0.7394 0.6882 591
pers 0.5801 0.7507 0.6545 357
org 0.2162 0.1013 0.1379 79
micro avg 0.6053 0.6943 0.6467 1027
macro avg 0.4800 0.5305 0.4935 1027
weighted avg 0.5886 0.6943 0.6341 1027
2023-10-14 22:52:50,738 ----------------------------------------------------------------------------------------------------
|