File size: 24,001 Bytes
e087466 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 |
2023-10-13 10:59:08,250 ----------------------------------------------------------------------------------------------------
2023-10-13 10:59:08,251 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 10:59:08,251 ----------------------------------------------------------------------------------------------------
2023-10-13 10:59:08,251 MultiCorpus: 966 train + 219 dev + 204 test sentences
- NER_HIPE_2022 Corpus: 966 train + 219 dev + 204 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/fr/with_doc_seperator
2023-10-13 10:59:08,251 ----------------------------------------------------------------------------------------------------
2023-10-13 10:59:08,251 Train: 966 sentences
2023-10-13 10:59:08,251 (train_with_dev=False, train_with_test=False)
2023-10-13 10:59:08,251 ----------------------------------------------------------------------------------------------------
2023-10-13 10:59:08,251 Training Params:
2023-10-13 10:59:08,251 - learning_rate: "5e-05"
2023-10-13 10:59:08,251 - mini_batch_size: "4"
2023-10-13 10:59:08,251 - max_epochs: "10"
2023-10-13 10:59:08,251 - shuffle: "True"
2023-10-13 10:59:08,251 ----------------------------------------------------------------------------------------------------
2023-10-13 10:59:08,251 Plugins:
2023-10-13 10:59:08,252 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 10:59:08,252 ----------------------------------------------------------------------------------------------------
2023-10-13 10:59:08,252 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 10:59:08,252 - metric: "('micro avg', 'f1-score')"
2023-10-13 10:59:08,252 ----------------------------------------------------------------------------------------------------
2023-10-13 10:59:08,252 Computation:
2023-10-13 10:59:08,252 - compute on device: cuda:0
2023-10-13 10:59:08,252 - embedding storage: none
2023-10-13 10:59:08,252 ----------------------------------------------------------------------------------------------------
2023-10-13 10:59:08,252 Model training base path: "hmbench-ajmc/fr-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-13 10:59:08,252 ----------------------------------------------------------------------------------------------------
2023-10-13 10:59:08,252 ----------------------------------------------------------------------------------------------------
2023-10-13 10:59:09,306 epoch 1 - iter 24/242 - loss 3.28733292 - time (sec): 1.05 - samples/sec: 2366.78 - lr: 0.000005 - momentum: 0.000000
2023-10-13 10:59:10,392 epoch 1 - iter 48/242 - loss 2.63402043 - time (sec): 2.14 - samples/sec: 2326.05 - lr: 0.000010 - momentum: 0.000000
2023-10-13 10:59:11,459 epoch 1 - iter 72/242 - loss 2.00586654 - time (sec): 3.21 - samples/sec: 2324.24 - lr: 0.000015 - momentum: 0.000000
2023-10-13 10:59:12,508 epoch 1 - iter 96/242 - loss 1.68954188 - time (sec): 4.26 - samples/sec: 2294.28 - lr: 0.000020 - momentum: 0.000000
2023-10-13 10:59:13,588 epoch 1 - iter 120/242 - loss 1.45281702 - time (sec): 5.33 - samples/sec: 2263.62 - lr: 0.000025 - momentum: 0.000000
2023-10-13 10:59:14,716 epoch 1 - iter 144/242 - loss 1.26827468 - time (sec): 6.46 - samples/sec: 2252.45 - lr: 0.000030 - momentum: 0.000000
2023-10-13 10:59:15,809 epoch 1 - iter 168/242 - loss 1.11884816 - time (sec): 7.56 - samples/sec: 2283.96 - lr: 0.000035 - momentum: 0.000000
2023-10-13 10:59:16,875 epoch 1 - iter 192/242 - loss 1.01409299 - time (sec): 8.62 - samples/sec: 2294.74 - lr: 0.000039 - momentum: 0.000000
2023-10-13 10:59:17,918 epoch 1 - iter 216/242 - loss 0.94260177 - time (sec): 9.67 - samples/sec: 2287.57 - lr: 0.000044 - momentum: 0.000000
2023-10-13 10:59:19,051 epoch 1 - iter 240/242 - loss 0.87025608 - time (sec): 10.80 - samples/sec: 2278.82 - lr: 0.000049 - momentum: 0.000000
2023-10-13 10:59:19,134 ----------------------------------------------------------------------------------------------------
2023-10-13 10:59:19,135 EPOCH 1 done: loss 0.8660 - lr: 0.000049
2023-10-13 10:59:19,874 DEV : loss 0.21214990317821503 - f1-score (micro avg) 0.5594
2023-10-13 10:59:19,879 saving best model
2023-10-13 10:59:20,221 ----------------------------------------------------------------------------------------------------
2023-10-13 10:59:21,273 epoch 2 - iter 24/242 - loss 0.22624820 - time (sec): 1.05 - samples/sec: 2433.95 - lr: 0.000049 - momentum: 0.000000
2023-10-13 10:59:22,350 epoch 2 - iter 48/242 - loss 0.21233454 - time (sec): 2.13 - samples/sec: 2375.85 - lr: 0.000049 - momentum: 0.000000
2023-10-13 10:59:23,396 epoch 2 - iter 72/242 - loss 0.18839820 - time (sec): 3.17 - samples/sec: 2339.79 - lr: 0.000048 - momentum: 0.000000
2023-10-13 10:59:24,448 epoch 2 - iter 96/242 - loss 0.17271089 - time (sec): 4.22 - samples/sec: 2330.52 - lr: 0.000048 - momentum: 0.000000
2023-10-13 10:59:25,552 epoch 2 - iter 120/242 - loss 0.18913709 - time (sec): 5.33 - samples/sec: 2356.25 - lr: 0.000047 - momentum: 0.000000
2023-10-13 10:59:26,636 epoch 2 - iter 144/242 - loss 0.17833993 - time (sec): 6.41 - samples/sec: 2388.36 - lr: 0.000047 - momentum: 0.000000
2023-10-13 10:59:27,689 epoch 2 - iter 168/242 - loss 0.17433724 - time (sec): 7.47 - samples/sec: 2363.68 - lr: 0.000046 - momentum: 0.000000
2023-10-13 10:59:28,762 epoch 2 - iter 192/242 - loss 0.17874942 - time (sec): 8.54 - samples/sec: 2321.69 - lr: 0.000046 - momentum: 0.000000
2023-10-13 10:59:29,824 epoch 2 - iter 216/242 - loss 0.18049272 - time (sec): 9.60 - samples/sec: 2297.49 - lr: 0.000045 - momentum: 0.000000
2023-10-13 10:59:30,905 epoch 2 - iter 240/242 - loss 0.17478721 - time (sec): 10.68 - samples/sec: 2301.40 - lr: 0.000045 - momentum: 0.000000
2023-10-13 10:59:30,990 ----------------------------------------------------------------------------------------------------
2023-10-13 10:59:30,991 EPOCH 2 done: loss 0.1742 - lr: 0.000045
2023-10-13 10:59:31,852 DEV : loss 0.12608306109905243 - f1-score (micro avg) 0.8045
2023-10-13 10:59:31,861 saving best model
2023-10-13 10:59:32,380 ----------------------------------------------------------------------------------------------------
2023-10-13 10:59:33,707 epoch 3 - iter 24/242 - loss 0.09298301 - time (sec): 1.32 - samples/sec: 1779.37 - lr: 0.000044 - momentum: 0.000000
2023-10-13 10:59:34,962 epoch 3 - iter 48/242 - loss 0.10636551 - time (sec): 2.58 - samples/sec: 1926.00 - lr: 0.000043 - momentum: 0.000000
2023-10-13 10:59:36,208 epoch 3 - iter 72/242 - loss 0.10343864 - time (sec): 3.83 - samples/sec: 1932.86 - lr: 0.000043 - momentum: 0.000000
2023-10-13 10:59:37,481 epoch 3 - iter 96/242 - loss 0.10112269 - time (sec): 5.10 - samples/sec: 1923.61 - lr: 0.000042 - momentum: 0.000000
2023-10-13 10:59:38,690 epoch 3 - iter 120/242 - loss 0.10150347 - time (sec): 6.31 - samples/sec: 1929.61 - lr: 0.000042 - momentum: 0.000000
2023-10-13 10:59:40,038 epoch 3 - iter 144/242 - loss 0.10549571 - time (sec): 7.66 - samples/sec: 1952.93 - lr: 0.000041 - momentum: 0.000000
2023-10-13 10:59:41,356 epoch 3 - iter 168/242 - loss 0.10215889 - time (sec): 8.97 - samples/sec: 1917.90 - lr: 0.000041 - momentum: 0.000000
2023-10-13 10:59:42,677 epoch 3 - iter 192/242 - loss 0.10211715 - time (sec): 10.29 - samples/sec: 1910.10 - lr: 0.000040 - momentum: 0.000000
2023-10-13 10:59:44,060 epoch 3 - iter 216/242 - loss 0.09881486 - time (sec): 11.68 - samples/sec: 1888.38 - lr: 0.000040 - momentum: 0.000000
2023-10-13 10:59:45,285 epoch 3 - iter 240/242 - loss 0.09977052 - time (sec): 12.90 - samples/sec: 1910.68 - lr: 0.000039 - momentum: 0.000000
2023-10-13 10:59:45,381 ----------------------------------------------------------------------------------------------------
2023-10-13 10:59:45,381 EPOCH 3 done: loss 0.0996 - lr: 0.000039
2023-10-13 10:59:46,225 DEV : loss 0.1327812224626541 - f1-score (micro avg) 0.8134
2023-10-13 10:59:46,230 saving best model
2023-10-13 10:59:46,704 ----------------------------------------------------------------------------------------------------
2023-10-13 10:59:47,820 epoch 4 - iter 24/242 - loss 0.04194900 - time (sec): 1.11 - samples/sec: 2245.65 - lr: 0.000038 - momentum: 0.000000
2023-10-13 10:59:48,933 epoch 4 - iter 48/242 - loss 0.06746592 - time (sec): 2.23 - samples/sec: 2269.70 - lr: 0.000038 - momentum: 0.000000
2023-10-13 10:59:50,079 epoch 4 - iter 72/242 - loss 0.07175249 - time (sec): 3.37 - samples/sec: 2327.56 - lr: 0.000037 - momentum: 0.000000
2023-10-13 10:59:51,181 epoch 4 - iter 96/242 - loss 0.07971414 - time (sec): 4.48 - samples/sec: 2280.25 - lr: 0.000037 - momentum: 0.000000
2023-10-13 10:59:52,322 epoch 4 - iter 120/242 - loss 0.07312796 - time (sec): 5.62 - samples/sec: 2279.12 - lr: 0.000036 - momentum: 0.000000
2023-10-13 10:59:53,455 epoch 4 - iter 144/242 - loss 0.07512113 - time (sec): 6.75 - samples/sec: 2270.23 - lr: 0.000036 - momentum: 0.000000
2023-10-13 10:59:54,569 epoch 4 - iter 168/242 - loss 0.07531151 - time (sec): 7.86 - samples/sec: 2286.32 - lr: 0.000035 - momentum: 0.000000
2023-10-13 10:59:55,639 epoch 4 - iter 192/242 - loss 0.07672808 - time (sec): 8.93 - samples/sec: 2230.72 - lr: 0.000035 - momentum: 0.000000
2023-10-13 10:59:56,705 epoch 4 - iter 216/242 - loss 0.07417428 - time (sec): 10.00 - samples/sec: 2234.15 - lr: 0.000034 - momentum: 0.000000
2023-10-13 10:59:57,756 epoch 4 - iter 240/242 - loss 0.07268832 - time (sec): 11.05 - samples/sec: 2231.23 - lr: 0.000033 - momentum: 0.000000
2023-10-13 10:59:57,843 ----------------------------------------------------------------------------------------------------
2023-10-13 10:59:57,843 EPOCH 4 done: loss 0.0724 - lr: 0.000033
2023-10-13 10:59:58,635 DEV : loss 0.1568627655506134 - f1-score (micro avg) 0.8312
2023-10-13 10:59:58,640 saving best model
2023-10-13 10:59:59,110 ----------------------------------------------------------------------------------------------------
2023-10-13 11:00:00,176 epoch 5 - iter 24/242 - loss 0.03483374 - time (sec): 1.06 - samples/sec: 1997.81 - lr: 0.000033 - momentum: 0.000000
2023-10-13 11:00:01,249 epoch 5 - iter 48/242 - loss 0.04413526 - time (sec): 2.13 - samples/sec: 2249.23 - lr: 0.000032 - momentum: 0.000000
2023-10-13 11:00:02,337 epoch 5 - iter 72/242 - loss 0.04976873 - time (sec): 3.22 - samples/sec: 2226.61 - lr: 0.000032 - momentum: 0.000000
2023-10-13 11:00:03,426 epoch 5 - iter 96/242 - loss 0.04982655 - time (sec): 4.31 - samples/sec: 2197.10 - lr: 0.000031 - momentum: 0.000000
2023-10-13 11:00:04,523 epoch 5 - iter 120/242 - loss 0.05222925 - time (sec): 5.40 - samples/sec: 2230.99 - lr: 0.000031 - momentum: 0.000000
2023-10-13 11:00:05,620 epoch 5 - iter 144/242 - loss 0.05547688 - time (sec): 6.50 - samples/sec: 2240.20 - lr: 0.000030 - momentum: 0.000000
2023-10-13 11:00:06,710 epoch 5 - iter 168/242 - loss 0.05574000 - time (sec): 7.59 - samples/sec: 2275.39 - lr: 0.000030 - momentum: 0.000000
2023-10-13 11:00:07,768 epoch 5 - iter 192/242 - loss 0.05515866 - time (sec): 8.65 - samples/sec: 2267.04 - lr: 0.000029 - momentum: 0.000000
2023-10-13 11:00:08,874 epoch 5 - iter 216/242 - loss 0.05441600 - time (sec): 9.75 - samples/sec: 2271.25 - lr: 0.000028 - momentum: 0.000000
2023-10-13 11:00:09,967 epoch 5 - iter 240/242 - loss 0.05441985 - time (sec): 10.85 - samples/sec: 2271.98 - lr: 0.000028 - momentum: 0.000000
2023-10-13 11:00:10,060 ----------------------------------------------------------------------------------------------------
2023-10-13 11:00:10,060 EPOCH 5 done: loss 0.0542 - lr: 0.000028
2023-10-13 11:00:10,812 DEV : loss 0.1912143975496292 - f1-score (micro avg) 0.8287
2023-10-13 11:00:10,817 ----------------------------------------------------------------------------------------------------
2023-10-13 11:00:11,938 epoch 6 - iter 24/242 - loss 0.05283667 - time (sec): 1.12 - samples/sec: 2277.81 - lr: 0.000027 - momentum: 0.000000
2023-10-13 11:00:13,012 epoch 6 - iter 48/242 - loss 0.03613276 - time (sec): 2.19 - samples/sec: 2198.78 - lr: 0.000027 - momentum: 0.000000
2023-10-13 11:00:14,098 epoch 6 - iter 72/242 - loss 0.03729743 - time (sec): 3.28 - samples/sec: 2240.13 - lr: 0.000026 - momentum: 0.000000
2023-10-13 11:00:15,232 epoch 6 - iter 96/242 - loss 0.03520942 - time (sec): 4.41 - samples/sec: 2228.42 - lr: 0.000026 - momentum: 0.000000
2023-10-13 11:00:16,386 epoch 6 - iter 120/242 - loss 0.03242642 - time (sec): 5.57 - samples/sec: 2253.95 - lr: 0.000025 - momentum: 0.000000
2023-10-13 11:00:17,478 epoch 6 - iter 144/242 - loss 0.03908906 - time (sec): 6.66 - samples/sec: 2229.69 - lr: 0.000025 - momentum: 0.000000
2023-10-13 11:00:18,575 epoch 6 - iter 168/242 - loss 0.03867449 - time (sec): 7.76 - samples/sec: 2214.95 - lr: 0.000024 - momentum: 0.000000
2023-10-13 11:00:19,656 epoch 6 - iter 192/242 - loss 0.03996907 - time (sec): 8.84 - samples/sec: 2236.10 - lr: 0.000023 - momentum: 0.000000
2023-10-13 11:00:20,703 epoch 6 - iter 216/242 - loss 0.04038098 - time (sec): 9.88 - samples/sec: 2239.86 - lr: 0.000023 - momentum: 0.000000
2023-10-13 11:00:21,779 epoch 6 - iter 240/242 - loss 0.03904930 - time (sec): 10.96 - samples/sec: 2231.07 - lr: 0.000022 - momentum: 0.000000
2023-10-13 11:00:21,872 ----------------------------------------------------------------------------------------------------
2023-10-13 11:00:21,872 EPOCH 6 done: loss 0.0386 - lr: 0.000022
2023-10-13 11:00:22,704 DEV : loss 0.18979744613170624 - f1-score (micro avg) 0.8401
2023-10-13 11:00:22,710 saving best model
2023-10-13 11:00:23,179 ----------------------------------------------------------------------------------------------------
2023-10-13 11:00:24,347 epoch 7 - iter 24/242 - loss 0.01995235 - time (sec): 1.16 - samples/sec: 2138.90 - lr: 0.000022 - momentum: 0.000000
2023-10-13 11:00:25,425 epoch 7 - iter 48/242 - loss 0.01639084 - time (sec): 2.24 - samples/sec: 2054.26 - lr: 0.000021 - momentum: 0.000000
2023-10-13 11:00:26,499 epoch 7 - iter 72/242 - loss 0.01766459 - time (sec): 3.31 - samples/sec: 2147.70 - lr: 0.000021 - momentum: 0.000000
2023-10-13 11:00:27,634 epoch 7 - iter 96/242 - loss 0.02077422 - time (sec): 4.45 - samples/sec: 2208.62 - lr: 0.000020 - momentum: 0.000000
2023-10-13 11:00:28,711 epoch 7 - iter 120/242 - loss 0.02499053 - time (sec): 5.53 - samples/sec: 2172.53 - lr: 0.000020 - momentum: 0.000000
2023-10-13 11:00:29,804 epoch 7 - iter 144/242 - loss 0.02688253 - time (sec): 6.62 - samples/sec: 2185.03 - lr: 0.000019 - momentum: 0.000000
2023-10-13 11:00:30,915 epoch 7 - iter 168/242 - loss 0.02690714 - time (sec): 7.73 - samples/sec: 2193.29 - lr: 0.000018 - momentum: 0.000000
2023-10-13 11:00:32,036 epoch 7 - iter 192/242 - loss 0.02722919 - time (sec): 8.85 - samples/sec: 2215.68 - lr: 0.000018 - momentum: 0.000000
2023-10-13 11:00:33,112 epoch 7 - iter 216/242 - loss 0.02576495 - time (sec): 9.93 - samples/sec: 2231.01 - lr: 0.000017 - momentum: 0.000000
2023-10-13 11:00:34,219 epoch 7 - iter 240/242 - loss 0.02602469 - time (sec): 11.03 - samples/sec: 2228.97 - lr: 0.000017 - momentum: 0.000000
2023-10-13 11:00:34,306 ----------------------------------------------------------------------------------------------------
2023-10-13 11:00:34,306 EPOCH 7 done: loss 0.0259 - lr: 0.000017
2023-10-13 11:00:35,271 DEV : loss 0.1991121768951416 - f1-score (micro avg) 0.8315
2023-10-13 11:00:35,277 ----------------------------------------------------------------------------------------------------
2023-10-13 11:00:36,352 epoch 8 - iter 24/242 - loss 0.01398267 - time (sec): 1.07 - samples/sec: 2428.11 - lr: 0.000016 - momentum: 0.000000
2023-10-13 11:00:37,410 epoch 8 - iter 48/242 - loss 0.02187050 - time (sec): 2.13 - samples/sec: 2261.18 - lr: 0.000016 - momentum: 0.000000
2023-10-13 11:00:38,497 epoch 8 - iter 72/242 - loss 0.02478786 - time (sec): 3.22 - samples/sec: 2316.34 - lr: 0.000015 - momentum: 0.000000
2023-10-13 11:00:39,565 epoch 8 - iter 96/242 - loss 0.02290241 - time (sec): 4.29 - samples/sec: 2350.71 - lr: 0.000015 - momentum: 0.000000
2023-10-13 11:00:40,644 epoch 8 - iter 120/242 - loss 0.01921730 - time (sec): 5.37 - samples/sec: 2302.90 - lr: 0.000014 - momentum: 0.000000
2023-10-13 11:00:41,758 epoch 8 - iter 144/242 - loss 0.01872496 - time (sec): 6.48 - samples/sec: 2327.98 - lr: 0.000013 - momentum: 0.000000
2023-10-13 11:00:42,819 epoch 8 - iter 168/242 - loss 0.01731024 - time (sec): 7.54 - samples/sec: 2299.72 - lr: 0.000013 - momentum: 0.000000
2023-10-13 11:00:43,890 epoch 8 - iter 192/242 - loss 0.01778436 - time (sec): 8.61 - samples/sec: 2301.69 - lr: 0.000012 - momentum: 0.000000
2023-10-13 11:00:44,962 epoch 8 - iter 216/242 - loss 0.01877575 - time (sec): 9.68 - samples/sec: 2299.60 - lr: 0.000012 - momentum: 0.000000
2023-10-13 11:00:46,026 epoch 8 - iter 240/242 - loss 0.01890326 - time (sec): 10.75 - samples/sec: 2292.00 - lr: 0.000011 - momentum: 0.000000
2023-10-13 11:00:46,109 ----------------------------------------------------------------------------------------------------
2023-10-13 11:00:46,109 EPOCH 8 done: loss 0.0188 - lr: 0.000011
2023-10-13 11:00:46,955 DEV : loss 0.18604768812656403 - f1-score (micro avg) 0.8455
2023-10-13 11:00:46,960 saving best model
2023-10-13 11:00:47,419 ----------------------------------------------------------------------------------------------------
2023-10-13 11:00:48,560 epoch 9 - iter 24/242 - loss 0.02467621 - time (sec): 1.14 - samples/sec: 1974.79 - lr: 0.000011 - momentum: 0.000000
2023-10-13 11:00:49,631 epoch 9 - iter 48/242 - loss 0.01478693 - time (sec): 2.21 - samples/sec: 2156.92 - lr: 0.000010 - momentum: 0.000000
2023-10-13 11:00:50,751 epoch 9 - iter 72/242 - loss 0.01735104 - time (sec): 3.33 - samples/sec: 2179.58 - lr: 0.000010 - momentum: 0.000000
2023-10-13 11:00:51,852 epoch 9 - iter 96/242 - loss 0.01333860 - time (sec): 4.43 - samples/sec: 2240.98 - lr: 0.000009 - momentum: 0.000000
2023-10-13 11:00:52,943 epoch 9 - iter 120/242 - loss 0.01223139 - time (sec): 5.52 - samples/sec: 2226.98 - lr: 0.000008 - momentum: 0.000000
2023-10-13 11:00:54,061 epoch 9 - iter 144/242 - loss 0.01157016 - time (sec): 6.64 - samples/sec: 2213.05 - lr: 0.000008 - momentum: 0.000000
2023-10-13 11:00:55,220 epoch 9 - iter 168/242 - loss 0.01108253 - time (sec): 7.80 - samples/sec: 2215.59 - lr: 0.000007 - momentum: 0.000000
2023-10-13 11:00:56,375 epoch 9 - iter 192/242 - loss 0.01061662 - time (sec): 8.95 - samples/sec: 2224.12 - lr: 0.000007 - momentum: 0.000000
2023-10-13 11:00:57,494 epoch 9 - iter 216/242 - loss 0.01072204 - time (sec): 10.07 - samples/sec: 2199.31 - lr: 0.000006 - momentum: 0.000000
2023-10-13 11:00:58,615 epoch 9 - iter 240/242 - loss 0.01115647 - time (sec): 11.19 - samples/sec: 2194.00 - lr: 0.000006 - momentum: 0.000000
2023-10-13 11:00:58,699 ----------------------------------------------------------------------------------------------------
2023-10-13 11:00:58,700 EPOCH 9 done: loss 0.0111 - lr: 0.000006
2023-10-13 11:00:59,516 DEV : loss 0.1872359663248062 - f1-score (micro avg) 0.8451
2023-10-13 11:00:59,522 ----------------------------------------------------------------------------------------------------
2023-10-13 11:01:00,652 epoch 10 - iter 24/242 - loss 0.00276810 - time (sec): 1.13 - samples/sec: 2006.47 - lr: 0.000005 - momentum: 0.000000
2023-10-13 11:01:01,740 epoch 10 - iter 48/242 - loss 0.00426126 - time (sec): 2.22 - samples/sec: 2154.76 - lr: 0.000005 - momentum: 0.000000
2023-10-13 11:01:02,891 epoch 10 - iter 72/242 - loss 0.00479354 - time (sec): 3.37 - samples/sec: 2183.96 - lr: 0.000004 - momentum: 0.000000
2023-10-13 11:01:04,056 epoch 10 - iter 96/242 - loss 0.01271530 - time (sec): 4.53 - samples/sec: 2172.12 - lr: 0.000003 - momentum: 0.000000
2023-10-13 11:01:05,129 epoch 10 - iter 120/242 - loss 0.01214986 - time (sec): 5.61 - samples/sec: 2153.56 - lr: 0.000003 - momentum: 0.000000
2023-10-13 11:01:06,195 epoch 10 - iter 144/242 - loss 0.01140887 - time (sec): 6.67 - samples/sec: 2108.49 - lr: 0.000002 - momentum: 0.000000
2023-10-13 11:01:07,303 epoch 10 - iter 168/242 - loss 0.00972095 - time (sec): 7.78 - samples/sec: 2146.63 - lr: 0.000002 - momentum: 0.000000
2023-10-13 11:01:08,402 epoch 10 - iter 192/242 - loss 0.00946718 - time (sec): 8.88 - samples/sec: 2155.32 - lr: 0.000001 - momentum: 0.000000
2023-10-13 11:01:09,485 epoch 10 - iter 216/242 - loss 0.00880560 - time (sec): 9.96 - samples/sec: 2184.60 - lr: 0.000001 - momentum: 0.000000
2023-10-13 11:01:10,583 epoch 10 - iter 240/242 - loss 0.00862072 - time (sec): 11.06 - samples/sec: 2219.90 - lr: 0.000000 - momentum: 0.000000
2023-10-13 11:01:10,668 ----------------------------------------------------------------------------------------------------
2023-10-13 11:01:10,668 EPOCH 10 done: loss 0.0086 - lr: 0.000000
2023-10-13 11:01:11,517 DEV : loss 0.19478590786457062 - f1-score (micro avg) 0.835
2023-10-13 11:01:11,880 ----------------------------------------------------------------------------------------------------
2023-10-13 11:01:11,881 Loading model from best epoch ...
2023-10-13 11:01:13,251 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 11:01:14,070
Results:
- F-score (micro) 0.8232
- F-score (macro) 0.4943
- Accuracy 0.7193
By class:
precision recall f1-score support
pers 0.8611 0.8921 0.8763 139
scope 0.8321 0.8837 0.8571 129
work 0.6848 0.7875 0.7326 80
loc 0.5714 0.4444 0.5000 9
date 0.0000 0.0000 0.0000 3
object 0.0000 0.0000 0.0000 0
micro avg 0.8005 0.8472 0.8232 360
macro avg 0.4916 0.5013 0.4943 360
weighted avg 0.7971 0.8472 0.8208 360
2023-10-13 11:01:14,070 ----------------------------------------------------------------------------------------------------
|