File size: 24,116 Bytes
6c81002 |
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 |
2023-10-17 10:02:06,493 ----------------------------------------------------------------------------------------------------
2023-10-17 10:02:06,495 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 10:02:06,495 ----------------------------------------------------------------------------------------------------
2023-10-17 10:02:06,495 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 10:02:06,495 ----------------------------------------------------------------------------------------------------
2023-10-17 10:02:06,495 Train: 1214 sentences
2023-10-17 10:02:06,495 (train_with_dev=False, train_with_test=False)
2023-10-17 10:02:06,495 ----------------------------------------------------------------------------------------------------
2023-10-17 10:02:06,495 Training Params:
2023-10-17 10:02:06,495 - learning_rate: "3e-05"
2023-10-17 10:02:06,495 - mini_batch_size: "4"
2023-10-17 10:02:06,495 - max_epochs: "10"
2023-10-17 10:02:06,495 - shuffle: "True"
2023-10-17 10:02:06,495 ----------------------------------------------------------------------------------------------------
2023-10-17 10:02:06,495 Plugins:
2023-10-17 10:02:06,496 - TensorboardLogger
2023-10-17 10:02:06,496 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 10:02:06,496 ----------------------------------------------------------------------------------------------------
2023-10-17 10:02:06,496 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 10:02:06,496 - metric: "('micro avg', 'f1-score')"
2023-10-17 10:02:06,496 ----------------------------------------------------------------------------------------------------
2023-10-17 10:02:06,496 Computation:
2023-10-17 10:02:06,496 - compute on device: cuda:0
2023-10-17 10:02:06,496 - embedding storage: none
2023-10-17 10:02:06,496 ----------------------------------------------------------------------------------------------------
2023-10-17 10:02:06,496 Model training base path: "hmbench-ajmc/en-hmteams/teams-base-historic-multilingual-discriminator-bs4-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-17 10:02:06,496 ----------------------------------------------------------------------------------------------------
2023-10-17 10:02:06,496 ----------------------------------------------------------------------------------------------------
2023-10-17 10:02:06,496 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 10:02:07,999 epoch 1 - iter 30/304 - loss 3.91471408 - time (sec): 1.50 - samples/sec: 1893.38 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:02:09,519 epoch 1 - iter 60/304 - loss 3.15356945 - time (sec): 3.02 - samples/sec: 1988.92 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:02:11,060 epoch 1 - iter 90/304 - loss 2.42737450 - time (sec): 4.56 - samples/sec: 1973.07 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:02:12,632 epoch 1 - iter 120/304 - loss 1.92457254 - time (sec): 6.13 - samples/sec: 1974.94 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:02:14,193 epoch 1 - iter 150/304 - loss 1.63369367 - time (sec): 7.70 - samples/sec: 1953.71 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:02:15,749 epoch 1 - iter 180/304 - loss 1.41758456 - time (sec): 9.25 - samples/sec: 1965.58 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:02:17,275 epoch 1 - iter 210/304 - loss 1.24794432 - time (sec): 10.78 - samples/sec: 2005.91 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:02:18,649 epoch 1 - iter 240/304 - loss 1.12096737 - time (sec): 12.15 - samples/sec: 2033.68 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:02:20,022 epoch 1 - iter 270/304 - loss 1.02357630 - time (sec): 13.52 - samples/sec: 2061.13 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:02:21,387 epoch 1 - iter 300/304 - loss 0.94687326 - time (sec): 14.89 - samples/sec: 2061.06 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:02:21,564 ----------------------------------------------------------------------------------------------------
2023-10-17 10:02:21,564 EPOCH 1 done: loss 0.9389 - lr: 0.000030
2023-10-17 10:02:22,726 DEV : loss 0.21260005235671997 - f1-score (micro avg) 0.5972
2023-10-17 10:02:22,733 saving best model
2023-10-17 10:02:23,111 ----------------------------------------------------------------------------------------------------
2023-10-17 10:02:24,494 epoch 2 - iter 30/304 - loss 0.17504565 - time (sec): 1.38 - samples/sec: 2386.47 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:02:25,873 epoch 2 - iter 60/304 - loss 0.19569853 - time (sec): 2.76 - samples/sec: 2275.26 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:02:27,289 epoch 2 - iter 90/304 - loss 0.17193454 - time (sec): 4.18 - samples/sec: 2233.12 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:02:28,670 epoch 2 - iter 120/304 - loss 0.16934135 - time (sec): 5.56 - samples/sec: 2241.59 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:02:30,052 epoch 2 - iter 150/304 - loss 0.17176923 - time (sec): 6.94 - samples/sec: 2241.61 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:02:31,440 epoch 2 - iter 180/304 - loss 0.16894244 - time (sec): 8.33 - samples/sec: 2262.59 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:02:32,802 epoch 2 - iter 210/304 - loss 0.15988629 - time (sec): 9.69 - samples/sec: 2220.87 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:02:34,148 epoch 2 - iter 240/304 - loss 0.15375049 - time (sec): 11.03 - samples/sec: 2254.22 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:02:35,519 epoch 2 - iter 270/304 - loss 0.14916921 - time (sec): 12.41 - samples/sec: 2258.76 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:02:36,884 epoch 2 - iter 300/304 - loss 0.14765257 - time (sec): 13.77 - samples/sec: 2227.32 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:02:37,069 ----------------------------------------------------------------------------------------------------
2023-10-17 10:02:37,069 EPOCH 2 done: loss 0.1465 - lr: 0.000027
2023-10-17 10:02:37,993 DEV : loss 0.16393417119979858 - f1-score (micro avg) 0.7701
2023-10-17 10:02:37,999 saving best model
2023-10-17 10:02:38,446 ----------------------------------------------------------------------------------------------------
2023-10-17 10:02:39,816 epoch 3 - iter 30/304 - loss 0.10374105 - time (sec): 1.37 - samples/sec: 2245.87 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:02:41,184 epoch 3 - iter 60/304 - loss 0.08006234 - time (sec): 2.74 - samples/sec: 2234.77 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:02:42,568 epoch 3 - iter 90/304 - loss 0.08767063 - time (sec): 4.12 - samples/sec: 2256.37 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:02:43,944 epoch 3 - iter 120/304 - loss 0.09068705 - time (sec): 5.50 - samples/sec: 2235.81 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:02:45,302 epoch 3 - iter 150/304 - loss 0.09135349 - time (sec): 6.85 - samples/sec: 2205.45 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:02:46,662 epoch 3 - iter 180/304 - loss 0.08696189 - time (sec): 8.21 - samples/sec: 2184.25 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:02:48,049 epoch 3 - iter 210/304 - loss 0.08741350 - time (sec): 9.60 - samples/sec: 2183.73 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:02:49,439 epoch 3 - iter 240/304 - loss 0.08724010 - time (sec): 10.99 - samples/sec: 2213.31 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:02:50,823 epoch 3 - iter 270/304 - loss 0.09120280 - time (sec): 12.37 - samples/sec: 2226.75 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:02:52,184 epoch 3 - iter 300/304 - loss 0.09523846 - time (sec): 13.74 - samples/sec: 2225.39 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:02:52,362 ----------------------------------------------------------------------------------------------------
2023-10-17 10:02:52,362 EPOCH 3 done: loss 0.0944 - lr: 0.000023
2023-10-17 10:02:53,289 DEV : loss 0.16942133009433746 - f1-score (micro avg) 0.8286
2023-10-17 10:02:53,296 saving best model
2023-10-17 10:02:53,760 ----------------------------------------------------------------------------------------------------
2023-10-17 10:02:55,191 epoch 4 - iter 30/304 - loss 0.04129966 - time (sec): 1.43 - samples/sec: 2476.54 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:02:56,590 epoch 4 - iter 60/304 - loss 0.06484886 - time (sec): 2.83 - samples/sec: 2319.89 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:02:57,960 epoch 4 - iter 90/304 - loss 0.06438072 - time (sec): 4.20 - samples/sec: 2273.59 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:02:59,338 epoch 4 - iter 120/304 - loss 0.06760369 - time (sec): 5.57 - samples/sec: 2226.87 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:03:00,715 epoch 4 - iter 150/304 - loss 0.06231282 - time (sec): 6.95 - samples/sec: 2234.11 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:03:02,062 epoch 4 - iter 180/304 - loss 0.06648038 - time (sec): 8.30 - samples/sec: 2255.68 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:03:03,426 epoch 4 - iter 210/304 - loss 0.06648870 - time (sec): 9.66 - samples/sec: 2253.44 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:03:04,800 epoch 4 - iter 240/304 - loss 0.06551934 - time (sec): 11.04 - samples/sec: 2247.17 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:03:06,155 epoch 4 - iter 270/304 - loss 0.06434722 - time (sec): 12.39 - samples/sec: 2241.03 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:03:07,521 epoch 4 - iter 300/304 - loss 0.06561613 - time (sec): 13.76 - samples/sec: 2221.15 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:03:07,701 ----------------------------------------------------------------------------------------------------
2023-10-17 10:03:07,701 EPOCH 4 done: loss 0.0668 - lr: 0.000020
2023-10-17 10:03:08,890 DEV : loss 0.1767215132713318 - f1-score (micro avg) 0.8341
2023-10-17 10:03:08,900 saving best model
2023-10-17 10:03:09,405 ----------------------------------------------------------------------------------------------------
2023-10-17 10:03:11,078 epoch 5 - iter 30/304 - loss 0.05590618 - time (sec): 1.67 - samples/sec: 1906.74 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:03:12,658 epoch 5 - iter 60/304 - loss 0.05771895 - time (sec): 3.25 - samples/sec: 1878.22 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:03:14,217 epoch 5 - iter 90/304 - loss 0.05929665 - time (sec): 4.81 - samples/sec: 1968.46 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:03:15,769 epoch 5 - iter 120/304 - loss 0.05479179 - time (sec): 6.36 - samples/sec: 1974.78 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:03:17,304 epoch 5 - iter 150/304 - loss 0.05431980 - time (sec): 7.89 - samples/sec: 1985.62 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:03:18,698 epoch 5 - iter 180/304 - loss 0.05638198 - time (sec): 9.29 - samples/sec: 2031.94 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:03:20,046 epoch 5 - iter 210/304 - loss 0.05557100 - time (sec): 10.63 - samples/sec: 2041.16 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:03:21,448 epoch 5 - iter 240/304 - loss 0.05729684 - time (sec): 12.04 - samples/sec: 2075.15 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:03:22,833 epoch 5 - iter 270/304 - loss 0.05637455 - time (sec): 13.42 - samples/sec: 2074.00 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:03:24,197 epoch 5 - iter 300/304 - loss 0.05370632 - time (sec): 14.79 - samples/sec: 2077.13 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:03:24,375 ----------------------------------------------------------------------------------------------------
2023-10-17 10:03:24,375 EPOCH 5 done: loss 0.0533 - lr: 0.000017
2023-10-17 10:03:25,363 DEV : loss 0.17217454314231873 - f1-score (micro avg) 0.849
2023-10-17 10:03:25,369 saving best model
2023-10-17 10:03:25,859 ----------------------------------------------------------------------------------------------------
2023-10-17 10:03:27,287 epoch 6 - iter 30/304 - loss 0.04404318 - time (sec): 1.43 - samples/sec: 2140.42 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:03:28,747 epoch 6 - iter 60/304 - loss 0.03709370 - time (sec): 2.89 - samples/sec: 2158.44 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:03:30,146 epoch 6 - iter 90/304 - loss 0.03226865 - time (sec): 4.28 - samples/sec: 2100.40 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:03:31,480 epoch 6 - iter 120/304 - loss 0.03019694 - time (sec): 5.62 - samples/sec: 2202.69 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:03:32,801 epoch 6 - iter 150/304 - loss 0.02705743 - time (sec): 6.94 - samples/sec: 2191.22 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:03:34,118 epoch 6 - iter 180/304 - loss 0.03498976 - time (sec): 8.26 - samples/sec: 2232.42 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:03:35,433 epoch 6 - iter 210/304 - loss 0.03557122 - time (sec): 9.57 - samples/sec: 2205.32 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:03:36,741 epoch 6 - iter 240/304 - loss 0.03641619 - time (sec): 10.88 - samples/sec: 2265.09 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:03:38,065 epoch 6 - iter 270/304 - loss 0.03625595 - time (sec): 12.20 - samples/sec: 2276.50 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:03:39,381 epoch 6 - iter 300/304 - loss 0.03763049 - time (sec): 13.52 - samples/sec: 2275.63 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:03:39,549 ----------------------------------------------------------------------------------------------------
2023-10-17 10:03:39,549 EPOCH 6 done: loss 0.0374 - lr: 0.000013
2023-10-17 10:03:40,537 DEV : loss 0.18530553579330444 - f1-score (micro avg) 0.8527
2023-10-17 10:03:40,544 saving best model
2023-10-17 10:03:41,042 ----------------------------------------------------------------------------------------------------
2023-10-17 10:03:42,342 epoch 7 - iter 30/304 - loss 0.03976232 - time (sec): 1.30 - samples/sec: 2304.43 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:03:43,660 epoch 7 - iter 60/304 - loss 0.02405073 - time (sec): 2.62 - samples/sec: 2395.83 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:03:44,965 epoch 7 - iter 90/304 - loss 0.02777625 - time (sec): 3.92 - samples/sec: 2278.17 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:03:46,339 epoch 7 - iter 120/304 - loss 0.03062262 - time (sec): 5.29 - samples/sec: 2293.00 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:03:47,717 epoch 7 - iter 150/304 - loss 0.03593935 - time (sec): 6.67 - samples/sec: 2291.32 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:03:49,104 epoch 7 - iter 180/304 - loss 0.03452574 - time (sec): 8.06 - samples/sec: 2282.21 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:03:50,497 epoch 7 - iter 210/304 - loss 0.03181745 - time (sec): 9.45 - samples/sec: 2279.81 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:03:51,923 epoch 7 - iter 240/304 - loss 0.03048670 - time (sec): 10.88 - samples/sec: 2268.64 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:03:53,315 epoch 7 - iter 270/304 - loss 0.03055172 - time (sec): 12.27 - samples/sec: 2249.14 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:03:54,687 epoch 7 - iter 300/304 - loss 0.03050690 - time (sec): 13.64 - samples/sec: 2248.61 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:03:54,885 ----------------------------------------------------------------------------------------------------
2023-10-17 10:03:54,885 EPOCH 7 done: loss 0.0302 - lr: 0.000010
2023-10-17 10:03:55,864 DEV : loss 0.19813692569732666 - f1-score (micro avg) 0.8565
2023-10-17 10:03:55,870 saving best model
2023-10-17 10:03:56,356 ----------------------------------------------------------------------------------------------------
2023-10-17 10:03:58,046 epoch 8 - iter 30/304 - loss 0.03823214 - time (sec): 1.69 - samples/sec: 1691.31 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:03:59,768 epoch 8 - iter 60/304 - loss 0.02482736 - time (sec): 3.41 - samples/sec: 1704.41 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:04:01,384 epoch 8 - iter 90/304 - loss 0.01791624 - time (sec): 5.03 - samples/sec: 1781.49 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:04:02,988 epoch 8 - iter 120/304 - loss 0.01629216 - time (sec): 6.63 - samples/sec: 1797.58 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:04:04,571 epoch 8 - iter 150/304 - loss 0.01543196 - time (sec): 8.21 - samples/sec: 1841.85 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:04:06,151 epoch 8 - iter 180/304 - loss 0.01356072 - time (sec): 9.79 - samples/sec: 1857.55 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:04:07,705 epoch 8 - iter 210/304 - loss 0.01371063 - time (sec): 11.35 - samples/sec: 1872.20 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:04:09,297 epoch 8 - iter 240/304 - loss 0.01484184 - time (sec): 12.94 - samples/sec: 1895.52 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:04:10,861 epoch 8 - iter 270/304 - loss 0.02034814 - time (sec): 14.50 - samples/sec: 1902.70 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:04:12,436 epoch 8 - iter 300/304 - loss 0.02174712 - time (sec): 16.08 - samples/sec: 1902.59 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:04:12,647 ----------------------------------------------------------------------------------------------------
2023-10-17 10:04:12,647 EPOCH 8 done: loss 0.0215 - lr: 0.000007
2023-10-17 10:04:13,607 DEV : loss 0.19218282401561737 - f1-score (micro avg) 0.849
2023-10-17 10:04:13,616 ----------------------------------------------------------------------------------------------------
2023-10-17 10:04:15,165 epoch 9 - iter 30/304 - loss 0.01048409 - time (sec): 1.55 - samples/sec: 1950.67 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:04:16,690 epoch 9 - iter 60/304 - loss 0.01542063 - time (sec): 3.07 - samples/sec: 2014.43 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:04:18,203 epoch 9 - iter 90/304 - loss 0.01289720 - time (sec): 4.59 - samples/sec: 2041.18 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:04:19,591 epoch 9 - iter 120/304 - loss 0.01607044 - time (sec): 5.97 - samples/sec: 2067.12 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:04:20,997 epoch 9 - iter 150/304 - loss 0.01530081 - time (sec): 7.38 - samples/sec: 2054.90 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:04:22,540 epoch 9 - iter 180/304 - loss 0.01745658 - time (sec): 8.92 - samples/sec: 2057.70 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:04:24,203 epoch 9 - iter 210/304 - loss 0.01717509 - time (sec): 10.59 - samples/sec: 2031.59 - lr: 0.000004 - momentum: 0.000000
2023-10-17 10:04:25,826 epoch 9 - iter 240/304 - loss 0.01617675 - time (sec): 12.21 - samples/sec: 2003.32 - lr: 0.000004 - momentum: 0.000000
2023-10-17 10:04:27,432 epoch 9 - iter 270/304 - loss 0.01706249 - time (sec): 13.82 - samples/sec: 1993.05 - lr: 0.000004 - momentum: 0.000000
2023-10-17 10:04:29,064 epoch 9 - iter 300/304 - loss 0.01619386 - time (sec): 15.45 - samples/sec: 1982.08 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:04:29,272 ----------------------------------------------------------------------------------------------------
2023-10-17 10:04:29,272 EPOCH 9 done: loss 0.0161 - lr: 0.000003
2023-10-17 10:04:30,289 DEV : loss 0.1948387622833252 - f1-score (micro avg) 0.8487
2023-10-17 10:04:30,303 ----------------------------------------------------------------------------------------------------
2023-10-17 10:04:31,889 epoch 10 - iter 30/304 - loss 0.01311954 - time (sec): 1.58 - samples/sec: 2032.56 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:04:33,472 epoch 10 - iter 60/304 - loss 0.01765805 - time (sec): 3.17 - samples/sec: 1948.18 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:04:35,109 epoch 10 - iter 90/304 - loss 0.01320737 - time (sec): 4.80 - samples/sec: 1883.58 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:04:36,741 epoch 10 - iter 120/304 - loss 0.01166367 - time (sec): 6.44 - samples/sec: 1927.18 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:04:38,344 epoch 10 - iter 150/304 - loss 0.00951810 - time (sec): 8.04 - samples/sec: 1947.82 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:04:39,964 epoch 10 - iter 180/304 - loss 0.01010278 - time (sec): 9.66 - samples/sec: 1936.51 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:04:41,533 epoch 10 - iter 210/304 - loss 0.00992968 - time (sec): 11.23 - samples/sec: 1921.63 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:04:43,132 epoch 10 - iter 240/304 - loss 0.01162759 - time (sec): 12.83 - samples/sec: 1912.66 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:04:44,736 epoch 10 - iter 270/304 - loss 0.01275249 - time (sec): 14.43 - samples/sec: 1908.14 - lr: 0.000000 - momentum: 0.000000
2023-10-17 10:04:46,364 epoch 10 - iter 300/304 - loss 0.01301596 - time (sec): 16.06 - samples/sec: 1903.08 - lr: 0.000000 - momentum: 0.000000
2023-10-17 10:04:46,578 ----------------------------------------------------------------------------------------------------
2023-10-17 10:04:46,579 EPOCH 10 done: loss 0.0132 - lr: 0.000000
2023-10-17 10:04:47,632 DEV : loss 0.1933870017528534 - f1-score (micro avg) 0.8609
2023-10-17 10:04:47,640 saving best model
2023-10-17 10:04:48,444 ----------------------------------------------------------------------------------------------------
2023-10-17 10:04:48,446 Loading model from best epoch ...
2023-10-17 10:04:49,925 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 10:04:50,913
Results:
- F-score (micro) 0.8139
- F-score (macro) 0.6574
- Accuracy 0.691
By class:
precision recall f1-score support
scope 0.7548 0.7748 0.7647 151
pers 0.8182 0.9375 0.8738 96
work 0.8155 0.8842 0.8485 95
loc 1.0000 0.6667 0.8000 3
date 0.0000 0.0000 0.0000 3
micro avg 0.7876 0.8420 0.8139 348
macro avg 0.6777 0.6526 0.6574 348
weighted avg 0.7845 0.8420 0.8114 348
2023-10-17 10:04:50,914 ----------------------------------------------------------------------------------------------------
|