File size: 24,160 Bytes
075c173 |
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 |
2023-10-17 09:54:13,867 ----------------------------------------------------------------------------------------------------
2023-10-17 09:54:13,868 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): ElectraModel(
(embeddings): ElectraEmbeddings(
(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): ElectraEncoder(
(layer): ModuleList(
(0-11): 12 x ElectraLayer(
(attention): ElectraAttention(
(self): ElectraSelfAttention(
(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): ElectraSelfOutput(
(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): ElectraIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): ElectraOutput(
(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)
)
)
)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 09:54:13,868 ----------------------------------------------------------------------------------------------------
2023-10-17 09:54:13,869 MultiCorpus: 1214 train + 266 dev + 251 test sentences
- NER_HIPE_2022 Corpus: 1214 train + 266 dev + 251 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/en/with_doc_seperator
2023-10-17 09:54:13,869 ----------------------------------------------------------------------------------------------------
2023-10-17 09:54:13,869 Train: 1214 sentences
2023-10-17 09:54:13,869 (train_with_dev=False, train_with_test=False)
2023-10-17 09:54:13,869 ----------------------------------------------------------------------------------------------------
2023-10-17 09:54:13,869 Training Params:
2023-10-17 09:54:13,869 - learning_rate: "5e-05"
2023-10-17 09:54:13,869 - mini_batch_size: "4"
2023-10-17 09:54:13,869 - max_epochs: "10"
2023-10-17 09:54:13,869 - shuffle: "True"
2023-10-17 09:54:13,869 ----------------------------------------------------------------------------------------------------
2023-10-17 09:54:13,869 Plugins:
2023-10-17 09:54:13,869 - TensorboardLogger
2023-10-17 09:54:13,869 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 09:54:13,869 ----------------------------------------------------------------------------------------------------
2023-10-17 09:54:13,869 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 09:54:13,869 - metric: "('micro avg', 'f1-score')"
2023-10-17 09:54:13,869 ----------------------------------------------------------------------------------------------------
2023-10-17 09:54:13,869 Computation:
2023-10-17 09:54:13,869 - compute on device: cuda:0
2023-10-17 09:54:13,869 - embedding storage: none
2023-10-17 09:54:13,869 ----------------------------------------------------------------------------------------------------
2023-10-17 09:54:13,869 Model training base path: "hmbench-ajmc/en-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-17 09:54:13,869 ----------------------------------------------------------------------------------------------------
2023-10-17 09:54:13,869 ----------------------------------------------------------------------------------------------------
2023-10-17 09:54:13,869 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 09:54:15,461 epoch 1 - iter 30/304 - loss 3.92541070 - time (sec): 1.59 - samples/sec: 1953.57 - lr: 0.000005 - momentum: 0.000000
2023-10-17 09:54:17,033 epoch 1 - iter 60/304 - loss 2.83002534 - time (sec): 3.16 - samples/sec: 1982.33 - lr: 0.000010 - momentum: 0.000000
2023-10-17 09:54:18,613 epoch 1 - iter 90/304 - loss 2.09110440 - time (sec): 4.74 - samples/sec: 1977.32 - lr: 0.000015 - momentum: 0.000000
2023-10-17 09:54:20,180 epoch 1 - iter 120/304 - loss 1.68334725 - time (sec): 6.31 - samples/sec: 1977.51 - lr: 0.000020 - momentum: 0.000000
2023-10-17 09:54:21,832 epoch 1 - iter 150/304 - loss 1.43709984 - time (sec): 7.96 - samples/sec: 1960.01 - lr: 0.000025 - momentum: 0.000000
2023-10-17 09:54:23,389 epoch 1 - iter 180/304 - loss 1.26564420 - time (sec): 9.52 - samples/sec: 1942.99 - lr: 0.000029 - momentum: 0.000000
2023-10-17 09:54:24,929 epoch 1 - iter 210/304 - loss 1.12056567 - time (sec): 11.06 - samples/sec: 1939.18 - lr: 0.000034 - momentum: 0.000000
2023-10-17 09:54:26,530 epoch 1 - iter 240/304 - loss 1.02447655 - time (sec): 12.66 - samples/sec: 1927.22 - lr: 0.000039 - momentum: 0.000000
2023-10-17 09:54:28,082 epoch 1 - iter 270/304 - loss 0.93337178 - time (sec): 14.21 - samples/sec: 1937.90 - lr: 0.000044 - momentum: 0.000000
2023-10-17 09:54:29,676 epoch 1 - iter 300/304 - loss 0.85762804 - time (sec): 15.81 - samples/sec: 1937.78 - lr: 0.000049 - momentum: 0.000000
2023-10-17 09:54:29,873 ----------------------------------------------------------------------------------------------------
2023-10-17 09:54:29,873 EPOCH 1 done: loss 0.8498 - lr: 0.000049
2023-10-17 09:54:30,898 DEV : loss 0.22860193252563477 - f1-score (micro avg) 0.6309
2023-10-17 09:54:30,907 saving best model
2023-10-17 09:54:31,343 ----------------------------------------------------------------------------------------------------
2023-10-17 09:54:32,716 epoch 2 - iter 30/304 - loss 0.17256119 - time (sec): 1.37 - samples/sec: 2189.96 - lr: 0.000049 - momentum: 0.000000
2023-10-17 09:54:34,134 epoch 2 - iter 60/304 - loss 0.17606219 - time (sec): 2.79 - samples/sec: 2176.91 - lr: 0.000049 - momentum: 0.000000
2023-10-17 09:54:35,573 epoch 2 - iter 90/304 - loss 0.16356435 - time (sec): 4.23 - samples/sec: 2168.56 - lr: 0.000048 - momentum: 0.000000
2023-10-17 09:54:37,179 epoch 2 - iter 120/304 - loss 0.15726544 - time (sec): 5.84 - samples/sec: 2064.08 - lr: 0.000048 - momentum: 0.000000
2023-10-17 09:54:38,816 epoch 2 - iter 150/304 - loss 0.15305299 - time (sec): 7.47 - samples/sec: 2045.50 - lr: 0.000047 - momentum: 0.000000
2023-10-17 09:54:40,335 epoch 2 - iter 180/304 - loss 0.14349988 - time (sec): 8.99 - samples/sec: 2048.80 - lr: 0.000047 - momentum: 0.000000
2023-10-17 09:54:41,745 epoch 2 - iter 210/304 - loss 0.14441468 - time (sec): 10.40 - samples/sec: 2073.44 - lr: 0.000046 - momentum: 0.000000
2023-10-17 09:54:43,222 epoch 2 - iter 240/304 - loss 0.13509089 - time (sec): 11.88 - samples/sec: 2071.61 - lr: 0.000046 - momentum: 0.000000
2023-10-17 09:54:44,816 epoch 2 - iter 270/304 - loss 0.13165195 - time (sec): 13.47 - samples/sec: 2051.25 - lr: 0.000045 - momentum: 0.000000
2023-10-17 09:54:46,437 epoch 2 - iter 300/304 - loss 0.13298263 - time (sec): 15.09 - samples/sec: 2031.22 - lr: 0.000045 - momentum: 0.000000
2023-10-17 09:54:46,656 ----------------------------------------------------------------------------------------------------
2023-10-17 09:54:46,656 EPOCH 2 done: loss 0.1329 - lr: 0.000045
2023-10-17 09:54:47,681 DEV : loss 0.16872060298919678 - f1-score (micro avg) 0.7932
2023-10-17 09:54:47,690 saving best model
2023-10-17 09:54:48,238 ----------------------------------------------------------------------------------------------------
2023-10-17 09:54:49,636 epoch 3 - iter 30/304 - loss 0.05274848 - time (sec): 1.40 - samples/sec: 2056.34 - lr: 0.000044 - momentum: 0.000000
2023-10-17 09:54:51,055 epoch 3 - iter 60/304 - loss 0.05982692 - time (sec): 2.81 - samples/sec: 2096.78 - lr: 0.000043 - momentum: 0.000000
2023-10-17 09:54:52,612 epoch 3 - iter 90/304 - loss 0.08228454 - time (sec): 4.37 - samples/sec: 2012.76 - lr: 0.000043 - momentum: 0.000000
2023-10-17 09:54:54,133 epoch 3 - iter 120/304 - loss 0.08042581 - time (sec): 5.89 - samples/sec: 2033.82 - lr: 0.000042 - momentum: 0.000000
2023-10-17 09:54:55,719 epoch 3 - iter 150/304 - loss 0.08178362 - time (sec): 7.48 - samples/sec: 2028.63 - lr: 0.000042 - momentum: 0.000000
2023-10-17 09:54:57,300 epoch 3 - iter 180/304 - loss 0.07941139 - time (sec): 9.06 - samples/sec: 2005.37 - lr: 0.000041 - momentum: 0.000000
2023-10-17 09:54:58,925 epoch 3 - iter 210/304 - loss 0.08265085 - time (sec): 10.69 - samples/sec: 1988.99 - lr: 0.000041 - momentum: 0.000000
2023-10-17 09:55:00,536 epoch 3 - iter 240/304 - loss 0.08573224 - time (sec): 12.30 - samples/sec: 1982.40 - lr: 0.000040 - momentum: 0.000000
2023-10-17 09:55:02,133 epoch 3 - iter 270/304 - loss 0.08997078 - time (sec): 13.89 - samples/sec: 1983.58 - lr: 0.000040 - momentum: 0.000000
2023-10-17 09:55:03,763 epoch 3 - iter 300/304 - loss 0.09055316 - time (sec): 15.52 - samples/sec: 1973.09 - lr: 0.000039 - momentum: 0.000000
2023-10-17 09:55:03,972 ----------------------------------------------------------------------------------------------------
2023-10-17 09:55:03,972 EPOCH 3 done: loss 0.0900 - lr: 0.000039
2023-10-17 09:55:04,903 DEV : loss 0.15869022905826569 - f1-score (micro avg) 0.8223
2023-10-17 09:55:04,910 saving best model
2023-10-17 09:55:05,438 ----------------------------------------------------------------------------------------------------
2023-10-17 09:55:07,051 epoch 4 - iter 30/304 - loss 0.22243641 - time (sec): 1.61 - samples/sec: 1882.75 - lr: 0.000038 - momentum: 0.000000
2023-10-17 09:55:08,660 epoch 4 - iter 60/304 - loss 0.18033808 - time (sec): 3.22 - samples/sec: 1820.17 - lr: 0.000038 - momentum: 0.000000
2023-10-17 09:55:10,291 epoch 4 - iter 90/304 - loss 0.14561604 - time (sec): 4.85 - samples/sec: 1879.90 - lr: 0.000037 - momentum: 0.000000
2023-10-17 09:55:11,877 epoch 4 - iter 120/304 - loss 0.11864431 - time (sec): 6.44 - samples/sec: 1918.44 - lr: 0.000037 - momentum: 0.000000
2023-10-17 09:55:13,540 epoch 4 - iter 150/304 - loss 0.10256781 - time (sec): 8.10 - samples/sec: 1903.01 - lr: 0.000036 - momentum: 0.000000
2023-10-17 09:55:15,143 epoch 4 - iter 180/304 - loss 0.09530623 - time (sec): 9.70 - samples/sec: 1894.36 - lr: 0.000036 - momentum: 0.000000
2023-10-17 09:55:16,608 epoch 4 - iter 210/304 - loss 0.09260037 - time (sec): 11.17 - samples/sec: 1937.91 - lr: 0.000035 - momentum: 0.000000
2023-10-17 09:55:18,006 epoch 4 - iter 240/304 - loss 0.08909074 - time (sec): 12.57 - samples/sec: 1956.66 - lr: 0.000035 - momentum: 0.000000
2023-10-17 09:55:19,439 epoch 4 - iter 270/304 - loss 0.08554726 - time (sec): 14.00 - samples/sec: 1961.12 - lr: 0.000034 - momentum: 0.000000
2023-10-17 09:55:20,866 epoch 4 - iter 300/304 - loss 0.08568790 - time (sec): 15.43 - samples/sec: 1983.06 - lr: 0.000033 - momentum: 0.000000
2023-10-17 09:55:21,051 ----------------------------------------------------------------------------------------------------
2023-10-17 09:55:21,051 EPOCH 4 done: loss 0.0880 - lr: 0.000033
2023-10-17 09:55:21,991 DEV : loss 0.1595626175403595 - f1-score (micro avg) 0.8343
2023-10-17 09:55:21,998 saving best model
2023-10-17 09:55:22,546 ----------------------------------------------------------------------------------------------------
2023-10-17 09:55:24,058 epoch 5 - iter 30/304 - loss 0.03697545 - time (sec): 1.51 - samples/sec: 1901.02 - lr: 0.000033 - momentum: 0.000000
2023-10-17 09:55:25,463 epoch 5 - iter 60/304 - loss 0.03777380 - time (sec): 2.92 - samples/sec: 2122.32 - lr: 0.000032 - momentum: 0.000000
2023-10-17 09:55:26,930 epoch 5 - iter 90/304 - loss 0.04110514 - time (sec): 4.38 - samples/sec: 2116.43 - lr: 0.000032 - momentum: 0.000000
2023-10-17 09:55:28,524 epoch 5 - iter 120/304 - loss 0.03961826 - time (sec): 5.98 - samples/sec: 2026.55 - lr: 0.000031 - momentum: 0.000000
2023-10-17 09:55:30,130 epoch 5 - iter 150/304 - loss 0.03751316 - time (sec): 7.58 - samples/sec: 2000.79 - lr: 0.000031 - momentum: 0.000000
2023-10-17 09:55:31,722 epoch 5 - iter 180/304 - loss 0.03548953 - time (sec): 9.17 - samples/sec: 1999.72 - lr: 0.000030 - momentum: 0.000000
2023-10-17 09:55:33,330 epoch 5 - iter 210/304 - loss 0.03902609 - time (sec): 10.78 - samples/sec: 1993.15 - lr: 0.000030 - momentum: 0.000000
2023-10-17 09:55:34,924 epoch 5 - iter 240/304 - loss 0.04171248 - time (sec): 12.38 - samples/sec: 1983.58 - lr: 0.000029 - momentum: 0.000000
2023-10-17 09:55:36,293 epoch 5 - iter 270/304 - loss 0.04498475 - time (sec): 13.75 - samples/sec: 2006.22 - lr: 0.000028 - momentum: 0.000000
2023-10-17 09:55:37,674 epoch 5 - iter 300/304 - loss 0.04539674 - time (sec): 15.13 - samples/sec: 2029.85 - lr: 0.000028 - momentum: 0.000000
2023-10-17 09:55:37,866 ----------------------------------------------------------------------------------------------------
2023-10-17 09:55:37,867 EPOCH 5 done: loss 0.0451 - lr: 0.000028
2023-10-17 09:55:38,795 DEV : loss 0.20325231552124023 - f1-score (micro avg) 0.8329
2023-10-17 09:55:38,802 ----------------------------------------------------------------------------------------------------
2023-10-17 09:55:40,200 epoch 6 - iter 30/304 - loss 0.02360420 - time (sec): 1.40 - samples/sec: 2411.55 - lr: 0.000027 - momentum: 0.000000
2023-10-17 09:55:41,733 epoch 6 - iter 60/304 - loss 0.05298460 - time (sec): 2.93 - samples/sec: 2193.31 - lr: 0.000027 - momentum: 0.000000
2023-10-17 09:55:43,186 epoch 6 - iter 90/304 - loss 0.04447507 - time (sec): 4.38 - samples/sec: 2108.37 - lr: 0.000026 - momentum: 0.000000
2023-10-17 09:55:44,803 epoch 6 - iter 120/304 - loss 0.04234440 - time (sec): 6.00 - samples/sec: 1974.90 - lr: 0.000026 - momentum: 0.000000
2023-10-17 09:55:46,379 epoch 6 - iter 150/304 - loss 0.04134043 - time (sec): 7.58 - samples/sec: 1975.51 - lr: 0.000025 - momentum: 0.000000
2023-10-17 09:55:47,935 epoch 6 - iter 180/304 - loss 0.03644266 - time (sec): 9.13 - samples/sec: 1981.53 - lr: 0.000025 - momentum: 0.000000
2023-10-17 09:55:49,333 epoch 6 - iter 210/304 - loss 0.03483795 - time (sec): 10.53 - samples/sec: 2026.41 - lr: 0.000024 - momentum: 0.000000
2023-10-17 09:55:50,701 epoch 6 - iter 240/304 - loss 0.03394206 - time (sec): 11.90 - samples/sec: 2018.44 - lr: 0.000023 - momentum: 0.000000
2023-10-17 09:55:52,129 epoch 6 - iter 270/304 - loss 0.03112237 - time (sec): 13.33 - samples/sec: 2054.51 - lr: 0.000023 - momentum: 0.000000
2023-10-17 09:55:53,532 epoch 6 - iter 300/304 - loss 0.03255731 - time (sec): 14.73 - samples/sec: 2074.43 - lr: 0.000022 - momentum: 0.000000
2023-10-17 09:55:53,717 ----------------------------------------------------------------------------------------------------
2023-10-17 09:55:53,717 EPOCH 6 done: loss 0.0331 - lr: 0.000022
2023-10-17 09:55:54,653 DEV : loss 0.19660867750644684 - f1-score (micro avg) 0.8392
2023-10-17 09:55:54,661 saving best model
2023-10-17 09:55:55,147 ----------------------------------------------------------------------------------------------------
2023-10-17 09:55:56,599 epoch 7 - iter 30/304 - loss 0.00971525 - time (sec): 1.45 - samples/sec: 2108.25 - lr: 0.000022 - momentum: 0.000000
2023-10-17 09:55:58,035 epoch 7 - iter 60/304 - loss 0.01823157 - time (sec): 2.88 - samples/sec: 2169.76 - lr: 0.000021 - momentum: 0.000000
2023-10-17 09:55:59,480 epoch 7 - iter 90/304 - loss 0.01503166 - time (sec): 4.33 - samples/sec: 2160.06 - lr: 0.000021 - momentum: 0.000000
2023-10-17 09:56:00,863 epoch 7 - iter 120/304 - loss 0.01826492 - time (sec): 5.71 - samples/sec: 2216.00 - lr: 0.000020 - momentum: 0.000000
2023-10-17 09:56:02,265 epoch 7 - iter 150/304 - loss 0.01729233 - time (sec): 7.11 - samples/sec: 2199.37 - lr: 0.000020 - momentum: 0.000000
2023-10-17 09:56:03,632 epoch 7 - iter 180/304 - loss 0.01866288 - time (sec): 8.48 - samples/sec: 2196.68 - lr: 0.000019 - momentum: 0.000000
2023-10-17 09:56:05,010 epoch 7 - iter 210/304 - loss 0.02159675 - time (sec): 9.86 - samples/sec: 2192.19 - lr: 0.000018 - momentum: 0.000000
2023-10-17 09:56:06,402 epoch 7 - iter 240/304 - loss 0.02325090 - time (sec): 11.25 - samples/sec: 2186.48 - lr: 0.000018 - momentum: 0.000000
2023-10-17 09:56:07,770 epoch 7 - iter 270/304 - loss 0.02334865 - time (sec): 12.62 - samples/sec: 2180.37 - lr: 0.000017 - momentum: 0.000000
2023-10-17 09:56:09,171 epoch 7 - iter 300/304 - loss 0.02307807 - time (sec): 14.02 - samples/sec: 2188.50 - lr: 0.000017 - momentum: 0.000000
2023-10-17 09:56:09,355 ----------------------------------------------------------------------------------------------------
2023-10-17 09:56:09,355 EPOCH 7 done: loss 0.0234 - lr: 0.000017
2023-10-17 09:56:10,274 DEV : loss 0.2072305530309677 - f1-score (micro avg) 0.84
2023-10-17 09:56:10,281 saving best model
2023-10-17 09:56:10,790 ----------------------------------------------------------------------------------------------------
2023-10-17 09:56:12,174 epoch 8 - iter 30/304 - loss 0.00679041 - time (sec): 1.38 - samples/sec: 2031.26 - lr: 0.000016 - momentum: 0.000000
2023-10-17 09:56:13,554 epoch 8 - iter 60/304 - loss 0.00994703 - time (sec): 2.76 - samples/sec: 2114.09 - lr: 0.000016 - momentum: 0.000000
2023-10-17 09:56:14,957 epoch 8 - iter 90/304 - loss 0.02010303 - time (sec): 4.17 - samples/sec: 2187.26 - lr: 0.000015 - momentum: 0.000000
2023-10-17 09:56:16,378 epoch 8 - iter 120/304 - loss 0.02061087 - time (sec): 5.59 - samples/sec: 2120.24 - lr: 0.000015 - momentum: 0.000000
2023-10-17 09:56:17,788 epoch 8 - iter 150/304 - loss 0.02246354 - time (sec): 7.00 - samples/sec: 2136.88 - lr: 0.000014 - momentum: 0.000000
2023-10-17 09:56:19,187 epoch 8 - iter 180/304 - loss 0.02369727 - time (sec): 8.40 - samples/sec: 2144.79 - lr: 0.000013 - momentum: 0.000000
2023-10-17 09:56:20,510 epoch 8 - iter 210/304 - loss 0.02258358 - time (sec): 9.72 - samples/sec: 2187.46 - lr: 0.000013 - momentum: 0.000000
2023-10-17 09:56:21,845 epoch 8 - iter 240/304 - loss 0.02108637 - time (sec): 11.05 - samples/sec: 2202.68 - lr: 0.000012 - momentum: 0.000000
2023-10-17 09:56:23,173 epoch 8 - iter 270/304 - loss 0.02132594 - time (sec): 12.38 - samples/sec: 2201.90 - lr: 0.000012 - momentum: 0.000000
2023-10-17 09:56:24,603 epoch 8 - iter 300/304 - loss 0.02121622 - time (sec): 13.81 - samples/sec: 2216.72 - lr: 0.000011 - momentum: 0.000000
2023-10-17 09:56:24,780 ----------------------------------------------------------------------------------------------------
2023-10-17 09:56:24,780 EPOCH 8 done: loss 0.0210 - lr: 0.000011
2023-10-17 09:56:25,723 DEV : loss 0.21292492747306824 - f1-score (micro avg) 0.848
2023-10-17 09:56:25,730 saving best model
2023-10-17 09:56:26,197 ----------------------------------------------------------------------------------------------------
2023-10-17 09:56:27,668 epoch 9 - iter 30/304 - loss 0.01587622 - time (sec): 1.47 - samples/sec: 2101.64 - lr: 0.000011 - momentum: 0.000000
2023-10-17 09:56:29,102 epoch 9 - iter 60/304 - loss 0.01070887 - time (sec): 2.90 - samples/sec: 2050.10 - lr: 0.000010 - momentum: 0.000000
2023-10-17 09:56:30,478 epoch 9 - iter 90/304 - loss 0.01496039 - time (sec): 4.28 - samples/sec: 2073.06 - lr: 0.000010 - momentum: 0.000000
2023-10-17 09:56:31,887 epoch 9 - iter 120/304 - loss 0.01124426 - time (sec): 5.69 - samples/sec: 2077.72 - lr: 0.000009 - momentum: 0.000000
2023-10-17 09:56:33,312 epoch 9 - iter 150/304 - loss 0.01410474 - time (sec): 7.11 - samples/sec: 2086.37 - lr: 0.000008 - momentum: 0.000000
2023-10-17 09:56:34,795 epoch 9 - iter 180/304 - loss 0.01244996 - time (sec): 8.60 - samples/sec: 2092.32 - lr: 0.000008 - momentum: 0.000000
2023-10-17 09:56:36,264 epoch 9 - iter 210/304 - loss 0.01248262 - time (sec): 10.06 - samples/sec: 2091.37 - lr: 0.000007 - momentum: 0.000000
2023-10-17 09:56:37,754 epoch 9 - iter 240/304 - loss 0.01213209 - time (sec): 11.55 - samples/sec: 2102.23 - lr: 0.000007 - momentum: 0.000000
2023-10-17 09:56:39,170 epoch 9 - iter 270/304 - loss 0.01353100 - time (sec): 12.97 - samples/sec: 2131.95 - lr: 0.000006 - momentum: 0.000000
2023-10-17 09:56:40,608 epoch 9 - iter 300/304 - loss 0.01378414 - time (sec): 14.41 - samples/sec: 2124.72 - lr: 0.000006 - momentum: 0.000000
2023-10-17 09:56:40,812 ----------------------------------------------------------------------------------------------------
2023-10-17 09:56:40,812 EPOCH 9 done: loss 0.0136 - lr: 0.000006
2023-10-17 09:56:41,731 DEV : loss 0.2044668346643448 - f1-score (micro avg) 0.8541
2023-10-17 09:56:41,738 saving best model
2023-10-17 09:56:42,227 ----------------------------------------------------------------------------------------------------
2023-10-17 09:56:43,715 epoch 10 - iter 30/304 - loss 0.01354305 - time (sec): 1.48 - samples/sec: 2072.22 - lr: 0.000005 - momentum: 0.000000
2023-10-17 09:56:45,124 epoch 10 - iter 60/304 - loss 0.00751199 - time (sec): 2.89 - samples/sec: 2071.21 - lr: 0.000005 - momentum: 0.000000
2023-10-17 09:56:46,546 epoch 10 - iter 90/304 - loss 0.00928881 - time (sec): 4.31 - samples/sec: 2161.61 - lr: 0.000004 - momentum: 0.000000
2023-10-17 09:56:47,844 epoch 10 - iter 120/304 - loss 0.01290944 - time (sec): 5.61 - samples/sec: 2182.17 - lr: 0.000003 - momentum: 0.000000
2023-10-17 09:56:49,142 epoch 10 - iter 150/304 - loss 0.01173024 - time (sec): 6.91 - samples/sec: 2205.14 - lr: 0.000003 - momentum: 0.000000
2023-10-17 09:56:50,445 epoch 10 - iter 180/304 - loss 0.01028826 - time (sec): 8.21 - samples/sec: 2220.18 - lr: 0.000002 - momentum: 0.000000
2023-10-17 09:56:51,743 epoch 10 - iter 210/304 - loss 0.00959073 - time (sec): 9.51 - samples/sec: 2213.19 - lr: 0.000002 - momentum: 0.000000
2023-10-17 09:56:53,047 epoch 10 - iter 240/304 - loss 0.00963868 - time (sec): 10.82 - samples/sec: 2205.85 - lr: 0.000001 - momentum: 0.000000
2023-10-17 09:56:54,380 epoch 10 - iter 270/304 - loss 0.00918610 - time (sec): 12.15 - samples/sec: 2248.93 - lr: 0.000001 - momentum: 0.000000
2023-10-17 09:56:55,708 epoch 10 - iter 300/304 - loss 0.01027470 - time (sec): 13.48 - samples/sec: 2269.04 - lr: 0.000000 - momentum: 0.000000
2023-10-17 09:56:55,887 ----------------------------------------------------------------------------------------------------
2023-10-17 09:56:55,887 EPOCH 10 done: loss 0.0101 - lr: 0.000000
2023-10-17 09:56:56,858 DEV : loss 0.21448352932929993 - f1-score (micro avg) 0.8595
2023-10-17 09:56:56,871 saving best model
2023-10-17 09:56:57,685 ----------------------------------------------------------------------------------------------------
2023-10-17 09:56:57,687 Loading model from best epoch ...
2023-10-17 09:56:59,606 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-date, B-date, E-date, I-date, S-object, B-object, E-object, I-object
2023-10-17 09:57:00,722
Results:
- F-score (micro) 0.8425
- F-score (macro) 0.7312
- Accuracy 0.7332
By class:
precision recall f1-score support
scope 0.7812 0.8278 0.8039 151
work 0.7944 0.8947 0.8416 95
pers 0.8932 0.9583 0.9246 96
date 0.2500 0.3333 0.2857 3
loc 1.0000 0.6667 0.8000 3
micro avg 0.8112 0.8764 0.8425 348
macro avg 0.7438 0.7362 0.7312 348
weighted avg 0.8130 0.8764 0.8430 348
2023-10-17 09:57:00,722 ----------------------------------------------------------------------------------------------------
|