File size: 24,069 Bytes
6007d78 |
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 |
2023-10-17 21:26:40,875 ----------------------------------------------------------------------------------------------------
2023-10-17 21:26:40,876 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=21, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 21:26:40,876 ----------------------------------------------------------------------------------------------------
2023-10-17 21:26:40,876 MultiCorpus: 5901 train + 1287 dev + 1505 test sentences
- NER_HIPE_2022 Corpus: 5901 train + 1287 dev + 1505 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/fr/with_doc_seperator
2023-10-17 21:26:40,876 ----------------------------------------------------------------------------------------------------
2023-10-17 21:26:40,876 Train: 5901 sentences
2023-10-17 21:26:40,876 (train_with_dev=False, train_with_test=False)
2023-10-17 21:26:40,876 ----------------------------------------------------------------------------------------------------
2023-10-17 21:26:40,876 Training Params:
2023-10-17 21:26:40,876 - learning_rate: "3e-05"
2023-10-17 21:26:40,876 - mini_batch_size: "8"
2023-10-17 21:26:40,876 - max_epochs: "10"
2023-10-17 21:26:40,876 - shuffle: "True"
2023-10-17 21:26:40,876 ----------------------------------------------------------------------------------------------------
2023-10-17 21:26:40,876 Plugins:
2023-10-17 21:26:40,877 - TensorboardLogger
2023-10-17 21:26:40,877 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 21:26:40,877 ----------------------------------------------------------------------------------------------------
2023-10-17 21:26:40,877 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 21:26:40,877 - metric: "('micro avg', 'f1-score')"
2023-10-17 21:26:40,877 ----------------------------------------------------------------------------------------------------
2023-10-17 21:26:40,877 Computation:
2023-10-17 21:26:40,877 - compute on device: cuda:0
2023-10-17 21:26:40,877 - embedding storage: none
2023-10-17 21:26:40,877 ----------------------------------------------------------------------------------------------------
2023-10-17 21:26:40,877 Model training base path: "hmbench-hipe2020/fr-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-17 21:26:40,877 ----------------------------------------------------------------------------------------------------
2023-10-17 21:26:40,877 ----------------------------------------------------------------------------------------------------
2023-10-17 21:26:40,877 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 21:26:45,615 epoch 1 - iter 73/738 - loss 3.49952752 - time (sec): 4.74 - samples/sec: 3401.60 - lr: 0.000003 - momentum: 0.000000
2023-10-17 21:26:51,525 epoch 1 - iter 146/738 - loss 2.16207004 - time (sec): 10.65 - samples/sec: 3375.23 - lr: 0.000006 - momentum: 0.000000
2023-10-17 21:26:56,443 epoch 1 - iter 219/738 - loss 1.66274405 - time (sec): 15.56 - samples/sec: 3326.52 - lr: 0.000009 - momentum: 0.000000
2023-10-17 21:27:00,716 epoch 1 - iter 292/738 - loss 1.38721801 - time (sec): 19.84 - samples/sec: 3357.65 - lr: 0.000012 - momentum: 0.000000
2023-10-17 21:27:05,522 epoch 1 - iter 365/738 - loss 1.18355146 - time (sec): 24.64 - samples/sec: 3350.31 - lr: 0.000015 - momentum: 0.000000
2023-10-17 21:27:10,816 epoch 1 - iter 438/738 - loss 1.02870123 - time (sec): 29.94 - samples/sec: 3358.32 - lr: 0.000018 - momentum: 0.000000
2023-10-17 21:27:15,600 epoch 1 - iter 511/738 - loss 0.91506435 - time (sec): 34.72 - samples/sec: 3358.58 - lr: 0.000021 - momentum: 0.000000
2023-10-17 21:27:20,128 epoch 1 - iter 584/738 - loss 0.83281866 - time (sec): 39.25 - samples/sec: 3372.24 - lr: 0.000024 - momentum: 0.000000
2023-10-17 21:27:24,700 epoch 1 - iter 657/738 - loss 0.76709624 - time (sec): 43.82 - samples/sec: 3378.50 - lr: 0.000027 - momentum: 0.000000
2023-10-17 21:27:29,463 epoch 1 - iter 730/738 - loss 0.70974504 - time (sec): 48.59 - samples/sec: 3382.19 - lr: 0.000030 - momentum: 0.000000
2023-10-17 21:27:29,971 ----------------------------------------------------------------------------------------------------
2023-10-17 21:27:29,972 EPOCH 1 done: loss 0.7025 - lr: 0.000030
2023-10-17 21:27:35,758 DEV : loss 0.11762610077857971 - f1-score (micro avg) 0.7482
2023-10-17 21:27:35,791 saving best model
2023-10-17 21:27:36,662 ----------------------------------------------------------------------------------------------------
2023-10-17 21:27:41,652 epoch 2 - iter 73/738 - loss 0.14984083 - time (sec): 4.99 - samples/sec: 3222.14 - lr: 0.000030 - momentum: 0.000000
2023-10-17 21:27:46,811 epoch 2 - iter 146/738 - loss 0.15409294 - time (sec): 10.15 - samples/sec: 3379.77 - lr: 0.000029 - momentum: 0.000000
2023-10-17 21:27:51,908 epoch 2 - iter 219/738 - loss 0.14322277 - time (sec): 15.24 - samples/sec: 3332.70 - lr: 0.000029 - momentum: 0.000000
2023-10-17 21:27:56,414 epoch 2 - iter 292/738 - loss 0.13804033 - time (sec): 19.75 - samples/sec: 3368.52 - lr: 0.000029 - momentum: 0.000000
2023-10-17 21:28:01,318 epoch 2 - iter 365/738 - loss 0.13219670 - time (sec): 24.65 - samples/sec: 3397.53 - lr: 0.000028 - momentum: 0.000000
2023-10-17 21:28:06,597 epoch 2 - iter 438/738 - loss 0.13134819 - time (sec): 29.93 - samples/sec: 3363.86 - lr: 0.000028 - momentum: 0.000000
2023-10-17 21:28:11,528 epoch 2 - iter 511/738 - loss 0.12866441 - time (sec): 34.86 - samples/sec: 3359.08 - lr: 0.000028 - momentum: 0.000000
2023-10-17 21:28:16,885 epoch 2 - iter 584/738 - loss 0.12449709 - time (sec): 40.22 - samples/sec: 3326.61 - lr: 0.000027 - momentum: 0.000000
2023-10-17 21:28:21,525 epoch 2 - iter 657/738 - loss 0.12254254 - time (sec): 44.86 - samples/sec: 3341.34 - lr: 0.000027 - momentum: 0.000000
2023-10-17 21:28:26,153 epoch 2 - iter 730/738 - loss 0.12293368 - time (sec): 49.49 - samples/sec: 3330.98 - lr: 0.000027 - momentum: 0.000000
2023-10-17 21:28:26,639 ----------------------------------------------------------------------------------------------------
2023-10-17 21:28:26,639 EPOCH 2 done: loss 0.1226 - lr: 0.000027
2023-10-17 21:28:37,830 DEV : loss 0.08953238278627396 - f1-score (micro avg) 0.8325
2023-10-17 21:28:37,863 saving best model
2023-10-17 21:28:38,345 ----------------------------------------------------------------------------------------------------
2023-10-17 21:28:43,166 epoch 3 - iter 73/738 - loss 0.06498284 - time (sec): 4.82 - samples/sec: 3316.94 - lr: 0.000026 - momentum: 0.000000
2023-10-17 21:28:48,268 epoch 3 - iter 146/738 - loss 0.07253182 - time (sec): 9.92 - samples/sec: 3292.11 - lr: 0.000026 - momentum: 0.000000
2023-10-17 21:28:53,207 epoch 3 - iter 219/738 - loss 0.07369571 - time (sec): 14.86 - samples/sec: 3307.50 - lr: 0.000026 - momentum: 0.000000
2023-10-17 21:28:58,390 epoch 3 - iter 292/738 - loss 0.07104734 - time (sec): 20.04 - samples/sec: 3291.04 - lr: 0.000025 - momentum: 0.000000
2023-10-17 21:29:03,747 epoch 3 - iter 365/738 - loss 0.07010578 - time (sec): 25.40 - samples/sec: 3289.69 - lr: 0.000025 - momentum: 0.000000
2023-10-17 21:29:08,795 epoch 3 - iter 438/738 - loss 0.07173198 - time (sec): 30.44 - samples/sec: 3280.44 - lr: 0.000025 - momentum: 0.000000
2023-10-17 21:29:13,423 epoch 3 - iter 511/738 - loss 0.07041391 - time (sec): 35.07 - samples/sec: 3281.82 - lr: 0.000024 - momentum: 0.000000
2023-10-17 21:29:18,816 epoch 3 - iter 584/738 - loss 0.06977851 - time (sec): 40.47 - samples/sec: 3282.39 - lr: 0.000024 - momentum: 0.000000
2023-10-17 21:29:23,695 epoch 3 - iter 657/738 - loss 0.06917953 - time (sec): 45.34 - samples/sec: 3292.70 - lr: 0.000024 - momentum: 0.000000
2023-10-17 21:29:28,226 epoch 3 - iter 730/738 - loss 0.07009721 - time (sec): 49.87 - samples/sec: 3304.46 - lr: 0.000023 - momentum: 0.000000
2023-10-17 21:29:28,708 ----------------------------------------------------------------------------------------------------
2023-10-17 21:29:28,709 EPOCH 3 done: loss 0.0701 - lr: 0.000023
2023-10-17 21:29:40,107 DEV : loss 0.1102706640958786 - f1-score (micro avg) 0.8405
2023-10-17 21:29:40,146 saving best model
2023-10-17 21:29:40,654 ----------------------------------------------------------------------------------------------------
2023-10-17 21:29:45,608 epoch 4 - iter 73/738 - loss 0.02968515 - time (sec): 4.95 - samples/sec: 3166.16 - lr: 0.000023 - momentum: 0.000000
2023-10-17 21:29:50,421 epoch 4 - iter 146/738 - loss 0.03527936 - time (sec): 9.76 - samples/sec: 3318.65 - lr: 0.000023 - momentum: 0.000000
2023-10-17 21:29:55,855 epoch 4 - iter 219/738 - loss 0.03661792 - time (sec): 15.20 - samples/sec: 3291.89 - lr: 0.000022 - momentum: 0.000000
2023-10-17 21:30:00,595 epoch 4 - iter 292/738 - loss 0.04111833 - time (sec): 19.94 - samples/sec: 3293.63 - lr: 0.000022 - momentum: 0.000000
2023-10-17 21:30:04,902 epoch 4 - iter 365/738 - loss 0.04094378 - time (sec): 24.25 - samples/sec: 3299.44 - lr: 0.000022 - momentum: 0.000000
2023-10-17 21:30:09,927 epoch 4 - iter 438/738 - loss 0.04325232 - time (sec): 29.27 - samples/sec: 3265.15 - lr: 0.000021 - momentum: 0.000000
2023-10-17 21:30:15,565 epoch 4 - iter 511/738 - loss 0.04681831 - time (sec): 34.91 - samples/sec: 3290.65 - lr: 0.000021 - momentum: 0.000000
2023-10-17 21:30:20,255 epoch 4 - iter 584/738 - loss 0.04786698 - time (sec): 39.60 - samples/sec: 3290.40 - lr: 0.000021 - momentum: 0.000000
2023-10-17 21:30:25,799 epoch 4 - iter 657/738 - loss 0.04774352 - time (sec): 45.14 - samples/sec: 3279.63 - lr: 0.000020 - momentum: 0.000000
2023-10-17 21:30:30,533 epoch 4 - iter 730/738 - loss 0.04825194 - time (sec): 49.88 - samples/sec: 3292.97 - lr: 0.000020 - momentum: 0.000000
2023-10-17 21:30:31,199 ----------------------------------------------------------------------------------------------------
2023-10-17 21:30:31,199 EPOCH 4 done: loss 0.0483 - lr: 0.000020
2023-10-17 21:30:42,413 DEV : loss 0.1353892683982849 - f1-score (micro avg) 0.8416
2023-10-17 21:30:42,446 saving best model
2023-10-17 21:30:42,928 ----------------------------------------------------------------------------------------------------
2023-10-17 21:30:47,459 epoch 5 - iter 73/738 - loss 0.03440303 - time (sec): 4.53 - samples/sec: 3382.11 - lr: 0.000020 - momentum: 0.000000
2023-10-17 21:30:52,224 epoch 5 - iter 146/738 - loss 0.03591970 - time (sec): 9.29 - samples/sec: 3346.20 - lr: 0.000019 - momentum: 0.000000
2023-10-17 21:30:57,359 epoch 5 - iter 219/738 - loss 0.03487105 - time (sec): 14.43 - samples/sec: 3341.94 - lr: 0.000019 - momentum: 0.000000
2023-10-17 21:31:02,051 epoch 5 - iter 292/738 - loss 0.03396132 - time (sec): 19.12 - samples/sec: 3302.02 - lr: 0.000019 - momentum: 0.000000
2023-10-17 21:31:07,353 epoch 5 - iter 365/738 - loss 0.03575017 - time (sec): 24.42 - samples/sec: 3292.46 - lr: 0.000018 - momentum: 0.000000
2023-10-17 21:31:13,225 epoch 5 - iter 438/738 - loss 0.03630382 - time (sec): 30.29 - samples/sec: 3317.94 - lr: 0.000018 - momentum: 0.000000
2023-10-17 21:31:17,948 epoch 5 - iter 511/738 - loss 0.03628542 - time (sec): 35.02 - samples/sec: 3306.47 - lr: 0.000018 - momentum: 0.000000
2023-10-17 21:31:22,791 epoch 5 - iter 584/738 - loss 0.03570918 - time (sec): 39.86 - samples/sec: 3311.13 - lr: 0.000017 - momentum: 0.000000
2023-10-17 21:31:27,930 epoch 5 - iter 657/738 - loss 0.03596680 - time (sec): 45.00 - samples/sec: 3295.33 - lr: 0.000017 - momentum: 0.000000
2023-10-17 21:31:32,885 epoch 5 - iter 730/738 - loss 0.03568567 - time (sec): 49.95 - samples/sec: 3298.95 - lr: 0.000017 - momentum: 0.000000
2023-10-17 21:31:33,323 ----------------------------------------------------------------------------------------------------
2023-10-17 21:31:33,323 EPOCH 5 done: loss 0.0355 - lr: 0.000017
2023-10-17 21:31:44,632 DEV : loss 0.14825424551963806 - f1-score (micro avg) 0.8515
2023-10-17 21:31:44,669 saving best model
2023-10-17 21:31:45,155 ----------------------------------------------------------------------------------------------------
2023-10-17 21:31:50,379 epoch 6 - iter 73/738 - loss 0.01815114 - time (sec): 5.22 - samples/sec: 3277.80 - lr: 0.000016 - momentum: 0.000000
2023-10-17 21:31:54,804 epoch 6 - iter 146/738 - loss 0.02485180 - time (sec): 9.64 - samples/sec: 3307.68 - lr: 0.000016 - momentum: 0.000000
2023-10-17 21:32:00,352 epoch 6 - iter 219/738 - loss 0.02316966 - time (sec): 15.19 - samples/sec: 3156.11 - lr: 0.000016 - momentum: 0.000000
2023-10-17 21:32:05,025 epoch 6 - iter 292/738 - loss 0.02044324 - time (sec): 19.87 - samples/sec: 3184.54 - lr: 0.000015 - momentum: 0.000000
2023-10-17 21:32:09,801 epoch 6 - iter 365/738 - loss 0.01978515 - time (sec): 24.64 - samples/sec: 3225.44 - lr: 0.000015 - momentum: 0.000000
2023-10-17 21:32:15,135 epoch 6 - iter 438/738 - loss 0.02024243 - time (sec): 29.98 - samples/sec: 3213.94 - lr: 0.000015 - momentum: 0.000000
2023-10-17 21:32:20,409 epoch 6 - iter 511/738 - loss 0.02251455 - time (sec): 35.25 - samples/sec: 3211.11 - lr: 0.000014 - momentum: 0.000000
2023-10-17 21:32:25,237 epoch 6 - iter 584/738 - loss 0.02378898 - time (sec): 40.08 - samples/sec: 3234.89 - lr: 0.000014 - momentum: 0.000000
2023-10-17 21:32:30,304 epoch 6 - iter 657/738 - loss 0.02399881 - time (sec): 45.14 - samples/sec: 3248.20 - lr: 0.000014 - momentum: 0.000000
2023-10-17 21:32:35,259 epoch 6 - iter 730/738 - loss 0.02430477 - time (sec): 50.10 - samples/sec: 3248.64 - lr: 0.000013 - momentum: 0.000000
2023-10-17 21:32:36,208 ----------------------------------------------------------------------------------------------------
2023-10-17 21:32:36,208 EPOCH 6 done: loss 0.0249 - lr: 0.000013
2023-10-17 21:32:47,447 DEV : loss 0.15200623869895935 - f1-score (micro avg) 0.8486
2023-10-17 21:32:47,479 ----------------------------------------------------------------------------------------------------
2023-10-17 21:32:52,519 epoch 7 - iter 73/738 - loss 0.02120244 - time (sec): 5.04 - samples/sec: 3271.07 - lr: 0.000013 - momentum: 0.000000
2023-10-17 21:32:57,621 epoch 7 - iter 146/738 - loss 0.01625364 - time (sec): 10.14 - samples/sec: 3284.37 - lr: 0.000013 - momentum: 0.000000
2023-10-17 21:33:02,523 epoch 7 - iter 219/738 - loss 0.01923714 - time (sec): 15.04 - samples/sec: 3233.79 - lr: 0.000012 - momentum: 0.000000
2023-10-17 21:33:07,707 epoch 7 - iter 292/738 - loss 0.01798684 - time (sec): 20.23 - samples/sec: 3282.97 - lr: 0.000012 - momentum: 0.000000
2023-10-17 21:33:12,635 epoch 7 - iter 365/738 - loss 0.01737053 - time (sec): 25.15 - samples/sec: 3289.30 - lr: 0.000012 - momentum: 0.000000
2023-10-17 21:33:17,385 epoch 7 - iter 438/738 - loss 0.01786620 - time (sec): 29.91 - samples/sec: 3292.00 - lr: 0.000011 - momentum: 0.000000
2023-10-17 21:33:22,339 epoch 7 - iter 511/738 - loss 0.01716618 - time (sec): 34.86 - samples/sec: 3311.50 - lr: 0.000011 - momentum: 0.000000
2023-10-17 21:33:27,483 epoch 7 - iter 584/738 - loss 0.01905246 - time (sec): 40.00 - samples/sec: 3293.23 - lr: 0.000011 - momentum: 0.000000
2023-10-17 21:33:32,841 epoch 7 - iter 657/738 - loss 0.01919180 - time (sec): 45.36 - samples/sec: 3291.41 - lr: 0.000010 - momentum: 0.000000
2023-10-17 21:33:37,560 epoch 7 - iter 730/738 - loss 0.01935287 - time (sec): 50.08 - samples/sec: 3281.30 - lr: 0.000010 - momentum: 0.000000
2023-10-17 21:33:38,229 ----------------------------------------------------------------------------------------------------
2023-10-17 21:33:38,229 EPOCH 7 done: loss 0.0192 - lr: 0.000010
2023-10-17 21:33:49,459 DEV : loss 0.18122123181819916 - f1-score (micro avg) 0.8484
2023-10-17 21:33:49,490 ----------------------------------------------------------------------------------------------------
2023-10-17 21:33:54,535 epoch 8 - iter 73/738 - loss 0.01103084 - time (sec): 5.04 - samples/sec: 3291.02 - lr: 0.000010 - momentum: 0.000000
2023-10-17 21:33:59,503 epoch 8 - iter 146/738 - loss 0.01060912 - time (sec): 10.01 - samples/sec: 3400.07 - lr: 0.000009 - momentum: 0.000000
2023-10-17 21:34:04,704 epoch 8 - iter 219/738 - loss 0.01630488 - time (sec): 15.21 - samples/sec: 3383.60 - lr: 0.000009 - momentum: 0.000000
2023-10-17 21:34:09,971 epoch 8 - iter 292/738 - loss 0.01761086 - time (sec): 20.48 - samples/sec: 3348.32 - lr: 0.000009 - momentum: 0.000000
2023-10-17 21:34:14,862 epoch 8 - iter 365/738 - loss 0.01660792 - time (sec): 25.37 - samples/sec: 3308.23 - lr: 0.000008 - momentum: 0.000000
2023-10-17 21:34:19,664 epoch 8 - iter 438/738 - loss 0.01520866 - time (sec): 30.17 - samples/sec: 3293.22 - lr: 0.000008 - momentum: 0.000000
2023-10-17 21:34:24,642 epoch 8 - iter 511/738 - loss 0.01405476 - time (sec): 35.15 - samples/sec: 3290.91 - lr: 0.000008 - momentum: 0.000000
2023-10-17 21:34:29,213 epoch 8 - iter 584/738 - loss 0.01480907 - time (sec): 39.72 - samples/sec: 3303.80 - lr: 0.000007 - momentum: 0.000000
2023-10-17 21:34:33,873 epoch 8 - iter 657/738 - loss 0.01466298 - time (sec): 44.38 - samples/sec: 3304.79 - lr: 0.000007 - momentum: 0.000000
2023-10-17 21:34:39,470 epoch 8 - iter 730/738 - loss 0.01428222 - time (sec): 49.98 - samples/sec: 3289.07 - lr: 0.000007 - momentum: 0.000000
2023-10-17 21:34:40,123 ----------------------------------------------------------------------------------------------------
2023-10-17 21:34:40,123 EPOCH 8 done: loss 0.0141 - lr: 0.000007
2023-10-17 21:34:51,322 DEV : loss 0.18850760161876678 - f1-score (micro avg) 0.8447
2023-10-17 21:34:51,355 ----------------------------------------------------------------------------------------------------
2023-10-17 21:34:56,327 epoch 9 - iter 73/738 - loss 0.00939305 - time (sec): 4.97 - samples/sec: 3353.78 - lr: 0.000006 - momentum: 0.000000
2023-10-17 21:35:01,986 epoch 9 - iter 146/738 - loss 0.00823870 - time (sec): 10.63 - samples/sec: 3284.37 - lr: 0.000006 - momentum: 0.000000
2023-10-17 21:35:07,616 epoch 9 - iter 219/738 - loss 0.00914707 - time (sec): 16.26 - samples/sec: 3291.38 - lr: 0.000006 - momentum: 0.000000
2023-10-17 21:35:12,908 epoch 9 - iter 292/738 - loss 0.00914178 - time (sec): 21.55 - samples/sec: 3324.64 - lr: 0.000005 - momentum: 0.000000
2023-10-17 21:35:17,828 epoch 9 - iter 365/738 - loss 0.00961735 - time (sec): 26.47 - samples/sec: 3320.82 - lr: 0.000005 - momentum: 0.000000
2023-10-17 21:35:22,462 epoch 9 - iter 438/738 - loss 0.01002769 - time (sec): 31.11 - samples/sec: 3325.10 - lr: 0.000005 - momentum: 0.000000
2023-10-17 21:35:27,590 epoch 9 - iter 511/738 - loss 0.01052868 - time (sec): 36.23 - samples/sec: 3303.82 - lr: 0.000004 - momentum: 0.000000
2023-10-17 21:35:32,468 epoch 9 - iter 584/738 - loss 0.01024063 - time (sec): 41.11 - samples/sec: 3277.99 - lr: 0.000004 - momentum: 0.000000
2023-10-17 21:35:37,124 epoch 9 - iter 657/738 - loss 0.01016591 - time (sec): 45.77 - samples/sec: 3273.54 - lr: 0.000004 - momentum: 0.000000
2023-10-17 21:35:41,638 epoch 9 - iter 730/738 - loss 0.00973227 - time (sec): 50.28 - samples/sec: 3279.12 - lr: 0.000003 - momentum: 0.000000
2023-10-17 21:35:42,121 ----------------------------------------------------------------------------------------------------
2023-10-17 21:35:42,122 EPOCH 9 done: loss 0.0096 - lr: 0.000003
2023-10-17 21:35:53,382 DEV : loss 0.19131025671958923 - f1-score (micro avg) 0.8526
2023-10-17 21:35:53,417 saving best model
2023-10-17 21:35:53,919 ----------------------------------------------------------------------------------------------------
2023-10-17 21:35:59,133 epoch 10 - iter 73/738 - loss 0.00368059 - time (sec): 5.21 - samples/sec: 3264.79 - lr: 0.000003 - momentum: 0.000000
2023-10-17 21:36:04,536 epoch 10 - iter 146/738 - loss 0.00688414 - time (sec): 10.61 - samples/sec: 3219.79 - lr: 0.000003 - momentum: 0.000000
2023-10-17 21:36:09,499 epoch 10 - iter 219/738 - loss 0.00600329 - time (sec): 15.58 - samples/sec: 3229.89 - lr: 0.000002 - momentum: 0.000000
2023-10-17 21:36:14,470 epoch 10 - iter 292/738 - loss 0.00593938 - time (sec): 20.55 - samples/sec: 3206.22 - lr: 0.000002 - momentum: 0.000000
2023-10-17 21:36:19,636 epoch 10 - iter 365/738 - loss 0.00609185 - time (sec): 25.71 - samples/sec: 3220.95 - lr: 0.000002 - momentum: 0.000000
2023-10-17 21:36:24,288 epoch 10 - iter 438/738 - loss 0.00665908 - time (sec): 30.36 - samples/sec: 3232.50 - lr: 0.000001 - momentum: 0.000000
2023-10-17 21:36:29,180 epoch 10 - iter 511/738 - loss 0.00605624 - time (sec): 35.26 - samples/sec: 3248.47 - lr: 0.000001 - momentum: 0.000000
2023-10-17 21:36:33,742 epoch 10 - iter 584/738 - loss 0.00606397 - time (sec): 39.82 - samples/sec: 3256.28 - lr: 0.000001 - momentum: 0.000000
2023-10-17 21:36:39,559 epoch 10 - iter 657/738 - loss 0.00643681 - time (sec): 45.64 - samples/sec: 3289.31 - lr: 0.000000 - momentum: 0.000000
2023-10-17 21:36:44,217 epoch 10 - iter 730/738 - loss 0.00772105 - time (sec): 50.29 - samples/sec: 3282.16 - lr: 0.000000 - momentum: 0.000000
2023-10-17 21:36:44,686 ----------------------------------------------------------------------------------------------------
2023-10-17 21:36:44,686 EPOCH 10 done: loss 0.0077 - lr: 0.000000
2023-10-17 21:36:56,957 DEV : loss 0.19779689610004425 - f1-score (micro avg) 0.8516
2023-10-17 21:36:57,390 ----------------------------------------------------------------------------------------------------
2023-10-17 21:36:57,391 Loading model from best epoch ...
2023-10-17 21:36:58,862 SequenceTagger predicts: Dictionary with 21 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org, S-time, B-time, E-time, I-time, S-prod, B-prod, E-prod, I-prod
2023-10-17 21:37:05,616
Results:
- F-score (micro) 0.806
- F-score (macro) 0.7107
- Accuracy 0.694
By class:
precision recall f1-score support
loc 0.8910 0.8671 0.8789 858
pers 0.7539 0.8156 0.7835 537
org 0.5473 0.6136 0.5786 132
time 0.5556 0.6481 0.5983 54
prod 0.7843 0.6557 0.7143 61
micro avg 0.7974 0.8149 0.8060 1642
macro avg 0.7064 0.7201 0.7107 1642
weighted avg 0.8035 0.8149 0.8082 1642
2023-10-17 21:37:05,616 ----------------------------------------------------------------------------------------------------
|