File size: 24,085 Bytes
aa2412e |
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 |
2023-10-14 19:05:16,272 ----------------------------------------------------------------------------------------------------
2023-10-14 19:05:16,273 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(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): BertEncoder(
(layer): ModuleList(
(0-11): 12 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(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): BertSelfOutput(
(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): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(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)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=768, out_features=768, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=13, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-14 19:05:16,273 ----------------------------------------------------------------------------------------------------
2023-10-14 19:05:16,273 MultiCorpus: 14465 train + 1392 dev + 2432 test sentences
- NER_HIPE_2022 Corpus: 14465 train + 1392 dev + 2432 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/letemps/fr/with_doc_seperator
2023-10-14 19:05:16,273 ----------------------------------------------------------------------------------------------------
2023-10-14 19:05:16,273 Train: 14465 sentences
2023-10-14 19:05:16,273 (train_with_dev=False, train_with_test=False)
2023-10-14 19:05:16,273 ----------------------------------------------------------------------------------------------------
2023-10-14 19:05:16,273 Training Params:
2023-10-14 19:05:16,273 - learning_rate: "5e-05"
2023-10-14 19:05:16,273 - mini_batch_size: "4"
2023-10-14 19:05:16,273 - max_epochs: "10"
2023-10-14 19:05:16,273 - shuffle: "True"
2023-10-14 19:05:16,273 ----------------------------------------------------------------------------------------------------
2023-10-14 19:05:16,273 Plugins:
2023-10-14 19:05:16,273 - LinearScheduler | warmup_fraction: '0.1'
2023-10-14 19:05:16,273 ----------------------------------------------------------------------------------------------------
2023-10-14 19:05:16,273 Final evaluation on model from best epoch (best-model.pt)
2023-10-14 19:05:16,274 - metric: "('micro avg', 'f1-score')"
2023-10-14 19:05:16,274 ----------------------------------------------------------------------------------------------------
2023-10-14 19:05:16,281 Computation:
2023-10-14 19:05:16,281 - compute on device: cuda:0
2023-10-14 19:05:16,281 - embedding storage: none
2023-10-14 19:05:16,281 ----------------------------------------------------------------------------------------------------
2023-10-14 19:05:16,281 Model training base path: "hmbench-letemps/fr-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-14 19:05:16,281 ----------------------------------------------------------------------------------------------------
2023-10-14 19:05:16,281 ----------------------------------------------------------------------------------------------------
2023-10-14 19:05:32,627 epoch 1 - iter 361/3617 - loss 1.23920084 - time (sec): 16.34 - samples/sec: 2350.13 - lr: 0.000005 - momentum: 0.000000
2023-10-14 19:05:48,957 epoch 1 - iter 722/3617 - loss 0.71530722 - time (sec): 32.67 - samples/sec: 2334.39 - lr: 0.000010 - momentum: 0.000000
2023-10-14 19:06:05,427 epoch 1 - iter 1083/3617 - loss 0.53270967 - time (sec): 49.14 - samples/sec: 2315.81 - lr: 0.000015 - momentum: 0.000000
2023-10-14 19:06:22,334 epoch 1 - iter 1444/3617 - loss 0.42703208 - time (sec): 66.05 - samples/sec: 2316.46 - lr: 0.000020 - momentum: 0.000000
2023-10-14 19:06:38,623 epoch 1 - iter 1805/3617 - loss 0.36797501 - time (sec): 82.34 - samples/sec: 2312.38 - lr: 0.000025 - momentum: 0.000000
2023-10-14 19:06:54,895 epoch 1 - iter 2166/3617 - loss 0.32750828 - time (sec): 98.61 - samples/sec: 2310.32 - lr: 0.000030 - momentum: 0.000000
2023-10-14 19:07:11,180 epoch 1 - iter 2527/3617 - loss 0.29684754 - time (sec): 114.90 - samples/sec: 2307.26 - lr: 0.000035 - momentum: 0.000000
2023-10-14 19:07:27,493 epoch 1 - iter 2888/3617 - loss 0.27335721 - time (sec): 131.21 - samples/sec: 2323.15 - lr: 0.000040 - momentum: 0.000000
2023-10-14 19:07:43,729 epoch 1 - iter 3249/3617 - loss 0.25684398 - time (sec): 147.45 - samples/sec: 2318.20 - lr: 0.000045 - momentum: 0.000000
2023-10-14 19:08:00,221 epoch 1 - iter 3610/3617 - loss 0.24344102 - time (sec): 163.94 - samples/sec: 2313.29 - lr: 0.000050 - momentum: 0.000000
2023-10-14 19:08:00,523 ----------------------------------------------------------------------------------------------------
2023-10-14 19:08:00,523 EPOCH 1 done: loss 0.2432 - lr: 0.000050
2023-10-14 19:08:04,935 DEV : loss 0.1392880529165268 - f1-score (micro avg) 0.5689
2023-10-14 19:08:04,964 saving best model
2023-10-14 19:08:05,441 ----------------------------------------------------------------------------------------------------
2023-10-14 19:08:22,439 epoch 2 - iter 361/3617 - loss 0.10163212 - time (sec): 17.00 - samples/sec: 2243.75 - lr: 0.000049 - momentum: 0.000000
2023-10-14 19:08:38,735 epoch 2 - iter 722/3617 - loss 0.10138989 - time (sec): 33.29 - samples/sec: 2294.93 - lr: 0.000049 - momentum: 0.000000
2023-10-14 19:08:55,123 epoch 2 - iter 1083/3617 - loss 0.10563254 - time (sec): 49.68 - samples/sec: 2312.87 - lr: 0.000048 - momentum: 0.000000
2023-10-14 19:09:11,321 epoch 2 - iter 1444/3617 - loss 0.10583746 - time (sec): 65.88 - samples/sec: 2308.77 - lr: 0.000048 - momentum: 0.000000
2023-10-14 19:09:28,181 epoch 2 - iter 1805/3617 - loss 0.10742122 - time (sec): 82.74 - samples/sec: 2285.53 - lr: 0.000047 - momentum: 0.000000
2023-10-14 19:09:44,590 epoch 2 - iter 2166/3617 - loss 0.10649535 - time (sec): 99.15 - samples/sec: 2297.31 - lr: 0.000047 - momentum: 0.000000
2023-10-14 19:10:00,921 epoch 2 - iter 2527/3617 - loss 0.10728153 - time (sec): 115.48 - samples/sec: 2303.87 - lr: 0.000046 - momentum: 0.000000
2023-10-14 19:10:17,234 epoch 2 - iter 2888/3617 - loss 0.10535240 - time (sec): 131.79 - samples/sec: 2303.35 - lr: 0.000046 - momentum: 0.000000
2023-10-14 19:10:33,540 epoch 2 - iter 3249/3617 - loss 0.10454269 - time (sec): 148.10 - samples/sec: 2307.34 - lr: 0.000045 - momentum: 0.000000
2023-10-14 19:10:49,774 epoch 2 - iter 3610/3617 - loss 0.10433026 - time (sec): 164.33 - samples/sec: 2307.84 - lr: 0.000044 - momentum: 0.000000
2023-10-14 19:10:50,080 ----------------------------------------------------------------------------------------------------
2023-10-14 19:10:50,080 EPOCH 2 done: loss 0.1044 - lr: 0.000044
2023-10-14 19:10:55,595 DEV : loss 0.13085970282554626 - f1-score (micro avg) 0.6143
2023-10-14 19:10:55,625 saving best model
2023-10-14 19:10:56,106 ----------------------------------------------------------------------------------------------------
2023-10-14 19:11:12,432 epoch 3 - iter 361/3617 - loss 0.07918586 - time (sec): 16.32 - samples/sec: 2170.00 - lr: 0.000044 - momentum: 0.000000
2023-10-14 19:11:28,688 epoch 3 - iter 722/3617 - loss 0.07745191 - time (sec): 32.58 - samples/sec: 2257.80 - lr: 0.000043 - momentum: 0.000000
2023-10-14 19:11:44,987 epoch 3 - iter 1083/3617 - loss 0.07935488 - time (sec): 48.88 - samples/sec: 2279.45 - lr: 0.000043 - momentum: 0.000000
2023-10-14 19:12:01,266 epoch 3 - iter 1444/3617 - loss 0.08119260 - time (sec): 65.16 - samples/sec: 2314.24 - lr: 0.000042 - momentum: 0.000000
2023-10-14 19:12:17,499 epoch 3 - iter 1805/3617 - loss 0.08212278 - time (sec): 81.39 - samples/sec: 2316.72 - lr: 0.000042 - momentum: 0.000000
2023-10-14 19:12:33,816 epoch 3 - iter 2166/3617 - loss 0.08334289 - time (sec): 97.71 - samples/sec: 2326.30 - lr: 0.000041 - momentum: 0.000000
2023-10-14 19:12:50,145 epoch 3 - iter 2527/3617 - loss 0.08219281 - time (sec): 114.04 - samples/sec: 2331.35 - lr: 0.000041 - momentum: 0.000000
2023-10-14 19:13:06,352 epoch 3 - iter 2888/3617 - loss 0.08216017 - time (sec): 130.24 - samples/sec: 2329.63 - lr: 0.000040 - momentum: 0.000000
2023-10-14 19:13:23,473 epoch 3 - iter 3249/3617 - loss 0.08248407 - time (sec): 147.37 - samples/sec: 2313.36 - lr: 0.000039 - momentum: 0.000000
2023-10-14 19:13:39,840 epoch 3 - iter 3610/3617 - loss 0.08153553 - time (sec): 163.73 - samples/sec: 2316.89 - lr: 0.000039 - momentum: 0.000000
2023-10-14 19:13:40,153 ----------------------------------------------------------------------------------------------------
2023-10-14 19:13:40,153 EPOCH 3 done: loss 0.0815 - lr: 0.000039
2023-10-14 19:13:46,359 DEV : loss 0.18807156383991241 - f1-score (micro avg) 0.6297
2023-10-14 19:13:46,388 saving best model
2023-10-14 19:13:46,968 ----------------------------------------------------------------------------------------------------
2023-10-14 19:14:03,475 epoch 4 - iter 361/3617 - loss 0.06057895 - time (sec): 16.50 - samples/sec: 2377.22 - lr: 0.000038 - momentum: 0.000000
2023-10-14 19:14:19,869 epoch 4 - iter 722/3617 - loss 0.06057279 - time (sec): 32.90 - samples/sec: 2329.53 - lr: 0.000038 - momentum: 0.000000
2023-10-14 19:14:36,163 epoch 4 - iter 1083/3617 - loss 0.06014925 - time (sec): 49.19 - samples/sec: 2321.32 - lr: 0.000037 - momentum: 0.000000
2023-10-14 19:14:52,280 epoch 4 - iter 1444/3617 - loss 0.06060661 - time (sec): 65.31 - samples/sec: 2305.72 - lr: 0.000037 - momentum: 0.000000
2023-10-14 19:15:08,428 epoch 4 - iter 1805/3617 - loss 0.06034321 - time (sec): 81.46 - samples/sec: 2323.14 - lr: 0.000036 - momentum: 0.000000
2023-10-14 19:15:24,537 epoch 4 - iter 2166/3617 - loss 0.06055420 - time (sec): 97.57 - samples/sec: 2322.80 - lr: 0.000036 - momentum: 0.000000
2023-10-14 19:15:40,905 epoch 4 - iter 2527/3617 - loss 0.06148213 - time (sec): 113.93 - samples/sec: 2325.01 - lr: 0.000035 - momentum: 0.000000
2023-10-14 19:15:57,092 epoch 4 - iter 2888/3617 - loss 0.06126538 - time (sec): 130.12 - samples/sec: 2331.57 - lr: 0.000034 - momentum: 0.000000
2023-10-14 19:16:13,259 epoch 4 - iter 3249/3617 - loss 0.06194798 - time (sec): 146.29 - samples/sec: 2334.61 - lr: 0.000034 - momentum: 0.000000
2023-10-14 19:16:29,462 epoch 4 - iter 3610/3617 - loss 0.06151356 - time (sec): 162.49 - samples/sec: 2333.50 - lr: 0.000033 - momentum: 0.000000
2023-10-14 19:16:29,772 ----------------------------------------------------------------------------------------------------
2023-10-14 19:16:29,773 EPOCH 4 done: loss 0.0614 - lr: 0.000033
2023-10-14 19:16:35,943 DEV : loss 0.2195165902376175 - f1-score (micro avg) 0.6197
2023-10-14 19:16:35,973 ----------------------------------------------------------------------------------------------------
2023-10-14 19:16:52,484 epoch 5 - iter 361/3617 - loss 0.04617918 - time (sec): 16.51 - samples/sec: 2385.58 - lr: 0.000033 - momentum: 0.000000
2023-10-14 19:17:08,608 epoch 5 - iter 722/3617 - loss 0.04463891 - time (sec): 32.63 - samples/sec: 2362.80 - lr: 0.000032 - momentum: 0.000000
2023-10-14 19:17:24,908 epoch 5 - iter 1083/3617 - loss 0.04774820 - time (sec): 48.93 - samples/sec: 2352.31 - lr: 0.000032 - momentum: 0.000000
2023-10-14 19:17:41,128 epoch 5 - iter 1444/3617 - loss 0.04829543 - time (sec): 65.15 - samples/sec: 2331.91 - lr: 0.000031 - momentum: 0.000000
2023-10-14 19:17:57,525 epoch 5 - iter 1805/3617 - loss 0.04629774 - time (sec): 81.55 - samples/sec: 2341.74 - lr: 0.000031 - momentum: 0.000000
2023-10-14 19:18:13,880 epoch 5 - iter 2166/3617 - loss 0.04539649 - time (sec): 97.91 - samples/sec: 2342.96 - lr: 0.000030 - momentum: 0.000000
2023-10-14 19:18:30,165 epoch 5 - iter 2527/3617 - loss 0.04517893 - time (sec): 114.19 - samples/sec: 2345.27 - lr: 0.000029 - momentum: 0.000000
2023-10-14 19:18:46,313 epoch 5 - iter 2888/3617 - loss 0.04431891 - time (sec): 130.34 - samples/sec: 2337.69 - lr: 0.000029 - momentum: 0.000000
2023-10-14 19:19:02,438 epoch 5 - iter 3249/3617 - loss 0.04447990 - time (sec): 146.46 - samples/sec: 2341.40 - lr: 0.000028 - momentum: 0.000000
2023-10-14 19:19:18,521 epoch 5 - iter 3610/3617 - loss 0.04436946 - time (sec): 162.55 - samples/sec: 2333.42 - lr: 0.000028 - momentum: 0.000000
2023-10-14 19:19:18,823 ----------------------------------------------------------------------------------------------------
2023-10-14 19:19:18,823 EPOCH 5 done: loss 0.0443 - lr: 0.000028
2023-10-14 19:19:25,023 DEV : loss 0.3201915919780731 - f1-score (micro avg) 0.6218
2023-10-14 19:19:25,052 ----------------------------------------------------------------------------------------------------
2023-10-14 19:19:41,359 epoch 6 - iter 361/3617 - loss 0.02847515 - time (sec): 16.31 - samples/sec: 2328.31 - lr: 0.000027 - momentum: 0.000000
2023-10-14 19:19:57,743 epoch 6 - iter 722/3617 - loss 0.03162381 - time (sec): 32.69 - samples/sec: 2301.42 - lr: 0.000027 - momentum: 0.000000
2023-10-14 19:20:13,935 epoch 6 - iter 1083/3617 - loss 0.03126177 - time (sec): 48.88 - samples/sec: 2288.28 - lr: 0.000026 - momentum: 0.000000
2023-10-14 19:20:30,140 epoch 6 - iter 1444/3617 - loss 0.03237564 - time (sec): 65.09 - samples/sec: 2296.72 - lr: 0.000026 - momentum: 0.000000
2023-10-14 19:20:46,426 epoch 6 - iter 1805/3617 - loss 0.03357153 - time (sec): 81.37 - samples/sec: 2314.60 - lr: 0.000025 - momentum: 0.000000
2023-10-14 19:21:02,702 epoch 6 - iter 2166/3617 - loss 0.03474249 - time (sec): 97.65 - samples/sec: 2321.83 - lr: 0.000024 - momentum: 0.000000
2023-10-14 19:21:18,924 epoch 6 - iter 2527/3617 - loss 0.03416368 - time (sec): 113.87 - samples/sec: 2318.63 - lr: 0.000024 - momentum: 0.000000
2023-10-14 19:21:35,300 epoch 6 - iter 2888/3617 - loss 0.03478335 - time (sec): 130.25 - samples/sec: 2333.53 - lr: 0.000023 - momentum: 0.000000
2023-10-14 19:21:51,568 epoch 6 - iter 3249/3617 - loss 0.03429456 - time (sec): 146.51 - samples/sec: 2326.16 - lr: 0.000023 - momentum: 0.000000
2023-10-14 19:22:07,803 epoch 6 - iter 3610/3617 - loss 0.03403117 - time (sec): 162.75 - samples/sec: 2330.17 - lr: 0.000022 - momentum: 0.000000
2023-10-14 19:22:08,110 ----------------------------------------------------------------------------------------------------
2023-10-14 19:22:08,110 EPOCH 6 done: loss 0.0340 - lr: 0.000022
2023-10-14 19:22:13,840 DEV : loss 0.295797199010849 - f1-score (micro avg) 0.6353
2023-10-14 19:22:13,873 saving best model
2023-10-14 19:22:15,198 ----------------------------------------------------------------------------------------------------
2023-10-14 19:22:31,709 epoch 7 - iter 361/3617 - loss 0.01782055 - time (sec): 16.51 - samples/sec: 2349.80 - lr: 0.000022 - momentum: 0.000000
2023-10-14 19:22:48,225 epoch 7 - iter 722/3617 - loss 0.02038942 - time (sec): 33.02 - samples/sec: 2334.00 - lr: 0.000021 - momentum: 0.000000
2023-10-14 19:23:04,594 epoch 7 - iter 1083/3617 - loss 0.02125683 - time (sec): 49.39 - samples/sec: 2328.23 - lr: 0.000021 - momentum: 0.000000
2023-10-14 19:23:20,803 epoch 7 - iter 1444/3617 - loss 0.02186389 - time (sec): 65.60 - samples/sec: 2329.56 - lr: 0.000020 - momentum: 0.000000
2023-10-14 19:23:37,105 epoch 7 - iter 1805/3617 - loss 0.02286446 - time (sec): 81.90 - samples/sec: 2327.67 - lr: 0.000019 - momentum: 0.000000
2023-10-14 19:23:53,394 epoch 7 - iter 2166/3617 - loss 0.02265517 - time (sec): 98.19 - samples/sec: 2327.00 - lr: 0.000019 - momentum: 0.000000
2023-10-14 19:24:09,594 epoch 7 - iter 2527/3617 - loss 0.02328181 - time (sec): 114.39 - samples/sec: 2325.06 - lr: 0.000018 - momentum: 0.000000
2023-10-14 19:24:25,973 epoch 7 - iter 2888/3617 - loss 0.02329662 - time (sec): 130.77 - samples/sec: 2325.37 - lr: 0.000018 - momentum: 0.000000
2023-10-14 19:24:42,223 epoch 7 - iter 3249/3617 - loss 0.02335070 - time (sec): 147.02 - samples/sec: 2321.86 - lr: 0.000017 - momentum: 0.000000
2023-10-14 19:24:58,552 epoch 7 - iter 3610/3617 - loss 0.02367072 - time (sec): 163.35 - samples/sec: 2321.55 - lr: 0.000017 - momentum: 0.000000
2023-10-14 19:24:58,858 ----------------------------------------------------------------------------------------------------
2023-10-14 19:24:58,858 EPOCH 7 done: loss 0.0237 - lr: 0.000017
2023-10-14 19:25:04,434 DEV : loss 0.2667960822582245 - f1-score (micro avg) 0.6398
2023-10-14 19:25:04,468 saving best model
2023-10-14 19:25:05,056 ----------------------------------------------------------------------------------------------------
2023-10-14 19:25:21,333 epoch 8 - iter 361/3617 - loss 0.01420959 - time (sec): 16.27 - samples/sec: 2271.05 - lr: 0.000016 - momentum: 0.000000
2023-10-14 19:25:37,646 epoch 8 - iter 722/3617 - loss 0.01438499 - time (sec): 32.59 - samples/sec: 2300.97 - lr: 0.000016 - momentum: 0.000000
2023-10-14 19:25:54,010 epoch 8 - iter 1083/3617 - loss 0.01612726 - time (sec): 48.95 - samples/sec: 2324.11 - lr: 0.000015 - momentum: 0.000000
2023-10-14 19:26:10,324 epoch 8 - iter 1444/3617 - loss 0.01522163 - time (sec): 65.26 - samples/sec: 2323.82 - lr: 0.000014 - momentum: 0.000000
2023-10-14 19:26:26,522 epoch 8 - iter 1805/3617 - loss 0.01523890 - time (sec): 81.46 - samples/sec: 2319.15 - lr: 0.000014 - momentum: 0.000000
2023-10-14 19:26:42,609 epoch 8 - iter 2166/3617 - loss 0.01636181 - time (sec): 97.55 - samples/sec: 2301.16 - lr: 0.000013 - momentum: 0.000000
2023-10-14 19:26:59,146 epoch 8 - iter 2527/3617 - loss 0.01647260 - time (sec): 114.09 - samples/sec: 2313.61 - lr: 0.000013 - momentum: 0.000000
2023-10-14 19:27:15,466 epoch 8 - iter 2888/3617 - loss 0.01588100 - time (sec): 130.41 - samples/sec: 2311.13 - lr: 0.000012 - momentum: 0.000000
2023-10-14 19:27:31,824 epoch 8 - iter 3249/3617 - loss 0.01566046 - time (sec): 146.76 - samples/sec: 2319.16 - lr: 0.000012 - momentum: 0.000000
2023-10-14 19:27:48,117 epoch 8 - iter 3610/3617 - loss 0.01532074 - time (sec): 163.06 - samples/sec: 2325.05 - lr: 0.000011 - momentum: 0.000000
2023-10-14 19:27:48,435 ----------------------------------------------------------------------------------------------------
2023-10-14 19:27:48,435 EPOCH 8 done: loss 0.0153 - lr: 0.000011
2023-10-14 19:27:54,685 DEV : loss 0.3515782952308655 - f1-score (micro avg) 0.6337
2023-10-14 19:27:54,715 ----------------------------------------------------------------------------------------------------
2023-10-14 19:28:11,336 epoch 9 - iter 361/3617 - loss 0.01078017 - time (sec): 16.62 - samples/sec: 2283.55 - lr: 0.000011 - momentum: 0.000000
2023-10-14 19:28:27,693 epoch 9 - iter 722/3617 - loss 0.01382949 - time (sec): 32.98 - samples/sec: 2335.85 - lr: 0.000010 - momentum: 0.000000
2023-10-14 19:28:43,933 epoch 9 - iter 1083/3617 - loss 0.01233271 - time (sec): 49.22 - samples/sec: 2346.52 - lr: 0.000009 - momentum: 0.000000
2023-10-14 19:29:00,350 epoch 9 - iter 1444/3617 - loss 0.01223895 - time (sec): 65.63 - samples/sec: 2317.17 - lr: 0.000009 - momentum: 0.000000
2023-10-14 19:29:16,569 epoch 9 - iter 1805/3617 - loss 0.01153695 - time (sec): 81.85 - samples/sec: 2317.35 - lr: 0.000008 - momentum: 0.000000
2023-10-14 19:29:32,978 epoch 9 - iter 2166/3617 - loss 0.01075403 - time (sec): 98.26 - samples/sec: 2311.68 - lr: 0.000008 - momentum: 0.000000
2023-10-14 19:29:49,302 epoch 9 - iter 2527/3617 - loss 0.01022522 - time (sec): 114.59 - samples/sec: 2302.94 - lr: 0.000007 - momentum: 0.000000
2023-10-14 19:30:05,794 epoch 9 - iter 2888/3617 - loss 0.01001724 - time (sec): 131.08 - samples/sec: 2302.46 - lr: 0.000007 - momentum: 0.000000
2023-10-14 19:30:22,232 epoch 9 - iter 3249/3617 - loss 0.01004538 - time (sec): 147.52 - samples/sec: 2308.06 - lr: 0.000006 - momentum: 0.000000
2023-10-14 19:30:38,601 epoch 9 - iter 3610/3617 - loss 0.00979458 - time (sec): 163.88 - samples/sec: 2314.28 - lr: 0.000006 - momentum: 0.000000
2023-10-14 19:30:38,911 ----------------------------------------------------------------------------------------------------
2023-10-14 19:30:38,911 EPOCH 9 done: loss 0.0098 - lr: 0.000006
2023-10-14 19:30:45,203 DEV : loss 0.365791916847229 - f1-score (micro avg) 0.6403
2023-10-14 19:30:45,260 saving best model
2023-10-14 19:30:45,882 ----------------------------------------------------------------------------------------------------
2023-10-14 19:31:02,260 epoch 10 - iter 361/3617 - loss 0.00465788 - time (sec): 16.38 - samples/sec: 2343.20 - lr: 0.000005 - momentum: 0.000000
2023-10-14 19:31:18,604 epoch 10 - iter 722/3617 - loss 0.00449815 - time (sec): 32.72 - samples/sec: 2361.91 - lr: 0.000004 - momentum: 0.000000
2023-10-14 19:31:34,930 epoch 10 - iter 1083/3617 - loss 0.00507346 - time (sec): 49.05 - samples/sec: 2330.64 - lr: 0.000004 - momentum: 0.000000
2023-10-14 19:31:51,267 epoch 10 - iter 1444/3617 - loss 0.00557339 - time (sec): 65.38 - samples/sec: 2328.26 - lr: 0.000003 - momentum: 0.000000
2023-10-14 19:32:07,730 epoch 10 - iter 1805/3617 - loss 0.00531563 - time (sec): 81.85 - samples/sec: 2328.59 - lr: 0.000003 - momentum: 0.000000
2023-10-14 19:32:24,093 epoch 10 - iter 2166/3617 - loss 0.00641317 - time (sec): 98.21 - samples/sec: 2330.04 - lr: 0.000002 - momentum: 0.000000
2023-10-14 19:32:40,290 epoch 10 - iter 2527/3617 - loss 0.00595602 - time (sec): 114.41 - samples/sec: 2318.77 - lr: 0.000002 - momentum: 0.000000
2023-10-14 19:32:56,448 epoch 10 - iter 2888/3617 - loss 0.00598682 - time (sec): 130.56 - samples/sec: 2304.67 - lr: 0.000001 - momentum: 0.000000
2023-10-14 19:33:12,731 epoch 10 - iter 3249/3617 - loss 0.00596379 - time (sec): 146.85 - samples/sec: 2312.77 - lr: 0.000001 - momentum: 0.000000
2023-10-14 19:33:29,112 epoch 10 - iter 3610/3617 - loss 0.00609464 - time (sec): 163.23 - samples/sec: 2323.98 - lr: 0.000000 - momentum: 0.000000
2023-10-14 19:33:29,412 ----------------------------------------------------------------------------------------------------
2023-10-14 19:33:29,412 EPOCH 10 done: loss 0.0061 - lr: 0.000000
2023-10-14 19:33:35,636 DEV : loss 0.3828698396682739 - f1-score (micro avg) 0.6321
2023-10-14 19:33:36,042 ----------------------------------------------------------------------------------------------------
2023-10-14 19:33:36,043 Loading model from best epoch ...
2023-10-14 19:33:37,423 SequenceTagger predicts: Dictionary with 13 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
2023-10-14 19:33:44,142
Results:
- F-score (micro) 0.6452
- F-score (macro) 0.4857
- Accuracy 0.4899
By class:
precision recall f1-score support
loc 0.6283 0.8037 0.7053 591
pers 0.5526 0.7507 0.6366 357
org 0.1333 0.1013 0.1151 79
micro avg 0.5772 0.7313 0.6452 1027
macro avg 0.4381 0.5519 0.4857 1027
weighted avg 0.5639 0.7313 0.6360 1027
2023-10-14 19:33:44,142 ----------------------------------------------------------------------------------------------------
|