File size: 24,083 Bytes
76b0cab |
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 |
2023-10-17 10:05:15,184 ----------------------------------------------------------------------------------------------------
2023-10-17 10:05:15,185 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 10:05:15,185 ----------------------------------------------------------------------------------------------------
2023-10-17 10:05:15,185 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 10:05:15,185 ----------------------------------------------------------------------------------------------------
2023-10-17 10:05:15,185 Train: 1214 sentences
2023-10-17 10:05:15,185 (train_with_dev=False, train_with_test=False)
2023-10-17 10:05:15,185 ----------------------------------------------------------------------------------------------------
2023-10-17 10:05:15,185 Training Params:
2023-10-17 10:05:15,185 - learning_rate: "5e-05"
2023-10-17 10:05:15,185 - mini_batch_size: "4"
2023-10-17 10:05:15,185 - max_epochs: "10"
2023-10-17 10:05:15,186 - shuffle: "True"
2023-10-17 10:05:15,186 ----------------------------------------------------------------------------------------------------
2023-10-17 10:05:15,186 Plugins:
2023-10-17 10:05:15,186 - TensorboardLogger
2023-10-17 10:05:15,186 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 10:05:15,186 ----------------------------------------------------------------------------------------------------
2023-10-17 10:05:15,186 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 10:05:15,186 - metric: "('micro avg', 'f1-score')"
2023-10-17 10:05:15,186 ----------------------------------------------------------------------------------------------------
2023-10-17 10:05:15,186 Computation:
2023-10-17 10:05:15,186 - compute on device: cuda:0
2023-10-17 10:05:15,186 - embedding storage: none
2023-10-17 10:05:15,186 ----------------------------------------------------------------------------------------------------
2023-10-17 10:05:15,186 Model training base path: "hmbench-ajmc/en-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-17 10:05:15,186 ----------------------------------------------------------------------------------------------------
2023-10-17 10:05:15,186 ----------------------------------------------------------------------------------------------------
2023-10-17 10:05:15,186 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 10:05:16,663 epoch 1 - iter 30/304 - loss 3.78710679 - time (sec): 1.48 - samples/sec: 1925.88 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:05:18,116 epoch 1 - iter 60/304 - loss 2.73070745 - time (sec): 2.93 - samples/sec: 2052.26 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:05:19,510 epoch 1 - iter 90/304 - loss 2.06124605 - time (sec): 4.32 - samples/sec: 2082.46 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:05:20,933 epoch 1 - iter 120/304 - loss 1.63016199 - time (sec): 5.75 - samples/sec: 2108.39 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:05:22,307 epoch 1 - iter 150/304 - loss 1.38162276 - time (sec): 7.12 - samples/sec: 2111.78 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:05:23,699 epoch 1 - iter 180/304 - loss 1.20050216 - time (sec): 8.51 - samples/sec: 2136.27 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:05:25,113 epoch 1 - iter 210/304 - loss 1.05612794 - time (sec): 9.93 - samples/sec: 2177.89 - lr: 0.000034 - momentum: 0.000000
2023-10-17 10:05:26,524 epoch 1 - iter 240/304 - loss 0.95025081 - time (sec): 11.34 - samples/sec: 2179.86 - lr: 0.000039 - momentum: 0.000000
2023-10-17 10:05:27,891 epoch 1 - iter 270/304 - loss 0.86773041 - time (sec): 12.70 - samples/sec: 2194.21 - lr: 0.000044 - momentum: 0.000000
2023-10-17 10:05:29,262 epoch 1 - iter 300/304 - loss 0.80247575 - time (sec): 14.08 - samples/sec: 2180.36 - lr: 0.000049 - momentum: 0.000000
2023-10-17 10:05:29,450 ----------------------------------------------------------------------------------------------------
2023-10-17 10:05:29,451 EPOCH 1 done: loss 0.7956 - lr: 0.000049
2023-10-17 10:05:30,543 DEV : loss 0.18940171599388123 - f1-score (micro avg) 0.6293
2023-10-17 10:05:30,552 saving best model
2023-10-17 10:05:30,950 ----------------------------------------------------------------------------------------------------
2023-10-17 10:05:32,505 epoch 2 - iter 30/304 - loss 0.18278762 - time (sec): 1.55 - samples/sec: 2122.52 - lr: 0.000049 - momentum: 0.000000
2023-10-17 10:05:34,094 epoch 2 - iter 60/304 - loss 0.18645643 - time (sec): 3.14 - samples/sec: 1998.47 - lr: 0.000049 - momentum: 0.000000
2023-10-17 10:05:35,698 epoch 2 - iter 90/304 - loss 0.15996554 - time (sec): 4.75 - samples/sec: 1964.82 - lr: 0.000048 - momentum: 0.000000
2023-10-17 10:05:37,285 epoch 2 - iter 120/304 - loss 0.15490686 - time (sec): 6.33 - samples/sec: 1966.73 - lr: 0.000048 - momentum: 0.000000
2023-10-17 10:05:38,850 epoch 2 - iter 150/304 - loss 0.15786215 - time (sec): 7.90 - samples/sec: 1969.34 - lr: 0.000047 - momentum: 0.000000
2023-10-17 10:05:40,407 epoch 2 - iter 180/304 - loss 0.15644962 - time (sec): 9.46 - samples/sec: 1992.69 - lr: 0.000047 - momentum: 0.000000
2023-10-17 10:05:41,954 epoch 2 - iter 210/304 - loss 0.14967596 - time (sec): 11.00 - samples/sec: 1955.64 - lr: 0.000046 - momentum: 0.000000
2023-10-17 10:05:43,498 epoch 2 - iter 240/304 - loss 0.14580967 - time (sec): 12.55 - samples/sec: 1982.75 - lr: 0.000046 - momentum: 0.000000
2023-10-17 10:05:45,087 epoch 2 - iter 270/304 - loss 0.14222833 - time (sec): 14.14 - samples/sec: 1982.37 - lr: 0.000045 - momentum: 0.000000
2023-10-17 10:05:46,670 epoch 2 - iter 300/304 - loss 0.14333843 - time (sec): 15.72 - samples/sec: 1951.35 - lr: 0.000045 - momentum: 0.000000
2023-10-17 10:05:46,880 ----------------------------------------------------------------------------------------------------
2023-10-17 10:05:46,880 EPOCH 2 done: loss 0.1420 - lr: 0.000045
2023-10-17 10:05:47,853 DEV : loss 0.16976089775562286 - f1-score (micro avg) 0.7528
2023-10-17 10:05:47,861 saving best model
2023-10-17 10:05:48,445 ----------------------------------------------------------------------------------------------------
2023-10-17 10:05:50,000 epoch 3 - iter 30/304 - loss 0.09322033 - time (sec): 1.55 - samples/sec: 1975.53 - lr: 0.000044 - momentum: 0.000000
2023-10-17 10:05:51,580 epoch 3 - iter 60/304 - loss 0.09273253 - time (sec): 3.13 - samples/sec: 1950.78 - lr: 0.000043 - momentum: 0.000000
2023-10-17 10:05:53,014 epoch 3 - iter 90/304 - loss 0.09363777 - time (sec): 4.57 - samples/sec: 2034.82 - lr: 0.000043 - momentum: 0.000000
2023-10-17 10:05:54,412 epoch 3 - iter 120/304 - loss 0.09298961 - time (sec): 5.97 - samples/sec: 2059.56 - lr: 0.000042 - momentum: 0.000000
2023-10-17 10:05:55,819 epoch 3 - iter 150/304 - loss 0.08922597 - time (sec): 7.37 - samples/sec: 2050.07 - lr: 0.000042 - momentum: 0.000000
2023-10-17 10:05:57,231 epoch 3 - iter 180/304 - loss 0.08730806 - time (sec): 8.78 - samples/sec: 2042.21 - lr: 0.000041 - momentum: 0.000000
2023-10-17 10:05:58,620 epoch 3 - iter 210/304 - loss 0.09127310 - time (sec): 10.17 - samples/sec: 2060.70 - lr: 0.000041 - momentum: 0.000000
2023-10-17 10:05:59,997 epoch 3 - iter 240/304 - loss 0.09364956 - time (sec): 11.55 - samples/sec: 2105.83 - lr: 0.000040 - momentum: 0.000000
2023-10-17 10:06:01,421 epoch 3 - iter 270/304 - loss 0.09589965 - time (sec): 12.97 - samples/sec: 2123.72 - lr: 0.000040 - momentum: 0.000000
2023-10-17 10:06:02,803 epoch 3 - iter 300/304 - loss 0.09784084 - time (sec): 14.36 - samples/sec: 2129.08 - lr: 0.000039 - momentum: 0.000000
2023-10-17 10:06:02,981 ----------------------------------------------------------------------------------------------------
2023-10-17 10:06:02,981 EPOCH 3 done: loss 0.0969 - lr: 0.000039
2023-10-17 10:06:03,949 DEV : loss 0.17592370510101318 - f1-score (micro avg) 0.8123
2023-10-17 10:06:03,957 saving best model
2023-10-17 10:06:04,402 ----------------------------------------------------------------------------------------------------
2023-10-17 10:06:05,997 epoch 4 - iter 30/304 - loss 0.05487273 - time (sec): 1.59 - samples/sec: 2222.32 - lr: 0.000038 - momentum: 0.000000
2023-10-17 10:06:07,579 epoch 4 - iter 60/304 - loss 0.06998359 - time (sec): 3.17 - samples/sec: 2065.77 - lr: 0.000038 - momentum: 0.000000
2023-10-17 10:06:09,156 epoch 4 - iter 90/304 - loss 0.06857176 - time (sec): 4.75 - samples/sec: 2008.65 - lr: 0.000037 - momentum: 0.000000
2023-10-17 10:06:10,820 epoch 4 - iter 120/304 - loss 0.06728517 - time (sec): 6.42 - samples/sec: 1935.14 - lr: 0.000037 - momentum: 0.000000
2023-10-17 10:06:12,399 epoch 4 - iter 150/304 - loss 0.06480906 - time (sec): 7.99 - samples/sec: 1943.06 - lr: 0.000036 - momentum: 0.000000
2023-10-17 10:06:14,047 epoch 4 - iter 180/304 - loss 0.07017801 - time (sec): 9.64 - samples/sec: 1941.32 - lr: 0.000036 - momentum: 0.000000
2023-10-17 10:06:15,633 epoch 4 - iter 210/304 - loss 0.06745500 - time (sec): 11.23 - samples/sec: 1939.28 - lr: 0.000035 - momentum: 0.000000
2023-10-17 10:06:17,187 epoch 4 - iter 240/304 - loss 0.06506326 - time (sec): 12.78 - samples/sec: 1940.37 - lr: 0.000035 - momentum: 0.000000
2023-10-17 10:06:18,833 epoch 4 - iter 270/304 - loss 0.06434291 - time (sec): 14.43 - samples/sec: 1924.82 - lr: 0.000034 - momentum: 0.000000
2023-10-17 10:06:20,490 epoch 4 - iter 300/304 - loss 0.06777296 - time (sec): 16.09 - samples/sec: 1899.81 - lr: 0.000033 - momentum: 0.000000
2023-10-17 10:06:20,700 ----------------------------------------------------------------------------------------------------
2023-10-17 10:06:20,700 EPOCH 4 done: loss 0.0689 - lr: 0.000033
2023-10-17 10:06:21,695 DEV : loss 0.20245088636875153 - f1-score (micro avg) 0.8253
2023-10-17 10:06:21,702 saving best model
2023-10-17 10:06:22,162 ----------------------------------------------------------------------------------------------------
2023-10-17 10:06:23,745 epoch 5 - iter 30/304 - loss 0.04992004 - time (sec): 1.58 - samples/sec: 2010.11 - lr: 0.000033 - momentum: 0.000000
2023-10-17 10:06:25,299 epoch 5 - iter 60/304 - loss 0.05014883 - time (sec): 3.14 - samples/sec: 1944.68 - lr: 0.000032 - momentum: 0.000000
2023-10-17 10:06:26,951 epoch 5 - iter 90/304 - loss 0.05743268 - time (sec): 4.79 - samples/sec: 1976.34 - lr: 0.000032 - momentum: 0.000000
2023-10-17 10:06:28,553 epoch 5 - iter 120/304 - loss 0.05409120 - time (sec): 6.39 - samples/sec: 1965.05 - lr: 0.000031 - momentum: 0.000000
2023-10-17 10:06:30,078 epoch 5 - iter 150/304 - loss 0.05061556 - time (sec): 7.91 - samples/sec: 1980.26 - lr: 0.000031 - momentum: 0.000000
2023-10-17 10:06:31,630 epoch 5 - iter 180/304 - loss 0.05270231 - time (sec): 9.47 - samples/sec: 1993.33 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:06:33,168 epoch 5 - iter 210/304 - loss 0.05109864 - time (sec): 11.00 - samples/sec: 1972.62 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:06:34,744 epoch 5 - iter 240/304 - loss 0.05359537 - time (sec): 12.58 - samples/sec: 1985.50 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:06:36,266 epoch 5 - iter 270/304 - loss 0.05236408 - time (sec): 14.10 - samples/sec: 1973.88 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:06:37,793 epoch 5 - iter 300/304 - loss 0.04758594 - time (sec): 15.63 - samples/sec: 1964.99 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:06:37,991 ----------------------------------------------------------------------------------------------------
2023-10-17 10:06:37,991 EPOCH 5 done: loss 0.0475 - lr: 0.000028
2023-10-17 10:06:38,998 DEV : loss 0.19856637716293335 - f1-score (micro avg) 0.8415
2023-10-17 10:06:39,007 saving best model
2023-10-17 10:06:39,447 ----------------------------------------------------------------------------------------------------
2023-10-17 10:06:40,994 epoch 6 - iter 30/304 - loss 0.03401226 - time (sec): 1.54 - samples/sec: 1974.87 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:06:42,386 epoch 6 - iter 60/304 - loss 0.02475800 - time (sec): 2.94 - samples/sec: 2120.75 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:06:43,768 epoch 6 - iter 90/304 - loss 0.02273325 - time (sec): 4.32 - samples/sec: 2083.45 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:06:45,186 epoch 6 - iter 120/304 - loss 0.02585818 - time (sec): 5.74 - samples/sec: 2156.94 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:06:46,722 epoch 6 - iter 150/304 - loss 0.02334142 - time (sec): 7.27 - samples/sec: 2090.79 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:06:48,285 epoch 6 - iter 180/304 - loss 0.03260567 - time (sec): 8.84 - samples/sec: 2085.81 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:06:49,839 epoch 6 - iter 210/304 - loss 0.03249194 - time (sec): 10.39 - samples/sec: 2031.52 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:06:51,420 epoch 6 - iter 240/304 - loss 0.03342660 - time (sec): 11.97 - samples/sec: 2058.44 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:06:52,951 epoch 6 - iter 270/304 - loss 0.03313726 - time (sec): 13.50 - samples/sec: 2057.54 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:06:54,477 epoch 6 - iter 300/304 - loss 0.03388454 - time (sec): 15.03 - samples/sec: 2047.16 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:06:54,690 ----------------------------------------------------------------------------------------------------
2023-10-17 10:06:54,690 EPOCH 6 done: loss 0.0336 - lr: 0.000022
2023-10-17 10:06:55,645 DEV : loss 0.22854739427566528 - f1-score (micro avg) 0.8323
2023-10-17 10:06:55,653 ----------------------------------------------------------------------------------------------------
2023-10-17 10:06:57,180 epoch 7 - iter 30/304 - loss 0.02727526 - time (sec): 1.53 - samples/sec: 1960.61 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:06:58,727 epoch 7 - iter 60/304 - loss 0.01954119 - time (sec): 3.07 - samples/sec: 2039.41 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:07:00,245 epoch 7 - iter 90/304 - loss 0.01949465 - time (sec): 4.59 - samples/sec: 1945.84 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:07:01,757 epoch 7 - iter 120/304 - loss 0.02626849 - time (sec): 6.10 - samples/sec: 1989.20 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:07:03,297 epoch 7 - iter 150/304 - loss 0.03160573 - time (sec): 7.64 - samples/sec: 2000.29 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:07:04,812 epoch 7 - iter 180/304 - loss 0.03165944 - time (sec): 9.16 - samples/sec: 2008.54 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:07:06,357 epoch 7 - iter 210/304 - loss 0.02809873 - time (sec): 10.70 - samples/sec: 2013.64 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:07:07,907 epoch 7 - iter 240/304 - loss 0.02841292 - time (sec): 12.25 - samples/sec: 2014.18 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:07:09,413 epoch 7 - iter 270/304 - loss 0.02688111 - time (sec): 13.76 - samples/sec: 2005.89 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:07:10,930 epoch 7 - iter 300/304 - loss 0.02676099 - time (sec): 15.28 - samples/sec: 2008.25 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:07:11,141 ----------------------------------------------------------------------------------------------------
2023-10-17 10:07:11,141 EPOCH 7 done: loss 0.0265 - lr: 0.000017
2023-10-17 10:07:12,131 DEV : loss 0.22182397544384003 - f1-score (micro avg) 0.8469
2023-10-17 10:07:12,139 saving best model
2023-10-17 10:07:12,610 ----------------------------------------------------------------------------------------------------
2023-10-17 10:07:14,097 epoch 8 - iter 30/304 - loss 0.03427352 - time (sec): 1.49 - samples/sec: 1921.31 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:07:15,659 epoch 8 - iter 60/304 - loss 0.02059897 - time (sec): 3.05 - samples/sec: 1907.52 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:07:17,222 epoch 8 - iter 90/304 - loss 0.01370743 - time (sec): 4.61 - samples/sec: 1942.25 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:07:18,786 epoch 8 - iter 120/304 - loss 0.01479090 - time (sec): 6.17 - samples/sec: 1930.02 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:07:20,347 epoch 8 - iter 150/304 - loss 0.01431101 - time (sec): 7.73 - samples/sec: 1955.61 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:07:21,881 epoch 8 - iter 180/304 - loss 0.01225557 - time (sec): 9.27 - samples/sec: 1962.45 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:07:23,413 epoch 8 - iter 210/304 - loss 0.01286757 - time (sec): 10.80 - samples/sec: 1966.89 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:07:24,911 epoch 8 - iter 240/304 - loss 0.01251269 - time (sec): 12.30 - samples/sec: 1994.09 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:07:26,433 epoch 8 - iter 270/304 - loss 0.01859633 - time (sec): 13.82 - samples/sec: 1996.49 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:07:27,952 epoch 8 - iter 300/304 - loss 0.02160579 - time (sec): 15.34 - samples/sec: 1994.18 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:07:28,145 ----------------------------------------------------------------------------------------------------
2023-10-17 10:07:28,145 EPOCH 8 done: loss 0.0213 - lr: 0.000011
2023-10-17 10:07:29,140 DEV : loss 0.20801495015621185 - f1-score (micro avg) 0.8656
2023-10-17 10:07:29,147 saving best model
2023-10-17 10:07:29,614 ----------------------------------------------------------------------------------------------------
2023-10-17 10:07:31,111 epoch 9 - iter 30/304 - loss 0.01053882 - time (sec): 1.50 - samples/sec: 2019.53 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:07:32,684 epoch 9 - iter 60/304 - loss 0.01047128 - time (sec): 3.07 - samples/sec: 2017.31 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:07:34,224 epoch 9 - iter 90/304 - loss 0.00969806 - time (sec): 4.61 - samples/sec: 2031.08 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:07:35,758 epoch 9 - iter 120/304 - loss 0.01549194 - time (sec): 6.14 - samples/sec: 2010.30 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:07:37,316 epoch 9 - iter 150/304 - loss 0.01484816 - time (sec): 7.70 - samples/sec: 1969.53 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:07:38,887 epoch 9 - iter 180/304 - loss 0.01706246 - time (sec): 9.27 - samples/sec: 1980.43 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:07:40,550 epoch 9 - iter 210/304 - loss 0.01557972 - time (sec): 10.93 - samples/sec: 1966.99 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:07:42,090 epoch 9 - iter 240/304 - loss 0.01455481 - time (sec): 12.47 - samples/sec: 1960.58 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:07:43,704 epoch 9 - iter 270/304 - loss 0.01487115 - time (sec): 14.09 - samples/sec: 1954.38 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:07:45,290 epoch 9 - iter 300/304 - loss 0.01407558 - time (sec): 15.67 - samples/sec: 1953.43 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:07:45,480 ----------------------------------------------------------------------------------------------------
2023-10-17 10:07:45,480 EPOCH 9 done: loss 0.0139 - lr: 0.000006
2023-10-17 10:07:46,449 DEV : loss 0.2179786115884781 - f1-score (micro avg) 0.8626
2023-10-17 10:07:46,456 ----------------------------------------------------------------------------------------------------
2023-10-17 10:07:47,970 epoch 10 - iter 30/304 - loss 0.00340291 - time (sec): 1.51 - samples/sec: 2129.64 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:07:49,551 epoch 10 - iter 60/304 - loss 0.01231855 - time (sec): 3.09 - samples/sec: 1994.84 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:07:51,112 epoch 10 - iter 90/304 - loss 0.00844573 - time (sec): 4.65 - samples/sec: 1944.47 - lr: 0.000004 - momentum: 0.000000
2023-10-17 10:07:52,670 epoch 10 - iter 120/304 - loss 0.00791771 - time (sec): 6.21 - samples/sec: 1996.50 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:07:54,261 epoch 10 - iter 150/304 - loss 0.00628740 - time (sec): 7.80 - samples/sec: 2006.65 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:07:55,860 epoch 10 - iter 180/304 - loss 0.00646574 - time (sec): 9.40 - samples/sec: 1989.37 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:07:57,418 epoch 10 - iter 210/304 - loss 0.00841620 - time (sec): 10.96 - samples/sec: 1968.42 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:07:58,983 epoch 10 - iter 240/304 - loss 0.01040953 - time (sec): 12.53 - samples/sec: 1958.76 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:08:00,532 epoch 10 - iter 270/304 - loss 0.01133920 - time (sec): 14.07 - samples/sec: 1956.54 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:08:02,099 epoch 10 - iter 300/304 - loss 0.01129145 - time (sec): 15.64 - samples/sec: 1953.92 - lr: 0.000000 - momentum: 0.000000
2023-10-17 10:08:02,307 ----------------------------------------------------------------------------------------------------
2023-10-17 10:08:02,307 EPOCH 10 done: loss 0.0113 - lr: 0.000000
2023-10-17 10:08:03,294 DEV : loss 0.21762032806873322 - f1-score (micro avg) 0.8616
2023-10-17 10:08:03,629 ----------------------------------------------------------------------------------------------------
2023-10-17 10:08:03,631 Loading model from best epoch ...
2023-10-17 10:08:05,420 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 10:08:06,625
Results:
- F-score (micro) 0.8242
- F-score (macro) 0.6634
- Accuracy 0.7059
By class:
precision recall f1-score support
scope 0.7771 0.8079 0.7922 151
pers 0.8142 0.9583 0.8804 96
work 0.8077 0.8842 0.8442 95
date 0.0000 0.0000 0.0000 3
loc 1.0000 0.6667 0.8000 3
micro avg 0.7895 0.8621 0.8242 348
macro avg 0.6798 0.6634 0.6634 348
weighted avg 0.7909 0.8621 0.8240 348
2023-10-17 10:08:06,625 ----------------------------------------------------------------------------------------------------
|