File size: 23,919 Bytes
490db8d |
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 |
2023-10-17 10:54:49,837 ----------------------------------------------------------------------------------------------------
2023-10-17 10:54:49,838 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 10:54:49,838 ----------------------------------------------------------------------------------------------------
2023-10-17 10:54:49,838 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
2023-10-17 10:54:49,838 ----------------------------------------------------------------------------------------------------
2023-10-17 10:54:49,839 Train: 966 sentences
2023-10-17 10:54:49,839 (train_with_dev=False, train_with_test=False)
2023-10-17 10:54:49,839 ----------------------------------------------------------------------------------------------------
2023-10-17 10:54:49,839 Training Params:
2023-10-17 10:54:49,839 - learning_rate: "5e-05"
2023-10-17 10:54:49,839 - mini_batch_size: "4"
2023-10-17 10:54:49,839 - max_epochs: "10"
2023-10-17 10:54:49,839 - shuffle: "True"
2023-10-17 10:54:49,839 ----------------------------------------------------------------------------------------------------
2023-10-17 10:54:49,839 Plugins:
2023-10-17 10:54:49,839 - TensorboardLogger
2023-10-17 10:54:49,839 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 10:54:49,839 ----------------------------------------------------------------------------------------------------
2023-10-17 10:54:49,839 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 10:54:49,839 - metric: "('micro avg', 'f1-score')"
2023-10-17 10:54:49,839 ----------------------------------------------------------------------------------------------------
2023-10-17 10:54:49,839 Computation:
2023-10-17 10:54:49,839 - compute on device: cuda:0
2023-10-17 10:54:49,839 - embedding storage: none
2023-10-17 10:54:49,839 ----------------------------------------------------------------------------------------------------
2023-10-17 10:54:49,839 Model training base path: "hmbench-ajmc/fr-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-17 10:54:49,839 ----------------------------------------------------------------------------------------------------
2023-10-17 10:54:49,839 ----------------------------------------------------------------------------------------------------
2023-10-17 10:54:49,839 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 10:54:50,943 epoch 1 - iter 24/242 - loss 3.92639461 - time (sec): 1.10 - samples/sec: 2185.69 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:54:52,047 epoch 1 - iter 48/242 - loss 3.04296997 - time (sec): 2.21 - samples/sec: 2152.70 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:54:53,168 epoch 1 - iter 72/242 - loss 2.24676922 - time (sec): 3.33 - samples/sec: 2125.43 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:54:54,279 epoch 1 - iter 96/242 - loss 1.75436762 - time (sec): 4.44 - samples/sec: 2199.27 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:54:55,414 epoch 1 - iter 120/242 - loss 1.50738487 - time (sec): 5.57 - samples/sec: 2161.27 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:54:56,565 epoch 1 - iter 144/242 - loss 1.29770685 - time (sec): 6.73 - samples/sec: 2174.19 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:54:57,742 epoch 1 - iter 168/242 - loss 1.13283400 - time (sec): 7.90 - samples/sec: 2190.62 - lr: 0.000035 - momentum: 0.000000
2023-10-17 10:54:58,906 epoch 1 - iter 192/242 - loss 1.01056784 - time (sec): 9.07 - samples/sec: 2219.15 - lr: 0.000039 - momentum: 0.000000
2023-10-17 10:55:00,077 epoch 1 - iter 216/242 - loss 0.94261585 - time (sec): 10.24 - samples/sec: 2183.51 - lr: 0.000044 - momentum: 0.000000
2023-10-17 10:55:01,189 epoch 1 - iter 240/242 - loss 0.87690795 - time (sec): 11.35 - samples/sec: 2168.80 - lr: 0.000049 - momentum: 0.000000
2023-10-17 10:55:01,279 ----------------------------------------------------------------------------------------------------
2023-10-17 10:55:01,279 EPOCH 1 done: loss 0.8745 - lr: 0.000049
2023-10-17 10:55:01,866 DEV : loss 0.17245863378047943 - f1-score (micro avg) 0.6897
2023-10-17 10:55:01,871 saving best model
2023-10-17 10:55:02,224 ----------------------------------------------------------------------------------------------------
2023-10-17 10:55:03,328 epoch 2 - iter 24/242 - loss 0.12720849 - time (sec): 1.10 - samples/sec: 2082.24 - lr: 0.000049 - momentum: 0.000000
2023-10-17 10:55:04,500 epoch 2 - iter 48/242 - loss 0.16247962 - time (sec): 2.27 - samples/sec: 2053.44 - lr: 0.000049 - momentum: 0.000000
2023-10-17 10:55:05,684 epoch 2 - iter 72/242 - loss 0.16688621 - time (sec): 3.46 - samples/sec: 2054.06 - lr: 0.000048 - momentum: 0.000000
2023-10-17 10:55:06,852 epoch 2 - iter 96/242 - loss 0.15980410 - time (sec): 4.63 - samples/sec: 2056.90 - lr: 0.000048 - momentum: 0.000000
2023-10-17 10:55:08,050 epoch 2 - iter 120/242 - loss 0.16033847 - time (sec): 5.82 - samples/sec: 2101.10 - lr: 0.000047 - momentum: 0.000000
2023-10-17 10:55:09,238 epoch 2 - iter 144/242 - loss 0.16292839 - time (sec): 7.01 - samples/sec: 2073.75 - lr: 0.000047 - momentum: 0.000000
2023-10-17 10:55:10,489 epoch 2 - iter 168/242 - loss 0.16233553 - time (sec): 8.26 - samples/sec: 2073.82 - lr: 0.000046 - momentum: 0.000000
2023-10-17 10:55:11,704 epoch 2 - iter 192/242 - loss 0.15837280 - time (sec): 9.48 - samples/sec: 2076.17 - lr: 0.000046 - momentum: 0.000000
2023-10-17 10:55:12,877 epoch 2 - iter 216/242 - loss 0.16692967 - time (sec): 10.65 - samples/sec: 2077.61 - lr: 0.000045 - momentum: 0.000000
2023-10-17 10:55:14,005 epoch 2 - iter 240/242 - loss 0.16298133 - time (sec): 11.78 - samples/sec: 2085.51 - lr: 0.000045 - momentum: 0.000000
2023-10-17 10:55:14,099 ----------------------------------------------------------------------------------------------------
2023-10-17 10:55:14,099 EPOCH 2 done: loss 0.1622 - lr: 0.000045
2023-10-17 10:55:14,847 DEV : loss 0.14140677452087402 - f1-score (micro avg) 0.817
2023-10-17 10:55:14,852 saving best model
2023-10-17 10:55:15,405 ----------------------------------------------------------------------------------------------------
2023-10-17 10:55:16,610 epoch 3 - iter 24/242 - loss 0.10393024 - time (sec): 1.20 - samples/sec: 2010.78 - lr: 0.000044 - momentum: 0.000000
2023-10-17 10:55:17,774 epoch 3 - iter 48/242 - loss 0.09928119 - time (sec): 2.36 - samples/sec: 2100.50 - lr: 0.000043 - momentum: 0.000000
2023-10-17 10:55:18,907 epoch 3 - iter 72/242 - loss 0.08611132 - time (sec): 3.50 - samples/sec: 2201.33 - lr: 0.000043 - momentum: 0.000000
2023-10-17 10:55:19,983 epoch 3 - iter 96/242 - loss 0.09746320 - time (sec): 4.57 - samples/sec: 2216.12 - lr: 0.000042 - momentum: 0.000000
2023-10-17 10:55:21,108 epoch 3 - iter 120/242 - loss 0.10170700 - time (sec): 5.70 - samples/sec: 2202.86 - lr: 0.000042 - momentum: 0.000000
2023-10-17 10:55:22,214 epoch 3 - iter 144/242 - loss 0.10049154 - time (sec): 6.80 - samples/sec: 2207.65 - lr: 0.000041 - momentum: 0.000000
2023-10-17 10:55:23,359 epoch 3 - iter 168/242 - loss 0.10292233 - time (sec): 7.95 - samples/sec: 2214.43 - lr: 0.000041 - momentum: 0.000000
2023-10-17 10:55:24,450 epoch 3 - iter 192/242 - loss 0.10253347 - time (sec): 9.04 - samples/sec: 2164.84 - lr: 0.000040 - momentum: 0.000000
2023-10-17 10:55:25,681 epoch 3 - iter 216/242 - loss 0.10587963 - time (sec): 10.27 - samples/sec: 2172.46 - lr: 0.000040 - momentum: 0.000000
2023-10-17 10:55:26,765 epoch 3 - iter 240/242 - loss 0.10863240 - time (sec): 11.36 - samples/sec: 2162.00 - lr: 0.000039 - momentum: 0.000000
2023-10-17 10:55:26,860 ----------------------------------------------------------------------------------------------------
2023-10-17 10:55:26,860 EPOCH 3 done: loss 0.1094 - lr: 0.000039
2023-10-17 10:55:27,755 DEV : loss 0.14144273102283478 - f1-score (micro avg) 0.7744
2023-10-17 10:55:27,761 ----------------------------------------------------------------------------------------------------
2023-10-17 10:55:28,875 epoch 4 - iter 24/242 - loss 0.08604125 - time (sec): 1.11 - samples/sec: 1861.28 - lr: 0.000038 - momentum: 0.000000
2023-10-17 10:55:30,027 epoch 4 - iter 48/242 - loss 0.06395054 - time (sec): 2.27 - samples/sec: 2012.14 - lr: 0.000038 - momentum: 0.000000
2023-10-17 10:55:31,147 epoch 4 - iter 72/242 - loss 0.05707292 - time (sec): 3.38 - samples/sec: 2055.84 - lr: 0.000037 - momentum: 0.000000
2023-10-17 10:55:32,303 epoch 4 - iter 96/242 - loss 0.06229815 - time (sec): 4.54 - samples/sec: 2104.22 - lr: 0.000037 - momentum: 0.000000
2023-10-17 10:55:33,429 epoch 4 - iter 120/242 - loss 0.07151279 - time (sec): 5.67 - samples/sec: 2156.06 - lr: 0.000036 - momentum: 0.000000
2023-10-17 10:55:34,510 epoch 4 - iter 144/242 - loss 0.07093158 - time (sec): 6.75 - samples/sec: 2135.41 - lr: 0.000036 - momentum: 0.000000
2023-10-17 10:55:35,646 epoch 4 - iter 168/242 - loss 0.07679301 - time (sec): 7.88 - samples/sec: 2132.70 - lr: 0.000035 - momentum: 0.000000
2023-10-17 10:55:36,781 epoch 4 - iter 192/242 - loss 0.07394390 - time (sec): 9.02 - samples/sec: 2145.66 - lr: 0.000035 - momentum: 0.000000
2023-10-17 10:55:37,969 epoch 4 - iter 216/242 - loss 0.07397380 - time (sec): 10.21 - samples/sec: 2145.94 - lr: 0.000034 - momentum: 0.000000
2023-10-17 10:55:39,066 epoch 4 - iter 240/242 - loss 0.07206965 - time (sec): 11.30 - samples/sec: 2178.72 - lr: 0.000033 - momentum: 0.000000
2023-10-17 10:55:39,154 ----------------------------------------------------------------------------------------------------
2023-10-17 10:55:39,154 EPOCH 4 done: loss 0.0717 - lr: 0.000033
2023-10-17 10:55:39,908 DEV : loss 0.17283324897289276 - f1-score (micro avg) 0.836
2023-10-17 10:55:39,913 saving best model
2023-10-17 10:55:40,419 ----------------------------------------------------------------------------------------------------
2023-10-17 10:55:41,545 epoch 5 - iter 24/242 - loss 0.05122857 - time (sec): 1.12 - samples/sec: 1932.26 - lr: 0.000033 - momentum: 0.000000
2023-10-17 10:55:42,636 epoch 5 - iter 48/242 - loss 0.05981318 - time (sec): 2.22 - samples/sec: 2005.79 - lr: 0.000032 - momentum: 0.000000
2023-10-17 10:55:43,777 epoch 5 - iter 72/242 - loss 0.05077187 - time (sec): 3.36 - samples/sec: 2097.46 - lr: 0.000032 - momentum: 0.000000
2023-10-17 10:55:44,923 epoch 5 - iter 96/242 - loss 0.05457487 - time (sec): 4.50 - samples/sec: 2081.38 - lr: 0.000031 - momentum: 0.000000
2023-10-17 10:55:46,017 epoch 5 - iter 120/242 - loss 0.04995074 - time (sec): 5.60 - samples/sec: 2103.37 - lr: 0.000031 - momentum: 0.000000
2023-10-17 10:55:47,131 epoch 5 - iter 144/242 - loss 0.05286292 - time (sec): 6.71 - samples/sec: 2172.68 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:55:48,242 epoch 5 - iter 168/242 - loss 0.05572414 - time (sec): 7.82 - samples/sec: 2154.82 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:55:49,359 epoch 5 - iter 192/242 - loss 0.05767409 - time (sec): 8.94 - samples/sec: 2164.19 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:55:50,471 epoch 5 - iter 216/242 - loss 0.05614593 - time (sec): 10.05 - samples/sec: 2176.22 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:55:51,624 epoch 5 - iter 240/242 - loss 0.05545451 - time (sec): 11.20 - samples/sec: 2200.52 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:55:51,711 ----------------------------------------------------------------------------------------------------
2023-10-17 10:55:51,711 EPOCH 5 done: loss 0.0553 - lr: 0.000028
2023-10-17 10:55:52,477 DEV : loss 0.20903627574443817 - f1-score (micro avg) 0.8327
2023-10-17 10:55:52,482 ----------------------------------------------------------------------------------------------------
2023-10-17 10:55:53,546 epoch 6 - iter 24/242 - loss 0.01747608 - time (sec): 1.06 - samples/sec: 2304.10 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:55:54,613 epoch 6 - iter 48/242 - loss 0.03186200 - time (sec): 2.13 - samples/sec: 2322.04 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:55:55,673 epoch 6 - iter 72/242 - loss 0.02827303 - time (sec): 3.19 - samples/sec: 2354.56 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:55:56,712 epoch 6 - iter 96/242 - loss 0.03116679 - time (sec): 4.23 - samples/sec: 2299.45 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:55:57,776 epoch 6 - iter 120/242 - loss 0.03185366 - time (sec): 5.29 - samples/sec: 2307.22 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:55:58,823 epoch 6 - iter 144/242 - loss 0.03273987 - time (sec): 6.34 - samples/sec: 2250.37 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:55:59,922 epoch 6 - iter 168/242 - loss 0.03376056 - time (sec): 7.44 - samples/sec: 2293.07 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:56:01,000 epoch 6 - iter 192/242 - loss 0.03120273 - time (sec): 8.52 - samples/sec: 2294.16 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:56:02,062 epoch 6 - iter 216/242 - loss 0.03176433 - time (sec): 9.58 - samples/sec: 2289.35 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:56:03,138 epoch 6 - iter 240/242 - loss 0.03443780 - time (sec): 10.65 - samples/sec: 2305.51 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:56:03,222 ----------------------------------------------------------------------------------------------------
2023-10-17 10:56:03,222 EPOCH 6 done: loss 0.0342 - lr: 0.000022
2023-10-17 10:56:03,970 DEV : loss 0.20771022140979767 - f1-score (micro avg) 0.8579
2023-10-17 10:56:03,976 saving best model
2023-10-17 10:56:04,527 ----------------------------------------------------------------------------------------------------
2023-10-17 10:56:05,657 epoch 7 - iter 24/242 - loss 0.01389319 - time (sec): 1.13 - samples/sec: 2125.86 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:56:06,806 epoch 7 - iter 48/242 - loss 0.03079333 - time (sec): 2.27 - samples/sec: 2048.45 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:56:07,974 epoch 7 - iter 72/242 - loss 0.02531122 - time (sec): 3.44 - samples/sec: 2100.59 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:56:09,106 epoch 7 - iter 96/242 - loss 0.02382481 - time (sec): 4.57 - samples/sec: 2112.74 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:56:10,217 epoch 7 - iter 120/242 - loss 0.02394786 - time (sec): 5.69 - samples/sec: 2156.15 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:56:11,370 epoch 7 - iter 144/242 - loss 0.02896998 - time (sec): 6.84 - samples/sec: 2131.58 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:56:12,497 epoch 7 - iter 168/242 - loss 0.02898444 - time (sec): 7.97 - samples/sec: 2124.17 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:56:13,633 epoch 7 - iter 192/242 - loss 0.02772235 - time (sec): 9.10 - samples/sec: 2136.23 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:56:14,779 epoch 7 - iter 216/242 - loss 0.03031956 - time (sec): 10.25 - samples/sec: 2151.31 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:56:15,892 epoch 7 - iter 240/242 - loss 0.02928220 - time (sec): 11.36 - samples/sec: 2168.91 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:56:15,984 ----------------------------------------------------------------------------------------------------
2023-10-17 10:56:15,984 EPOCH 7 done: loss 0.0292 - lr: 0.000017
2023-10-17 10:56:16,738 DEV : loss 0.2137136161327362 - f1-score (micro avg) 0.851
2023-10-17 10:56:16,743 ----------------------------------------------------------------------------------------------------
2023-10-17 10:56:17,874 epoch 8 - iter 24/242 - loss 0.03087610 - time (sec): 1.13 - samples/sec: 1895.79 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:56:19,021 epoch 8 - iter 48/242 - loss 0.02349420 - time (sec): 2.28 - samples/sec: 2107.16 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:56:20,219 epoch 8 - iter 72/242 - loss 0.01962876 - time (sec): 3.47 - samples/sec: 2060.49 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:56:21,391 epoch 8 - iter 96/242 - loss 0.01966456 - time (sec): 4.65 - samples/sec: 2156.26 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:56:22,513 epoch 8 - iter 120/242 - loss 0.01867760 - time (sec): 5.77 - samples/sec: 2171.79 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:56:23,627 epoch 8 - iter 144/242 - loss 0.01698197 - time (sec): 6.88 - samples/sec: 2177.10 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:56:24,717 epoch 8 - iter 168/242 - loss 0.01799341 - time (sec): 7.97 - samples/sec: 2153.06 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:56:25,815 epoch 8 - iter 192/242 - loss 0.01797490 - time (sec): 9.07 - samples/sec: 2164.11 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:56:26,937 epoch 8 - iter 216/242 - loss 0.01687743 - time (sec): 10.19 - samples/sec: 2183.19 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:56:28,061 epoch 8 - iter 240/242 - loss 0.01884973 - time (sec): 11.32 - samples/sec: 2174.54 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:56:28,151 ----------------------------------------------------------------------------------------------------
2023-10-17 10:56:28,151 EPOCH 8 done: loss 0.0188 - lr: 0.000011
2023-10-17 10:56:28,897 DEV : loss 0.23500145971775055 - f1-score (micro avg) 0.8308
2023-10-17 10:56:28,902 ----------------------------------------------------------------------------------------------------
2023-10-17 10:56:30,021 epoch 9 - iter 24/242 - loss 0.01912375 - time (sec): 1.12 - samples/sec: 2236.59 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:56:31,185 epoch 9 - iter 48/242 - loss 0.01160460 - time (sec): 2.28 - samples/sec: 2103.10 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:56:32,436 epoch 9 - iter 72/242 - loss 0.01393317 - time (sec): 3.53 - samples/sec: 1984.39 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:56:33,660 epoch 9 - iter 96/242 - loss 0.01317219 - time (sec): 4.76 - samples/sec: 1958.55 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:56:34,862 epoch 9 - iter 120/242 - loss 0.01130146 - time (sec): 5.96 - samples/sec: 1941.54 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:56:36,027 epoch 9 - iter 144/242 - loss 0.01143745 - time (sec): 7.12 - samples/sec: 1998.38 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:56:37,136 epoch 9 - iter 168/242 - loss 0.01338574 - time (sec): 8.23 - samples/sec: 2043.43 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:56:38,306 epoch 9 - iter 192/242 - loss 0.01238712 - time (sec): 9.40 - samples/sec: 2074.54 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:56:39,479 epoch 9 - iter 216/242 - loss 0.01393675 - time (sec): 10.58 - samples/sec: 2099.41 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:56:40,612 epoch 9 - iter 240/242 - loss 0.01450758 - time (sec): 11.71 - samples/sec: 2100.88 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:56:40,709 ----------------------------------------------------------------------------------------------------
2023-10-17 10:56:40,709 EPOCH 9 done: loss 0.0144 - lr: 0.000006
2023-10-17 10:56:41,455 DEV : loss 0.2329409271478653 - f1-score (micro avg) 0.8317
2023-10-17 10:56:41,460 ----------------------------------------------------------------------------------------------------
2023-10-17 10:56:42,621 epoch 10 - iter 24/242 - loss 0.00912702 - time (sec): 1.16 - samples/sec: 2040.82 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:56:43,716 epoch 10 - iter 48/242 - loss 0.01040602 - time (sec): 2.25 - samples/sec: 2024.67 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:56:44,872 epoch 10 - iter 72/242 - loss 0.01081807 - time (sec): 3.41 - samples/sec: 2099.24 - lr: 0.000004 - momentum: 0.000000
2023-10-17 10:56:45,995 epoch 10 - iter 96/242 - loss 0.01109546 - time (sec): 4.53 - samples/sec: 2137.61 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:56:47,133 epoch 10 - iter 120/242 - loss 0.01296140 - time (sec): 5.67 - samples/sec: 2241.61 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:56:48,242 epoch 10 - iter 144/242 - loss 0.01117121 - time (sec): 6.78 - samples/sec: 2192.51 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:56:49,346 epoch 10 - iter 168/242 - loss 0.00974678 - time (sec): 7.88 - samples/sec: 2164.81 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:56:50,489 epoch 10 - iter 192/242 - loss 0.00970789 - time (sec): 9.03 - samples/sec: 2182.58 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:56:51,625 epoch 10 - iter 216/242 - loss 0.00940542 - time (sec): 10.16 - samples/sec: 2206.48 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:56:52,740 epoch 10 - iter 240/242 - loss 0.00865941 - time (sec): 11.28 - samples/sec: 2178.99 - lr: 0.000000 - momentum: 0.000000
2023-10-17 10:56:52,823 ----------------------------------------------------------------------------------------------------
2023-10-17 10:56:52,824 EPOCH 10 done: loss 0.0086 - lr: 0.000000
2023-10-17 10:56:53,569 DEV : loss 0.24286319315433502 - f1-score (micro avg) 0.8363
2023-10-17 10:56:53,947 ----------------------------------------------------------------------------------------------------
2023-10-17 10:56:53,949 Loading model from best epoch ...
2023-10-17 10:56:55,284 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
2023-10-17 10:56:56,147
Results:
- F-score (micro) 0.8379
- F-score (macro) 0.575
- Accuracy 0.7332
By class:
precision recall f1-score support
pers 0.8750 0.9065 0.8905 139
scope 0.8321 0.8837 0.8571 129
work 0.7412 0.7875 0.7636 80
loc 1.0000 0.2222 0.3636 9
date 0.0000 0.0000 0.0000 3
micro avg 0.8288 0.8472 0.8379 360
macro avg 0.6897 0.5600 0.5750 360
weighted avg 0.8257 0.8472 0.8297 360
2023-10-17 10:56:56,147 ----------------------------------------------------------------------------------------------------
|