File size: 23,900 Bytes
09058d4 |
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 |
2023-10-17 10:32:40,008 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:40,009 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 10:32:40,009 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:40,009 MultiCorpus: 966 train + 219 dev + 204 test sentences
- NER_HIPE_2022 Corpus: 966 train + 219 dev + 204 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/fr/with_doc_seperator
2023-10-17 10:32:40,009 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:40,009 Train: 966 sentences
2023-10-17 10:32:40,009 (train_with_dev=False, train_with_test=False)
2023-10-17 10:32:40,009 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:40,009 Training Params:
2023-10-17 10:32:40,009 - learning_rate: "5e-05"
2023-10-17 10:32:40,009 - mini_batch_size: "8"
2023-10-17 10:32:40,009 - max_epochs: "10"
2023-10-17 10:32:40,009 - shuffle: "True"
2023-10-17 10:32:40,010 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:40,010 Plugins:
2023-10-17 10:32:40,010 - TensorboardLogger
2023-10-17 10:32:40,010 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 10:32:40,010 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:40,010 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 10:32:40,010 - metric: "('micro avg', 'f1-score')"
2023-10-17 10:32:40,010 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:40,010 Computation:
2023-10-17 10:32:40,010 - compute on device: cuda:0
2023-10-17 10:32:40,010 - embedding storage: none
2023-10-17 10:32:40,010 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:40,010 Model training base path: "hmbench-ajmc/fr-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-17 10:32:40,010 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:40,010 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:40,010 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 10:32:40,704 epoch 1 - iter 12/121 - loss 3.26006474 - time (sec): 0.69 - samples/sec: 3358.62 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:32:41,416 epoch 1 - iter 24/121 - loss 2.80714465 - time (sec): 1.41 - samples/sec: 3179.73 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:32:42,131 epoch 1 - iter 36/121 - loss 2.26357289 - time (sec): 2.12 - samples/sec: 3226.17 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:32:42,895 epoch 1 - iter 48/121 - loss 1.81294219 - time (sec): 2.88 - samples/sec: 3258.12 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:32:43,687 epoch 1 - iter 60/121 - loss 1.52221299 - time (sec): 3.68 - samples/sec: 3268.71 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:32:44,420 epoch 1 - iter 72/121 - loss 1.34077527 - time (sec): 4.41 - samples/sec: 3270.74 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:32:45,172 epoch 1 - iter 84/121 - loss 1.19489547 - time (sec): 5.16 - samples/sec: 3271.74 - lr: 0.000034 - momentum: 0.000000
2023-10-17 10:32:45,955 epoch 1 - iter 96/121 - loss 1.07021795 - time (sec): 5.94 - samples/sec: 3278.00 - lr: 0.000039 - momentum: 0.000000
2023-10-17 10:32:46,679 epoch 1 - iter 108/121 - loss 0.98756273 - time (sec): 6.67 - samples/sec: 3291.74 - lr: 0.000044 - momentum: 0.000000
2023-10-17 10:32:47,457 epoch 1 - iter 120/121 - loss 0.91125100 - time (sec): 7.45 - samples/sec: 3301.42 - lr: 0.000049 - momentum: 0.000000
2023-10-17 10:32:47,527 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:47,527 EPOCH 1 done: loss 0.9066 - lr: 0.000049
2023-10-17 10:32:48,393 DEV : loss 0.28961601853370667 - f1-score (micro avg) 0.5606
2023-10-17 10:32:48,400 saving best model
2023-10-17 10:32:48,825 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:49,560 epoch 2 - iter 12/121 - loss 0.22875945 - time (sec): 0.73 - samples/sec: 3357.71 - lr: 0.000049 - momentum: 0.000000
2023-10-17 10:32:50,258 epoch 2 - iter 24/121 - loss 0.22465711 - time (sec): 1.43 - samples/sec: 3181.20 - lr: 0.000049 - momentum: 0.000000
2023-10-17 10:32:50,958 epoch 2 - iter 36/121 - loss 0.21804974 - time (sec): 2.13 - samples/sec: 3256.77 - lr: 0.000048 - momentum: 0.000000
2023-10-17 10:32:51,716 epoch 2 - iter 48/121 - loss 0.20232516 - time (sec): 2.89 - samples/sec: 3363.68 - lr: 0.000048 - momentum: 0.000000
2023-10-17 10:32:52,516 epoch 2 - iter 60/121 - loss 0.19376496 - time (sec): 3.69 - samples/sec: 3323.26 - lr: 0.000047 - momentum: 0.000000
2023-10-17 10:32:53,255 epoch 2 - iter 72/121 - loss 0.19087101 - time (sec): 4.43 - samples/sec: 3345.29 - lr: 0.000047 - momentum: 0.000000
2023-10-17 10:32:54,009 epoch 2 - iter 84/121 - loss 0.18357053 - time (sec): 5.18 - samples/sec: 3309.54 - lr: 0.000046 - momentum: 0.000000
2023-10-17 10:32:54,817 epoch 2 - iter 96/121 - loss 0.18101522 - time (sec): 5.99 - samples/sec: 3292.67 - lr: 0.000046 - momentum: 0.000000
2023-10-17 10:32:55,656 epoch 2 - iter 108/121 - loss 0.17795186 - time (sec): 6.83 - samples/sec: 3254.49 - lr: 0.000045 - momentum: 0.000000
2023-10-17 10:32:56,429 epoch 2 - iter 120/121 - loss 0.17481566 - time (sec): 7.60 - samples/sec: 3230.58 - lr: 0.000045 - momentum: 0.000000
2023-10-17 10:32:56,482 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:56,482 EPOCH 2 done: loss 0.1739 - lr: 0.000045
2023-10-17 10:32:57,266 DEV : loss 0.13590769469738007 - f1-score (micro avg) 0.7866
2023-10-17 10:32:57,271 saving best model
2023-10-17 10:32:57,790 ----------------------------------------------------------------------------------------------------
2023-10-17 10:32:58,585 epoch 3 - iter 12/121 - loss 0.11327432 - time (sec): 0.79 - samples/sec: 3159.61 - lr: 0.000044 - momentum: 0.000000
2023-10-17 10:32:59,372 epoch 3 - iter 24/121 - loss 0.09727054 - time (sec): 1.58 - samples/sec: 3082.45 - lr: 0.000043 - momentum: 0.000000
2023-10-17 10:33:00,270 epoch 3 - iter 36/121 - loss 0.10169243 - time (sec): 2.48 - samples/sec: 3057.53 - lr: 0.000043 - momentum: 0.000000
2023-10-17 10:33:01,022 epoch 3 - iter 48/121 - loss 0.10278596 - time (sec): 3.23 - samples/sec: 3112.97 - lr: 0.000042 - momentum: 0.000000
2023-10-17 10:33:01,798 epoch 3 - iter 60/121 - loss 0.10669048 - time (sec): 4.00 - samples/sec: 3095.14 - lr: 0.000042 - momentum: 0.000000
2023-10-17 10:33:02,626 epoch 3 - iter 72/121 - loss 0.10457757 - time (sec): 4.83 - samples/sec: 3139.89 - lr: 0.000041 - momentum: 0.000000
2023-10-17 10:33:03,390 epoch 3 - iter 84/121 - loss 0.10430623 - time (sec): 5.60 - samples/sec: 3087.28 - lr: 0.000041 - momentum: 0.000000
2023-10-17 10:33:04,108 epoch 3 - iter 96/121 - loss 0.10325922 - time (sec): 6.31 - samples/sec: 3119.76 - lr: 0.000040 - momentum: 0.000000
2023-10-17 10:33:04,861 epoch 3 - iter 108/121 - loss 0.10206642 - time (sec): 7.07 - samples/sec: 3171.64 - lr: 0.000040 - momentum: 0.000000
2023-10-17 10:33:05,604 epoch 3 - iter 120/121 - loss 0.10270216 - time (sec): 7.81 - samples/sec: 3158.98 - lr: 0.000039 - momentum: 0.000000
2023-10-17 10:33:05,650 ----------------------------------------------------------------------------------------------------
2023-10-17 10:33:05,650 EPOCH 3 done: loss 0.1031 - lr: 0.000039
2023-10-17 10:33:06,430 DEV : loss 0.12963081896305084 - f1-score (micro avg) 0.8244
2023-10-17 10:33:06,436 saving best model
2023-10-17 10:33:06,980 ----------------------------------------------------------------------------------------------------
2023-10-17 10:33:07,716 epoch 4 - iter 12/121 - loss 0.10310305 - time (sec): 0.73 - samples/sec: 3448.31 - lr: 0.000038 - momentum: 0.000000
2023-10-17 10:33:08,423 epoch 4 - iter 24/121 - loss 0.07881906 - time (sec): 1.44 - samples/sec: 3412.90 - lr: 0.000038 - momentum: 0.000000
2023-10-17 10:33:09,156 epoch 4 - iter 36/121 - loss 0.06988981 - time (sec): 2.17 - samples/sec: 3421.12 - lr: 0.000037 - momentum: 0.000000
2023-10-17 10:33:09,918 epoch 4 - iter 48/121 - loss 0.08057507 - time (sec): 2.94 - samples/sec: 3305.58 - lr: 0.000037 - momentum: 0.000000
2023-10-17 10:33:10,641 epoch 4 - iter 60/121 - loss 0.07780496 - time (sec): 3.66 - samples/sec: 3399.05 - lr: 0.000036 - momentum: 0.000000
2023-10-17 10:33:11,431 epoch 4 - iter 72/121 - loss 0.06914986 - time (sec): 4.45 - samples/sec: 3354.70 - lr: 0.000036 - momentum: 0.000000
2023-10-17 10:33:12,226 epoch 4 - iter 84/121 - loss 0.06675876 - time (sec): 5.24 - samples/sec: 3323.77 - lr: 0.000035 - momentum: 0.000000
2023-10-17 10:33:12,958 epoch 4 - iter 96/121 - loss 0.07186866 - time (sec): 5.97 - samples/sec: 3319.67 - lr: 0.000035 - momentum: 0.000000
2023-10-17 10:33:13,732 epoch 4 - iter 108/121 - loss 0.07109931 - time (sec): 6.75 - samples/sec: 3281.32 - lr: 0.000034 - momentum: 0.000000
2023-10-17 10:33:14,537 epoch 4 - iter 120/121 - loss 0.06880122 - time (sec): 7.55 - samples/sec: 3250.67 - lr: 0.000034 - momentum: 0.000000
2023-10-17 10:33:14,587 ----------------------------------------------------------------------------------------------------
2023-10-17 10:33:14,587 EPOCH 4 done: loss 0.0683 - lr: 0.000034
2023-10-17 10:33:15,354 DEV : loss 0.15439237654209137 - f1-score (micro avg) 0.8221
2023-10-17 10:33:15,360 ----------------------------------------------------------------------------------------------------
2023-10-17 10:33:16,180 epoch 5 - iter 12/121 - loss 0.04724636 - time (sec): 0.82 - samples/sec: 3409.82 - lr: 0.000033 - momentum: 0.000000
2023-10-17 10:33:16,873 epoch 5 - iter 24/121 - loss 0.05331324 - time (sec): 1.51 - samples/sec: 3276.74 - lr: 0.000032 - momentum: 0.000000
2023-10-17 10:33:17,648 epoch 5 - iter 36/121 - loss 0.05313566 - time (sec): 2.29 - samples/sec: 3289.81 - lr: 0.000032 - momentum: 0.000000
2023-10-17 10:33:18,419 epoch 5 - iter 48/121 - loss 0.04906792 - time (sec): 3.06 - samples/sec: 3255.55 - lr: 0.000031 - momentum: 0.000000
2023-10-17 10:33:19,121 epoch 5 - iter 60/121 - loss 0.04719752 - time (sec): 3.76 - samples/sec: 3285.53 - lr: 0.000031 - momentum: 0.000000
2023-10-17 10:33:19,881 epoch 5 - iter 72/121 - loss 0.04375019 - time (sec): 4.52 - samples/sec: 3277.57 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:33:20,596 epoch 5 - iter 84/121 - loss 0.04656749 - time (sec): 5.23 - samples/sec: 3337.99 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:33:21,353 epoch 5 - iter 96/121 - loss 0.04663450 - time (sec): 5.99 - samples/sec: 3282.19 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:33:22,094 epoch 5 - iter 108/121 - loss 0.04648098 - time (sec): 6.73 - samples/sec: 3281.60 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:33:22,871 epoch 5 - iter 120/121 - loss 0.04818959 - time (sec): 7.51 - samples/sec: 3269.52 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:33:22,942 ----------------------------------------------------------------------------------------------------
2023-10-17 10:33:22,943 EPOCH 5 done: loss 0.0479 - lr: 0.000028
2023-10-17 10:33:23,707 DEV : loss 0.163357213139534 - f1-score (micro avg) 0.8361
2023-10-17 10:33:23,714 saving best model
2023-10-17 10:33:24,261 ----------------------------------------------------------------------------------------------------
2023-10-17 10:33:25,004 epoch 6 - iter 12/121 - loss 0.01706041 - time (sec): 0.74 - samples/sec: 3576.24 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:33:25,803 epoch 6 - iter 24/121 - loss 0.03413402 - time (sec): 1.54 - samples/sec: 3332.48 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:33:26,578 epoch 6 - iter 36/121 - loss 0.03081119 - time (sec): 2.32 - samples/sec: 3316.18 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:33:27,307 epoch 6 - iter 48/121 - loss 0.03234101 - time (sec): 3.04 - samples/sec: 3260.29 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:33:28,098 epoch 6 - iter 60/121 - loss 0.02974588 - time (sec): 3.84 - samples/sec: 3234.64 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:33:28,849 epoch 6 - iter 72/121 - loss 0.03224484 - time (sec): 4.59 - samples/sec: 3239.87 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:33:29,568 epoch 6 - iter 84/121 - loss 0.02986217 - time (sec): 5.31 - samples/sec: 3253.41 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:33:30,327 epoch 6 - iter 96/121 - loss 0.03273185 - time (sec): 6.06 - samples/sec: 3261.07 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:33:31,029 epoch 6 - iter 108/121 - loss 0.03525937 - time (sec): 6.77 - samples/sec: 3274.06 - lr: 0.000023 - momentum: 0.000000
2023-10-17 10:33:31,759 epoch 6 - iter 120/121 - loss 0.03648931 - time (sec): 7.50 - samples/sec: 3284.01 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:33:31,806 ----------------------------------------------------------------------------------------------------
2023-10-17 10:33:31,806 EPOCH 6 done: loss 0.0365 - lr: 0.000022
2023-10-17 10:33:32,581 DEV : loss 0.17233064770698547 - f1-score (micro avg) 0.821
2023-10-17 10:33:32,588 ----------------------------------------------------------------------------------------------------
2023-10-17 10:33:33,337 epoch 7 - iter 12/121 - loss 0.02404734 - time (sec): 0.75 - samples/sec: 3528.67 - lr: 0.000022 - momentum: 0.000000
2023-10-17 10:33:34,091 epoch 7 - iter 24/121 - loss 0.02779211 - time (sec): 1.50 - samples/sec: 3359.91 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:33:34,846 epoch 7 - iter 36/121 - loss 0.02659868 - time (sec): 2.26 - samples/sec: 3334.23 - lr: 0.000021 - momentum: 0.000000
2023-10-17 10:33:35,569 epoch 7 - iter 48/121 - loss 0.02679515 - time (sec): 2.98 - samples/sec: 3338.31 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:33:36,378 epoch 7 - iter 60/121 - loss 0.02553453 - time (sec): 3.79 - samples/sec: 3341.44 - lr: 0.000020 - momentum: 0.000000
2023-10-17 10:33:37,130 epoch 7 - iter 72/121 - loss 0.02409986 - time (sec): 4.54 - samples/sec: 3328.59 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:33:37,933 epoch 7 - iter 84/121 - loss 0.02334057 - time (sec): 5.34 - samples/sec: 3288.33 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:33:38,655 epoch 7 - iter 96/121 - loss 0.02309239 - time (sec): 6.07 - samples/sec: 3244.95 - lr: 0.000018 - momentum: 0.000000
2023-10-17 10:33:39,405 epoch 7 - iter 108/121 - loss 0.02399850 - time (sec): 6.82 - samples/sec: 3253.36 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:33:40,140 epoch 7 - iter 120/121 - loss 0.02302854 - time (sec): 7.55 - samples/sec: 3250.94 - lr: 0.000017 - momentum: 0.000000
2023-10-17 10:33:40,198 ----------------------------------------------------------------------------------------------------
2023-10-17 10:33:40,198 EPOCH 7 done: loss 0.0230 - lr: 0.000017
2023-10-17 10:33:41,033 DEV : loss 0.2143603265285492 - f1-score (micro avg) 0.825
2023-10-17 10:33:41,038 ----------------------------------------------------------------------------------------------------
2023-10-17 10:33:41,775 epoch 8 - iter 12/121 - loss 0.01577638 - time (sec): 0.74 - samples/sec: 3213.30 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:33:42,491 epoch 8 - iter 24/121 - loss 0.01178881 - time (sec): 1.45 - samples/sec: 3325.68 - lr: 0.000016 - momentum: 0.000000
2023-10-17 10:33:43,313 epoch 8 - iter 36/121 - loss 0.01688569 - time (sec): 2.27 - samples/sec: 3259.36 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:33:44,086 epoch 8 - iter 48/121 - loss 0.01747875 - time (sec): 3.05 - samples/sec: 3291.33 - lr: 0.000015 - momentum: 0.000000
2023-10-17 10:33:44,821 epoch 8 - iter 60/121 - loss 0.01973594 - time (sec): 3.78 - samples/sec: 3319.04 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:33:45,548 epoch 8 - iter 72/121 - loss 0.01860066 - time (sec): 4.51 - samples/sec: 3353.85 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:33:46,289 epoch 8 - iter 84/121 - loss 0.01781938 - time (sec): 5.25 - samples/sec: 3332.89 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:33:47,007 epoch 8 - iter 96/121 - loss 0.02044574 - time (sec): 5.97 - samples/sec: 3285.14 - lr: 0.000013 - momentum: 0.000000
2023-10-17 10:33:47,838 epoch 8 - iter 108/121 - loss 0.01883755 - time (sec): 6.80 - samples/sec: 3304.17 - lr: 0.000012 - momentum: 0.000000
2023-10-17 10:33:48,739 epoch 8 - iter 120/121 - loss 0.01804760 - time (sec): 7.70 - samples/sec: 3194.63 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:33:48,791 ----------------------------------------------------------------------------------------------------
2023-10-17 10:33:48,791 EPOCH 8 done: loss 0.0179 - lr: 0.000011
2023-10-17 10:33:49,588 DEV : loss 0.2186920940876007 - f1-score (micro avg) 0.8375
2023-10-17 10:33:49,593 saving best model
2023-10-17 10:33:50,089 ----------------------------------------------------------------------------------------------------
2023-10-17 10:33:50,818 epoch 9 - iter 12/121 - loss 0.01090955 - time (sec): 0.72 - samples/sec: 3218.18 - lr: 0.000011 - momentum: 0.000000
2023-10-17 10:33:51,572 epoch 9 - iter 24/121 - loss 0.01187864 - time (sec): 1.48 - samples/sec: 3083.15 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:33:52,357 epoch 9 - iter 36/121 - loss 0.01271093 - time (sec): 2.26 - samples/sec: 3149.15 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:33:53,127 epoch 9 - iter 48/121 - loss 0.01137607 - time (sec): 3.03 - samples/sec: 3211.30 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:33:53,911 epoch 9 - iter 60/121 - loss 0.01522958 - time (sec): 3.82 - samples/sec: 3160.21 - lr: 0.000009 - momentum: 0.000000
2023-10-17 10:33:54,709 epoch 9 - iter 72/121 - loss 0.01431421 - time (sec): 4.61 - samples/sec: 3156.82 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:33:55,437 epoch 9 - iter 84/121 - loss 0.01403599 - time (sec): 5.34 - samples/sec: 3161.08 - lr: 0.000008 - momentum: 0.000000
2023-10-17 10:33:56,197 epoch 9 - iter 96/121 - loss 0.01371746 - time (sec): 6.10 - samples/sec: 3194.06 - lr: 0.000007 - momentum: 0.000000
2023-10-17 10:33:56,952 epoch 9 - iter 108/121 - loss 0.01265669 - time (sec): 6.86 - samples/sec: 3215.69 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:33:57,739 epoch 9 - iter 120/121 - loss 0.01203950 - time (sec): 7.64 - samples/sec: 3210.74 - lr: 0.000006 - momentum: 0.000000
2023-10-17 10:33:57,802 ----------------------------------------------------------------------------------------------------
2023-10-17 10:33:57,802 EPOCH 9 done: loss 0.0120 - lr: 0.000006
2023-10-17 10:33:58,619 DEV : loss 0.22531166672706604 - f1-score (micro avg) 0.831
2023-10-17 10:33:58,625 ----------------------------------------------------------------------------------------------------
2023-10-17 10:33:59,359 epoch 10 - iter 12/121 - loss 0.00410764 - time (sec): 0.73 - samples/sec: 3231.86 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:34:00,117 epoch 10 - iter 24/121 - loss 0.00455508 - time (sec): 1.49 - samples/sec: 3295.25 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:34:00,855 epoch 10 - iter 36/121 - loss 0.00428222 - time (sec): 2.23 - samples/sec: 3295.81 - lr: 0.000004 - momentum: 0.000000
2023-10-17 10:34:01,694 epoch 10 - iter 48/121 - loss 0.00361187 - time (sec): 3.07 - samples/sec: 3229.25 - lr: 0.000004 - momentum: 0.000000
2023-10-17 10:34:02,545 epoch 10 - iter 60/121 - loss 0.00549113 - time (sec): 3.92 - samples/sec: 3181.74 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:34:03,295 epoch 10 - iter 72/121 - loss 0.00620171 - time (sec): 4.67 - samples/sec: 3192.91 - lr: 0.000003 - momentum: 0.000000
2023-10-17 10:34:04,040 epoch 10 - iter 84/121 - loss 0.00747202 - time (sec): 5.41 - samples/sec: 3206.36 - lr: 0.000002 - momentum: 0.000000
2023-10-17 10:34:04,836 epoch 10 - iter 96/121 - loss 0.00746805 - time (sec): 6.21 - samples/sec: 3222.04 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:34:05,588 epoch 10 - iter 108/121 - loss 0.00920783 - time (sec): 6.96 - samples/sec: 3215.46 - lr: 0.000001 - momentum: 0.000000
2023-10-17 10:34:06,319 epoch 10 - iter 120/121 - loss 0.00895767 - time (sec): 7.69 - samples/sec: 3205.43 - lr: 0.000000 - momentum: 0.000000
2023-10-17 10:34:06,366 ----------------------------------------------------------------------------------------------------
2023-10-17 10:34:06,366 EPOCH 10 done: loss 0.0089 - lr: 0.000000
2023-10-17 10:34:07,175 DEV : loss 0.23311887681484222 - f1-score (micro avg) 0.836
2023-10-17 10:34:07,605 ----------------------------------------------------------------------------------------------------
2023-10-17 10:34:07,606 Loading model from best epoch ...
2023-10-17 10:34:09,049 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
2023-10-17 10:34:09,865
Results:
- F-score (micro) 0.8081
- F-score (macro) 0.5805
- Accuracy 0.6952
By class:
precision recall f1-score support
pers 0.8367 0.8849 0.8601 139
scope 0.8099 0.8915 0.8487 129
work 0.6452 0.7500 0.6936 80
loc 1.0000 0.3333 0.5000 9
date 0.0000 0.0000 0.0000 3
micro avg 0.7818 0.8361 0.8081 360
macro avg 0.6584 0.5719 0.5805 360
weighted avg 0.7816 0.8361 0.8029 360
2023-10-17 10:34:09,865 ----------------------------------------------------------------------------------------------------
|