File size: 24,043 Bytes
4f292fe |
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 15:22:40,778 ----------------------------------------------------------------------------------------------------
2023-10-17 15:22:40,779 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=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 15:22:40,779 ----------------------------------------------------------------------------------------------------
2023-10-17 15:22:40,779 MultiCorpus: 7142 train + 698 dev + 2570 test sentences
- NER_HIPE_2022 Corpus: 7142 train + 698 dev + 2570 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fr/with_doc_seperator
2023-10-17 15:22:40,779 ----------------------------------------------------------------------------------------------------
2023-10-17 15:22:40,779 Train: 7142 sentences
2023-10-17 15:22:40,779 (train_with_dev=False, train_with_test=False)
2023-10-17 15:22:40,779 ----------------------------------------------------------------------------------------------------
2023-10-17 15:22:40,780 Training Params:
2023-10-17 15:22:40,780 - learning_rate: "5e-05"
2023-10-17 15:22:40,780 - mini_batch_size: "8"
2023-10-17 15:22:40,780 - max_epochs: "10"
2023-10-17 15:22:40,780 - shuffle: "True"
2023-10-17 15:22:40,780 ----------------------------------------------------------------------------------------------------
2023-10-17 15:22:40,780 Plugins:
2023-10-17 15:22:40,780 - TensorboardLogger
2023-10-17 15:22:40,780 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 15:22:40,780 ----------------------------------------------------------------------------------------------------
2023-10-17 15:22:40,780 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 15:22:40,780 - metric: "('micro avg', 'f1-score')"
2023-10-17 15:22:40,780 ----------------------------------------------------------------------------------------------------
2023-10-17 15:22:40,780 Computation:
2023-10-17 15:22:40,780 - compute on device: cuda:0
2023-10-17 15:22:40,780 - embedding storage: none
2023-10-17 15:22:40,780 ----------------------------------------------------------------------------------------------------
2023-10-17 15:22:40,780 Model training base path: "hmbench-newseye/fr-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-17 15:22:40,780 ----------------------------------------------------------------------------------------------------
2023-10-17 15:22:40,780 ----------------------------------------------------------------------------------------------------
2023-10-17 15:22:40,780 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 15:22:48,008 epoch 1 - iter 89/893 - loss 2.84068290 - time (sec): 7.23 - samples/sec: 3501.73 - lr: 0.000005 - momentum: 0.000000
2023-10-17 15:22:55,293 epoch 1 - iter 178/893 - loss 1.76508006 - time (sec): 14.51 - samples/sec: 3476.96 - lr: 0.000010 - momentum: 0.000000
2023-10-17 15:23:02,323 epoch 1 - iter 267/893 - loss 1.32887323 - time (sec): 21.54 - samples/sec: 3444.56 - lr: 0.000015 - momentum: 0.000000
2023-10-17 15:23:09,344 epoch 1 - iter 356/893 - loss 1.07961683 - time (sec): 28.56 - samples/sec: 3429.95 - lr: 0.000020 - momentum: 0.000000
2023-10-17 15:23:15,876 epoch 1 - iter 445/893 - loss 0.91537277 - time (sec): 35.09 - samples/sec: 3450.08 - lr: 0.000025 - momentum: 0.000000
2023-10-17 15:23:22,550 epoch 1 - iter 534/893 - loss 0.79441400 - time (sec): 41.77 - samples/sec: 3488.61 - lr: 0.000030 - momentum: 0.000000
2023-10-17 15:23:29,582 epoch 1 - iter 623/893 - loss 0.70440578 - time (sec): 48.80 - samples/sec: 3492.34 - lr: 0.000035 - momentum: 0.000000
2023-10-17 15:23:37,058 epoch 1 - iter 712/893 - loss 0.62842482 - time (sec): 56.28 - samples/sec: 3498.94 - lr: 0.000040 - momentum: 0.000000
2023-10-17 15:23:44,694 epoch 1 - iter 801/893 - loss 0.57176466 - time (sec): 63.91 - samples/sec: 3491.49 - lr: 0.000045 - momentum: 0.000000
2023-10-17 15:23:52,068 epoch 1 - iter 890/893 - loss 0.52978902 - time (sec): 71.29 - samples/sec: 3481.64 - lr: 0.000050 - momentum: 0.000000
2023-10-17 15:23:52,243 ----------------------------------------------------------------------------------------------------
2023-10-17 15:23:52,243 EPOCH 1 done: loss 0.5288 - lr: 0.000050
2023-10-17 15:23:55,932 DEV : loss 0.11837812513113022 - f1-score (micro avg) 0.6897
2023-10-17 15:23:55,949 saving best model
2023-10-17 15:23:56,359 ----------------------------------------------------------------------------------------------------
2023-10-17 15:24:03,910 epoch 2 - iter 89/893 - loss 0.11416371 - time (sec): 7.55 - samples/sec: 3668.50 - lr: 0.000049 - momentum: 0.000000
2023-10-17 15:24:10,734 epoch 2 - iter 178/893 - loss 0.11054462 - time (sec): 14.37 - samples/sec: 3596.28 - lr: 0.000049 - momentum: 0.000000
2023-10-17 15:24:17,679 epoch 2 - iter 267/893 - loss 0.10944617 - time (sec): 21.32 - samples/sec: 3587.36 - lr: 0.000048 - momentum: 0.000000
2023-10-17 15:24:24,969 epoch 2 - iter 356/893 - loss 0.10974003 - time (sec): 28.61 - samples/sec: 3510.45 - lr: 0.000048 - momentum: 0.000000
2023-10-17 15:24:31,656 epoch 2 - iter 445/893 - loss 0.12542965 - time (sec): 35.29 - samples/sec: 3506.52 - lr: 0.000047 - momentum: 0.000000
2023-10-17 15:24:38,799 epoch 2 - iter 534/893 - loss 0.12518621 - time (sec): 42.44 - samples/sec: 3480.14 - lr: 0.000047 - momentum: 0.000000
2023-10-17 15:24:46,264 epoch 2 - iter 623/893 - loss 0.12341348 - time (sec): 49.90 - samples/sec: 3444.96 - lr: 0.000046 - momentum: 0.000000
2023-10-17 15:24:54,011 epoch 2 - iter 712/893 - loss 0.12016815 - time (sec): 57.65 - samples/sec: 3435.92 - lr: 0.000046 - momentum: 0.000000
2023-10-17 15:25:01,036 epoch 2 - iter 801/893 - loss 0.11705570 - time (sec): 64.67 - samples/sec: 3438.85 - lr: 0.000045 - momentum: 0.000000
2023-10-17 15:25:08,460 epoch 2 - iter 890/893 - loss 0.11562128 - time (sec): 72.10 - samples/sec: 3443.42 - lr: 0.000044 - momentum: 0.000000
2023-10-17 15:25:08,631 ----------------------------------------------------------------------------------------------------
2023-10-17 15:25:08,632 EPOCH 2 done: loss 0.1156 - lr: 0.000044
2023-10-17 15:25:14,003 DEV : loss 0.09630837291479111 - f1-score (micro avg) 0.7789
2023-10-17 15:25:14,021 saving best model
2023-10-17 15:25:14,500 ----------------------------------------------------------------------------------------------------
2023-10-17 15:25:21,435 epoch 3 - iter 89/893 - loss 0.07621002 - time (sec): 6.93 - samples/sec: 3510.78 - lr: 0.000044 - momentum: 0.000000
2023-10-17 15:25:28,317 epoch 3 - iter 178/893 - loss 0.07028836 - time (sec): 13.81 - samples/sec: 3606.54 - lr: 0.000043 - momentum: 0.000000
2023-10-17 15:25:35,746 epoch 3 - iter 267/893 - loss 0.06895379 - time (sec): 21.24 - samples/sec: 3639.95 - lr: 0.000043 - momentum: 0.000000
2023-10-17 15:25:42,631 epoch 3 - iter 356/893 - loss 0.06871513 - time (sec): 28.12 - samples/sec: 3624.74 - lr: 0.000042 - momentum: 0.000000
2023-10-17 15:25:50,043 epoch 3 - iter 445/893 - loss 0.07020606 - time (sec): 35.54 - samples/sec: 3608.17 - lr: 0.000042 - momentum: 0.000000
2023-10-17 15:25:56,658 epoch 3 - iter 534/893 - loss 0.07084888 - time (sec): 42.15 - samples/sec: 3570.61 - lr: 0.000041 - momentum: 0.000000
2023-10-17 15:26:03,774 epoch 3 - iter 623/893 - loss 0.07076245 - time (sec): 49.27 - samples/sec: 3525.50 - lr: 0.000041 - momentum: 0.000000
2023-10-17 15:26:11,199 epoch 3 - iter 712/893 - loss 0.07034170 - time (sec): 56.69 - samples/sec: 3514.66 - lr: 0.000040 - momentum: 0.000000
2023-10-17 15:26:19,031 epoch 3 - iter 801/893 - loss 0.06997454 - time (sec): 64.52 - samples/sec: 3487.38 - lr: 0.000039 - momentum: 0.000000
2023-10-17 15:26:25,774 epoch 3 - iter 890/893 - loss 0.06927984 - time (sec): 71.27 - samples/sec: 3479.19 - lr: 0.000039 - momentum: 0.000000
2023-10-17 15:26:26,014 ----------------------------------------------------------------------------------------------------
2023-10-17 15:26:26,015 EPOCH 3 done: loss 0.0693 - lr: 0.000039
2023-10-17 15:26:30,286 DEV : loss 0.12525002658367157 - f1-score (micro avg) 0.7809
2023-10-17 15:26:30,306 saving best model
2023-10-17 15:26:30,824 ----------------------------------------------------------------------------------------------------
2023-10-17 15:26:38,462 epoch 4 - iter 89/893 - loss 0.05510630 - time (sec): 7.63 - samples/sec: 3358.06 - lr: 0.000038 - momentum: 0.000000
2023-10-17 15:26:45,580 epoch 4 - iter 178/893 - loss 0.04948991 - time (sec): 14.75 - samples/sec: 3444.13 - lr: 0.000038 - momentum: 0.000000
2023-10-17 15:26:52,737 epoch 4 - iter 267/893 - loss 0.04962954 - time (sec): 21.91 - samples/sec: 3459.17 - lr: 0.000037 - momentum: 0.000000
2023-10-17 15:26:59,532 epoch 4 - iter 356/893 - loss 0.04790656 - time (sec): 28.70 - samples/sec: 3465.08 - lr: 0.000037 - momentum: 0.000000
2023-10-17 15:27:06,747 epoch 4 - iter 445/893 - loss 0.04939163 - time (sec): 35.92 - samples/sec: 3439.54 - lr: 0.000036 - momentum: 0.000000
2023-10-17 15:27:13,550 epoch 4 - iter 534/893 - loss 0.04991515 - time (sec): 42.72 - samples/sec: 3455.63 - lr: 0.000036 - momentum: 0.000000
2023-10-17 15:27:20,682 epoch 4 - iter 623/893 - loss 0.04845630 - time (sec): 49.85 - samples/sec: 3476.52 - lr: 0.000035 - momentum: 0.000000
2023-10-17 15:27:27,805 epoch 4 - iter 712/893 - loss 0.04875852 - time (sec): 56.98 - samples/sec: 3480.22 - lr: 0.000034 - momentum: 0.000000
2023-10-17 15:27:35,059 epoch 4 - iter 801/893 - loss 0.04852640 - time (sec): 64.23 - samples/sec: 3481.36 - lr: 0.000034 - momentum: 0.000000
2023-10-17 15:27:42,111 epoch 4 - iter 890/893 - loss 0.04900835 - time (sec): 71.28 - samples/sec: 3476.67 - lr: 0.000033 - momentum: 0.000000
2023-10-17 15:27:42,383 ----------------------------------------------------------------------------------------------------
2023-10-17 15:27:42,383 EPOCH 4 done: loss 0.0490 - lr: 0.000033
2023-10-17 15:27:47,452 DEV : loss 0.14032277464866638 - f1-score (micro avg) 0.7902
2023-10-17 15:27:47,469 saving best model
2023-10-17 15:27:48,074 ----------------------------------------------------------------------------------------------------
2023-10-17 15:27:55,354 epoch 5 - iter 89/893 - loss 0.03298580 - time (sec): 7.28 - samples/sec: 3427.76 - lr: 0.000033 - momentum: 0.000000
2023-10-17 15:28:02,257 epoch 5 - iter 178/893 - loss 0.03246417 - time (sec): 14.18 - samples/sec: 3456.52 - lr: 0.000032 - momentum: 0.000000
2023-10-17 15:28:09,321 epoch 5 - iter 267/893 - loss 0.03531081 - time (sec): 21.25 - samples/sec: 3460.53 - lr: 0.000032 - momentum: 0.000000
2023-10-17 15:28:15,814 epoch 5 - iter 356/893 - loss 0.03662109 - time (sec): 27.74 - samples/sec: 3493.64 - lr: 0.000031 - momentum: 0.000000
2023-10-17 15:28:22,989 epoch 5 - iter 445/893 - loss 0.03546838 - time (sec): 34.91 - samples/sec: 3475.11 - lr: 0.000031 - momentum: 0.000000
2023-10-17 15:28:30,063 epoch 5 - iter 534/893 - loss 0.03683038 - time (sec): 41.99 - samples/sec: 3489.91 - lr: 0.000030 - momentum: 0.000000
2023-10-17 15:28:37,101 epoch 5 - iter 623/893 - loss 0.03593909 - time (sec): 49.03 - samples/sec: 3508.58 - lr: 0.000029 - momentum: 0.000000
2023-10-17 15:28:44,180 epoch 5 - iter 712/893 - loss 0.03641271 - time (sec): 56.10 - samples/sec: 3514.28 - lr: 0.000029 - momentum: 0.000000
2023-10-17 15:28:51,506 epoch 5 - iter 801/893 - loss 0.03619958 - time (sec): 63.43 - samples/sec: 3519.41 - lr: 0.000028 - momentum: 0.000000
2023-10-17 15:28:58,451 epoch 5 - iter 890/893 - loss 0.03553096 - time (sec): 70.38 - samples/sec: 3526.38 - lr: 0.000028 - momentum: 0.000000
2023-10-17 15:28:58,615 ----------------------------------------------------------------------------------------------------
2023-10-17 15:28:58,615 EPOCH 5 done: loss 0.0356 - lr: 0.000028
2023-10-17 15:29:03,077 DEV : loss 0.15836651623249054 - f1-score (micro avg) 0.796
2023-10-17 15:29:03,100 saving best model
2023-10-17 15:29:04,345 ----------------------------------------------------------------------------------------------------
2023-10-17 15:29:11,305 epoch 6 - iter 89/893 - loss 0.01800969 - time (sec): 6.96 - samples/sec: 3567.96 - lr: 0.000027 - momentum: 0.000000
2023-10-17 15:29:17,844 epoch 6 - iter 178/893 - loss 0.02091387 - time (sec): 13.50 - samples/sec: 3542.18 - lr: 0.000027 - momentum: 0.000000
2023-10-17 15:29:25,110 epoch 6 - iter 267/893 - loss 0.02236790 - time (sec): 20.76 - samples/sec: 3520.46 - lr: 0.000026 - momentum: 0.000000
2023-10-17 15:29:32,572 epoch 6 - iter 356/893 - loss 0.02351990 - time (sec): 28.23 - samples/sec: 3490.62 - lr: 0.000026 - momentum: 0.000000
2023-10-17 15:29:39,486 epoch 6 - iter 445/893 - loss 0.02468375 - time (sec): 35.14 - samples/sec: 3507.96 - lr: 0.000025 - momentum: 0.000000
2023-10-17 15:29:46,768 epoch 6 - iter 534/893 - loss 0.02606959 - time (sec): 42.42 - samples/sec: 3524.54 - lr: 0.000024 - momentum: 0.000000
2023-10-17 15:29:54,075 epoch 6 - iter 623/893 - loss 0.02639857 - time (sec): 49.73 - samples/sec: 3513.88 - lr: 0.000024 - momentum: 0.000000
2023-10-17 15:30:00,956 epoch 6 - iter 712/893 - loss 0.02599939 - time (sec): 56.61 - samples/sec: 3524.55 - lr: 0.000023 - momentum: 0.000000
2023-10-17 15:30:07,994 epoch 6 - iter 801/893 - loss 0.02730520 - time (sec): 63.65 - samples/sec: 3523.50 - lr: 0.000023 - momentum: 0.000000
2023-10-17 15:30:14,951 epoch 6 - iter 890/893 - loss 0.02699016 - time (sec): 70.60 - samples/sec: 3512.72 - lr: 0.000022 - momentum: 0.000000
2023-10-17 15:30:15,147 ----------------------------------------------------------------------------------------------------
2023-10-17 15:30:15,147 EPOCH 6 done: loss 0.0269 - lr: 0.000022
2023-10-17 15:30:19,463 DEV : loss 0.18827223777770996 - f1-score (micro avg) 0.8073
2023-10-17 15:30:19,483 saving best model
2023-10-17 15:30:20,021 ----------------------------------------------------------------------------------------------------
2023-10-17 15:30:27,186 epoch 7 - iter 89/893 - loss 0.01580982 - time (sec): 7.16 - samples/sec: 3649.38 - lr: 0.000022 - momentum: 0.000000
2023-10-17 15:30:34,189 epoch 7 - iter 178/893 - loss 0.01835420 - time (sec): 14.17 - samples/sec: 3574.06 - lr: 0.000021 - momentum: 0.000000
2023-10-17 15:30:41,197 epoch 7 - iter 267/893 - loss 0.01852637 - time (sec): 21.17 - samples/sec: 3502.01 - lr: 0.000021 - momentum: 0.000000
2023-10-17 15:30:48,198 epoch 7 - iter 356/893 - loss 0.01824809 - time (sec): 28.17 - samples/sec: 3531.46 - lr: 0.000020 - momentum: 0.000000
2023-10-17 15:30:55,196 epoch 7 - iter 445/893 - loss 0.01887441 - time (sec): 35.17 - samples/sec: 3536.41 - lr: 0.000019 - momentum: 0.000000
2023-10-17 15:31:01,894 epoch 7 - iter 534/893 - loss 0.01977477 - time (sec): 41.87 - samples/sec: 3560.21 - lr: 0.000019 - momentum: 0.000000
2023-10-17 15:31:08,667 epoch 7 - iter 623/893 - loss 0.01999407 - time (sec): 48.64 - samples/sec: 3563.02 - lr: 0.000018 - momentum: 0.000000
2023-10-17 15:31:15,718 epoch 7 - iter 712/893 - loss 0.02000396 - time (sec): 55.69 - samples/sec: 3540.87 - lr: 0.000018 - momentum: 0.000000
2023-10-17 15:31:23,232 epoch 7 - iter 801/893 - loss 0.02042434 - time (sec): 63.21 - samples/sec: 3529.38 - lr: 0.000017 - momentum: 0.000000
2023-10-17 15:31:30,175 epoch 7 - iter 890/893 - loss 0.01995623 - time (sec): 70.15 - samples/sec: 3538.79 - lr: 0.000017 - momentum: 0.000000
2023-10-17 15:31:30,393 ----------------------------------------------------------------------------------------------------
2023-10-17 15:31:30,393 EPOCH 7 done: loss 0.0200 - lr: 0.000017
2023-10-17 15:31:35,286 DEV : loss 0.20268814265727997 - f1-score (micro avg) 0.8297
2023-10-17 15:31:35,303 saving best model
2023-10-17 15:31:35,839 ----------------------------------------------------------------------------------------------------
2023-10-17 15:31:42,909 epoch 8 - iter 89/893 - loss 0.01596133 - time (sec): 7.07 - samples/sec: 3402.93 - lr: 0.000016 - momentum: 0.000000
2023-10-17 15:31:49,701 epoch 8 - iter 178/893 - loss 0.01480964 - time (sec): 13.86 - samples/sec: 3462.00 - lr: 0.000016 - momentum: 0.000000
2023-10-17 15:31:57,025 epoch 8 - iter 267/893 - loss 0.01675137 - time (sec): 21.18 - samples/sec: 3458.33 - lr: 0.000015 - momentum: 0.000000
2023-10-17 15:32:04,151 epoch 8 - iter 356/893 - loss 0.01522643 - time (sec): 28.31 - samples/sec: 3507.66 - lr: 0.000014 - momentum: 0.000000
2023-10-17 15:32:11,825 epoch 8 - iter 445/893 - loss 0.01508987 - time (sec): 35.98 - samples/sec: 3519.28 - lr: 0.000014 - momentum: 0.000000
2023-10-17 15:32:19,048 epoch 8 - iter 534/893 - loss 0.01436513 - time (sec): 43.21 - samples/sec: 3523.03 - lr: 0.000013 - momentum: 0.000000
2023-10-17 15:32:26,166 epoch 8 - iter 623/893 - loss 0.01425423 - time (sec): 50.32 - samples/sec: 3515.51 - lr: 0.000013 - momentum: 0.000000
2023-10-17 15:32:33,130 epoch 8 - iter 712/893 - loss 0.01422439 - time (sec): 57.29 - samples/sec: 3501.42 - lr: 0.000012 - momentum: 0.000000
2023-10-17 15:32:40,051 epoch 8 - iter 801/893 - loss 0.01401622 - time (sec): 64.21 - samples/sec: 3498.11 - lr: 0.000012 - momentum: 0.000000
2023-10-17 15:32:46,708 epoch 8 - iter 890/893 - loss 0.01447625 - time (sec): 70.87 - samples/sec: 3496.21 - lr: 0.000011 - momentum: 0.000000
2023-10-17 15:32:46,993 ----------------------------------------------------------------------------------------------------
2023-10-17 15:32:46,993 EPOCH 8 done: loss 0.0145 - lr: 0.000011
2023-10-17 15:32:51,297 DEV : loss 0.18800464272499084 - f1-score (micro avg) 0.8212
2023-10-17 15:32:51,315 ----------------------------------------------------------------------------------------------------
2023-10-17 15:32:58,164 epoch 9 - iter 89/893 - loss 0.01188764 - time (sec): 6.85 - samples/sec: 3518.48 - lr: 0.000011 - momentum: 0.000000
2023-10-17 15:33:04,766 epoch 9 - iter 178/893 - loss 0.01149941 - time (sec): 13.45 - samples/sec: 3575.04 - lr: 0.000010 - momentum: 0.000000
2023-10-17 15:33:11,658 epoch 9 - iter 267/893 - loss 0.01223126 - time (sec): 20.34 - samples/sec: 3572.33 - lr: 0.000009 - momentum: 0.000000
2023-10-17 15:33:18,377 epoch 9 - iter 356/893 - loss 0.01060371 - time (sec): 27.06 - samples/sec: 3578.76 - lr: 0.000009 - momentum: 0.000000
2023-10-17 15:33:26,375 epoch 9 - iter 445/893 - loss 0.01014695 - time (sec): 35.06 - samples/sec: 3508.51 - lr: 0.000008 - momentum: 0.000000
2023-10-17 15:33:33,290 epoch 9 - iter 534/893 - loss 0.01151378 - time (sec): 41.97 - samples/sec: 3544.73 - lr: 0.000008 - momentum: 0.000000
2023-10-17 15:33:40,363 epoch 9 - iter 623/893 - loss 0.01164043 - time (sec): 49.05 - samples/sec: 3520.41 - lr: 0.000007 - momentum: 0.000000
2023-10-17 15:33:47,206 epoch 9 - iter 712/893 - loss 0.01090414 - time (sec): 55.89 - samples/sec: 3536.84 - lr: 0.000007 - momentum: 0.000000
2023-10-17 15:33:54,230 epoch 9 - iter 801/893 - loss 0.01064507 - time (sec): 62.91 - samples/sec: 3534.84 - lr: 0.000006 - momentum: 0.000000
2023-10-17 15:34:01,797 epoch 9 - iter 890/893 - loss 0.01012024 - time (sec): 70.48 - samples/sec: 3516.33 - lr: 0.000006 - momentum: 0.000000
2023-10-17 15:34:02,023 ----------------------------------------------------------------------------------------------------
2023-10-17 15:34:02,023 EPOCH 9 done: loss 0.0101 - lr: 0.000006
2023-10-17 15:34:06,256 DEV : loss 0.20825740694999695 - f1-score (micro avg) 0.8223
2023-10-17 15:34:06,274 ----------------------------------------------------------------------------------------------------
2023-10-17 15:34:13,338 epoch 10 - iter 89/893 - loss 0.01106391 - time (sec): 7.06 - samples/sec: 3428.06 - lr: 0.000005 - momentum: 0.000000
2023-10-17 15:34:19,986 epoch 10 - iter 178/893 - loss 0.00869302 - time (sec): 13.71 - samples/sec: 3528.09 - lr: 0.000004 - momentum: 0.000000
2023-10-17 15:34:27,221 epoch 10 - iter 267/893 - loss 0.00786190 - time (sec): 20.95 - samples/sec: 3535.88 - lr: 0.000004 - momentum: 0.000000
2023-10-17 15:34:33,970 epoch 10 - iter 356/893 - loss 0.00840030 - time (sec): 27.69 - samples/sec: 3499.03 - lr: 0.000003 - momentum: 0.000000
2023-10-17 15:34:40,742 epoch 10 - iter 445/893 - loss 0.00768042 - time (sec): 34.47 - samples/sec: 3487.42 - lr: 0.000003 - momentum: 0.000000
2023-10-17 15:34:48,133 epoch 10 - iter 534/893 - loss 0.00760238 - time (sec): 41.86 - samples/sec: 3472.82 - lr: 0.000002 - momentum: 0.000000
2023-10-17 15:34:55,488 epoch 10 - iter 623/893 - loss 0.00744283 - time (sec): 49.21 - samples/sec: 3468.31 - lr: 0.000002 - momentum: 0.000000
2023-10-17 15:35:02,728 epoch 10 - iter 712/893 - loss 0.00710376 - time (sec): 56.45 - samples/sec: 3457.79 - lr: 0.000001 - momentum: 0.000000
2023-10-17 15:35:09,784 epoch 10 - iter 801/893 - loss 0.00664551 - time (sec): 63.51 - samples/sec: 3472.68 - lr: 0.000001 - momentum: 0.000000
2023-10-17 15:35:17,319 epoch 10 - iter 890/893 - loss 0.00642147 - time (sec): 71.04 - samples/sec: 3490.94 - lr: 0.000000 - momentum: 0.000000
2023-10-17 15:35:17,555 ----------------------------------------------------------------------------------------------------
2023-10-17 15:35:17,555 EPOCH 10 done: loss 0.0064 - lr: 0.000000
2023-10-17 15:35:22,442 DEV : loss 0.20835766196250916 - f1-score (micro avg) 0.8245
2023-10-17 15:35:22,857 ----------------------------------------------------------------------------------------------------
2023-10-17 15:35:22,859 Loading model from best epoch ...
2023-10-17 15:35:24,412 SequenceTagger predicts: Dictionary with 17 tags: O, S-PER, B-PER, E-PER, I-PER, S-LOC, B-LOC, E-LOC, I-LOC, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-17 15:35:34,665
Results:
- F-score (micro) 0.721
- F-score (macro) 0.6426
- Accuracy 0.5799
By class:
precision recall f1-score support
LOC 0.7364 0.7397 0.7380 1095
PER 0.7921 0.7717 0.7818 1012
ORG 0.5012 0.5826 0.5389 357
HumanProd 0.4151 0.6667 0.5116 33
micro avg 0.7130 0.7293 0.7210 2497
macro avg 0.6112 0.6902 0.6426 2497
weighted avg 0.7211 0.7293 0.7243 2497
2023-10-17 15:35:34,665 ----------------------------------------------------------------------------------------------------
|