File size: 24,232 Bytes
a3d316d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 |
2023-10-16 08:59:27,927 ----------------------------------------------------------------------------------------------------
2023-10-16 08:59:27,928 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=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-16 08:59:27,928 ----------------------------------------------------------------------------------------------------
2023-10-16 08:59:27,928 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
2023-10-16 08:59:27,928 ----------------------------------------------------------------------------------------------------
2023-10-16 08:59:27,928 Train: 7142 sentences
2023-10-16 08:59:27,928 (train_with_dev=False, train_with_test=False)
2023-10-16 08:59:27,928 ----------------------------------------------------------------------------------------------------
2023-10-16 08:59:27,928 Training Params:
2023-10-16 08:59:27,928 - learning_rate: "5e-05"
2023-10-16 08:59:27,928 - mini_batch_size: "4"
2023-10-16 08:59:27,928 - max_epochs: "10"
2023-10-16 08:59:27,928 - shuffle: "True"
2023-10-16 08:59:27,929 ----------------------------------------------------------------------------------------------------
2023-10-16 08:59:27,929 Plugins:
2023-10-16 08:59:27,929 - LinearScheduler | warmup_fraction: '0.1'
2023-10-16 08:59:27,929 ----------------------------------------------------------------------------------------------------
2023-10-16 08:59:27,929 Final evaluation on model from best epoch (best-model.pt)
2023-10-16 08:59:27,929 - metric: "('micro avg', 'f1-score')"
2023-10-16 08:59:27,929 ----------------------------------------------------------------------------------------------------
2023-10-16 08:59:27,929 Computation:
2023-10-16 08:59:27,929 - compute on device: cuda:0
2023-10-16 08:59:27,929 - embedding storage: none
2023-10-16 08:59:27,929 ----------------------------------------------------------------------------------------------------
2023-10-16 08:59:27,929 Model training base path: "hmbench-newseye/fr-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-16 08:59:27,929 ----------------------------------------------------------------------------------------------------
2023-10-16 08:59:27,929 ----------------------------------------------------------------------------------------------------
2023-10-16 08:59:36,701 epoch 1 - iter 178/1786 - loss 1.93950853 - time (sec): 8.77 - samples/sec: 2806.48 - lr: 0.000005 - momentum: 0.000000
2023-10-16 08:59:45,303 epoch 1 - iter 356/1786 - loss 1.19684096 - time (sec): 17.37 - samples/sec: 2873.23 - lr: 0.000010 - momentum: 0.000000
2023-10-16 08:59:53,918 epoch 1 - iter 534/1786 - loss 0.90350754 - time (sec): 25.99 - samples/sec: 2868.02 - lr: 0.000015 - momentum: 0.000000
2023-10-16 09:00:02,649 epoch 1 - iter 712/1786 - loss 0.73207711 - time (sec): 34.72 - samples/sec: 2899.46 - lr: 0.000020 - momentum: 0.000000
2023-10-16 09:00:11,223 epoch 1 - iter 890/1786 - loss 0.63211784 - time (sec): 43.29 - samples/sec: 2877.60 - lr: 0.000025 - momentum: 0.000000
2023-10-16 09:00:19,866 epoch 1 - iter 1068/1786 - loss 0.55770395 - time (sec): 51.94 - samples/sec: 2860.36 - lr: 0.000030 - momentum: 0.000000
2023-10-16 09:00:28,458 epoch 1 - iter 1246/1786 - loss 0.50479715 - time (sec): 60.53 - samples/sec: 2864.56 - lr: 0.000035 - momentum: 0.000000
2023-10-16 09:00:37,259 epoch 1 - iter 1424/1786 - loss 0.45844199 - time (sec): 69.33 - samples/sec: 2856.58 - lr: 0.000040 - momentum: 0.000000
2023-10-16 09:00:46,035 epoch 1 - iter 1602/1786 - loss 0.42492404 - time (sec): 78.11 - samples/sec: 2872.10 - lr: 0.000045 - momentum: 0.000000
2023-10-16 09:00:54,534 epoch 1 - iter 1780/1786 - loss 0.40035384 - time (sec): 86.60 - samples/sec: 2862.69 - lr: 0.000050 - momentum: 0.000000
2023-10-16 09:00:54,814 ----------------------------------------------------------------------------------------------------
2023-10-16 09:00:54,815 EPOCH 1 done: loss 0.3994 - lr: 0.000050
2023-10-16 09:00:57,828 DEV : loss 0.164632648229599 - f1-score (micro avg) 0.6901
2023-10-16 09:00:57,844 saving best model
2023-10-16 09:00:58,206 ----------------------------------------------------------------------------------------------------
2023-10-16 09:01:06,863 epoch 2 - iter 178/1786 - loss 0.11487137 - time (sec): 8.66 - samples/sec: 2797.96 - lr: 0.000049 - momentum: 0.000000
2023-10-16 09:01:15,536 epoch 2 - iter 356/1786 - loss 0.10865483 - time (sec): 17.33 - samples/sec: 2865.27 - lr: 0.000049 - momentum: 0.000000
2023-10-16 09:01:24,138 epoch 2 - iter 534/1786 - loss 0.11685553 - time (sec): 25.93 - samples/sec: 2839.41 - lr: 0.000048 - momentum: 0.000000
2023-10-16 09:01:32,735 epoch 2 - iter 712/1786 - loss 0.11885426 - time (sec): 34.53 - samples/sec: 2838.78 - lr: 0.000048 - momentum: 0.000000
2023-10-16 09:01:41,455 epoch 2 - iter 890/1786 - loss 0.11920458 - time (sec): 43.25 - samples/sec: 2830.37 - lr: 0.000047 - momentum: 0.000000
2023-10-16 09:01:50,204 epoch 2 - iter 1068/1786 - loss 0.11979459 - time (sec): 52.00 - samples/sec: 2832.05 - lr: 0.000047 - momentum: 0.000000
2023-10-16 09:01:58,827 epoch 2 - iter 1246/1786 - loss 0.12470555 - time (sec): 60.62 - samples/sec: 2847.73 - lr: 0.000046 - momentum: 0.000000
2023-10-16 09:02:07,725 epoch 2 - iter 1424/1786 - loss 0.12324524 - time (sec): 69.52 - samples/sec: 2848.49 - lr: 0.000046 - momentum: 0.000000
2023-10-16 09:02:16,897 epoch 2 - iter 1602/1786 - loss 0.12369002 - time (sec): 78.69 - samples/sec: 2834.51 - lr: 0.000045 - momentum: 0.000000
2023-10-16 09:02:25,680 epoch 2 - iter 1780/1786 - loss 0.12238966 - time (sec): 87.47 - samples/sec: 2833.26 - lr: 0.000044 - momentum: 0.000000
2023-10-16 09:02:25,944 ----------------------------------------------------------------------------------------------------
2023-10-16 09:02:25,944 EPOCH 2 done: loss 0.1228 - lr: 0.000044
2023-10-16 09:02:30,616 DEV : loss 0.1199851781129837 - f1-score (micro avg) 0.7637
2023-10-16 09:02:30,633 saving best model
2023-10-16 09:02:31,073 ----------------------------------------------------------------------------------------------------
2023-10-16 09:02:39,971 epoch 3 - iter 178/1786 - loss 0.08422764 - time (sec): 8.90 - samples/sec: 2650.63 - lr: 0.000044 - momentum: 0.000000
2023-10-16 09:02:48,902 epoch 3 - iter 356/1786 - loss 0.08832799 - time (sec): 17.83 - samples/sec: 2666.78 - lr: 0.000043 - momentum: 0.000000
2023-10-16 09:02:57,286 epoch 3 - iter 534/1786 - loss 0.08603336 - time (sec): 26.21 - samples/sec: 2720.96 - lr: 0.000043 - momentum: 0.000000
2023-10-16 09:03:06,353 epoch 3 - iter 712/1786 - loss 0.08439697 - time (sec): 35.28 - samples/sec: 2768.02 - lr: 0.000042 - momentum: 0.000000
2023-10-16 09:03:15,079 epoch 3 - iter 890/1786 - loss 0.08403027 - time (sec): 44.00 - samples/sec: 2776.45 - lr: 0.000042 - momentum: 0.000000
2023-10-16 09:03:23,832 epoch 3 - iter 1068/1786 - loss 0.08339811 - time (sec): 52.76 - samples/sec: 2796.66 - lr: 0.000041 - momentum: 0.000000
2023-10-16 09:03:32,601 epoch 3 - iter 1246/1786 - loss 0.08366626 - time (sec): 61.53 - samples/sec: 2796.00 - lr: 0.000041 - momentum: 0.000000
2023-10-16 09:03:41,648 epoch 3 - iter 1424/1786 - loss 0.08520626 - time (sec): 70.57 - samples/sec: 2811.10 - lr: 0.000040 - momentum: 0.000000
2023-10-16 09:03:50,239 epoch 3 - iter 1602/1786 - loss 0.08861563 - time (sec): 79.16 - samples/sec: 2795.22 - lr: 0.000039 - momentum: 0.000000
2023-10-16 09:03:59,241 epoch 3 - iter 1780/1786 - loss 0.08777026 - time (sec): 88.17 - samples/sec: 2814.81 - lr: 0.000039 - momentum: 0.000000
2023-10-16 09:03:59,514 ----------------------------------------------------------------------------------------------------
2023-10-16 09:03:59,514 EPOCH 3 done: loss 0.0876 - lr: 0.000039
2023-10-16 09:04:03,562 DEV : loss 0.12482591718435287 - f1-score (micro avg) 0.7656
2023-10-16 09:04:03,577 saving best model
2023-10-16 09:04:04,030 ----------------------------------------------------------------------------------------------------
2023-10-16 09:04:12,850 epoch 4 - iter 178/1786 - loss 0.07490226 - time (sec): 8.82 - samples/sec: 2826.41 - lr: 0.000038 - momentum: 0.000000
2023-10-16 09:04:21,987 epoch 4 - iter 356/1786 - loss 0.07388713 - time (sec): 17.96 - samples/sec: 2819.39 - lr: 0.000038 - momentum: 0.000000
2023-10-16 09:04:30,657 epoch 4 - iter 534/1786 - loss 0.07199900 - time (sec): 26.63 - samples/sec: 2796.78 - lr: 0.000037 - momentum: 0.000000
2023-10-16 09:04:39,443 epoch 4 - iter 712/1786 - loss 0.06917859 - time (sec): 35.41 - samples/sec: 2848.90 - lr: 0.000037 - momentum: 0.000000
2023-10-16 09:04:48,062 epoch 4 - iter 890/1786 - loss 0.06579901 - time (sec): 44.03 - samples/sec: 2843.02 - lr: 0.000036 - momentum: 0.000000
2023-10-16 09:04:57,041 epoch 4 - iter 1068/1786 - loss 0.06377113 - time (sec): 53.01 - samples/sec: 2843.47 - lr: 0.000036 - momentum: 0.000000
2023-10-16 09:05:05,388 epoch 4 - iter 1246/1786 - loss 0.06597385 - time (sec): 61.36 - samples/sec: 2819.29 - lr: 0.000035 - momentum: 0.000000
2023-10-16 09:05:14,250 epoch 4 - iter 1424/1786 - loss 0.06743665 - time (sec): 70.22 - samples/sec: 2821.07 - lr: 0.000034 - momentum: 0.000000
2023-10-16 09:05:23,540 epoch 4 - iter 1602/1786 - loss 0.06657675 - time (sec): 79.51 - samples/sec: 2802.30 - lr: 0.000034 - momentum: 0.000000
2023-10-16 09:05:32,491 epoch 4 - iter 1780/1786 - loss 0.06664923 - time (sec): 88.46 - samples/sec: 2801.56 - lr: 0.000033 - momentum: 0.000000
2023-10-16 09:05:32,817 ----------------------------------------------------------------------------------------------------
2023-10-16 09:05:32,817 EPOCH 4 done: loss 0.0669 - lr: 0.000033
2023-10-16 09:05:36,817 DEV : loss 0.17001311480998993 - f1-score (micro avg) 0.7773
2023-10-16 09:05:36,833 saving best model
2023-10-16 09:05:37,282 ----------------------------------------------------------------------------------------------------
2023-10-16 09:05:46,244 epoch 5 - iter 178/1786 - loss 0.05321277 - time (sec): 8.96 - samples/sec: 2962.54 - lr: 0.000033 - momentum: 0.000000
2023-10-16 09:05:55,007 epoch 5 - iter 356/1786 - loss 0.05027400 - time (sec): 17.72 - samples/sec: 2914.67 - lr: 0.000032 - momentum: 0.000000
2023-10-16 09:06:03,991 epoch 5 - iter 534/1786 - loss 0.05055258 - time (sec): 26.70 - samples/sec: 2883.74 - lr: 0.000032 - momentum: 0.000000
2023-10-16 09:06:12,573 epoch 5 - iter 712/1786 - loss 0.04836095 - time (sec): 35.29 - samples/sec: 2930.97 - lr: 0.000031 - momentum: 0.000000
2023-10-16 09:06:21,516 epoch 5 - iter 890/1786 - loss 0.04806037 - time (sec): 44.23 - samples/sec: 2907.81 - lr: 0.000031 - momentum: 0.000000
2023-10-16 09:06:30,294 epoch 5 - iter 1068/1786 - loss 0.04703404 - time (sec): 53.01 - samples/sec: 2880.93 - lr: 0.000030 - momentum: 0.000000
2023-10-16 09:06:38,968 epoch 5 - iter 1246/1786 - loss 0.04922091 - time (sec): 61.68 - samples/sec: 2876.27 - lr: 0.000029 - momentum: 0.000000
2023-10-16 09:06:47,627 epoch 5 - iter 1424/1786 - loss 0.05007639 - time (sec): 70.34 - samples/sec: 2846.92 - lr: 0.000029 - momentum: 0.000000
2023-10-16 09:06:56,313 epoch 5 - iter 1602/1786 - loss 0.05063967 - time (sec): 79.03 - samples/sec: 2824.96 - lr: 0.000028 - momentum: 0.000000
2023-10-16 09:07:04,982 epoch 5 - iter 1780/1786 - loss 0.05136262 - time (sec): 87.70 - samples/sec: 2821.84 - lr: 0.000028 - momentum: 0.000000
2023-10-16 09:07:05,332 ----------------------------------------------------------------------------------------------------
2023-10-16 09:07:05,332 EPOCH 5 done: loss 0.0514 - lr: 0.000028
2023-10-16 09:07:09,890 DEV : loss 0.1479008048772812 - f1-score (micro avg) 0.7853
2023-10-16 09:07:09,906 saving best model
2023-10-16 09:07:10,355 ----------------------------------------------------------------------------------------------------
2023-10-16 09:07:19,191 epoch 6 - iter 178/1786 - loss 0.04084234 - time (sec): 8.83 - samples/sec: 2787.69 - lr: 0.000027 - momentum: 0.000000
2023-10-16 09:07:27,857 epoch 6 - iter 356/1786 - loss 0.03988235 - time (sec): 17.50 - samples/sec: 2819.19 - lr: 0.000027 - momentum: 0.000000
2023-10-16 09:07:36,931 epoch 6 - iter 534/1786 - loss 0.04230182 - time (sec): 26.57 - samples/sec: 2812.14 - lr: 0.000026 - momentum: 0.000000
2023-10-16 09:07:45,917 epoch 6 - iter 712/1786 - loss 0.04315507 - time (sec): 35.56 - samples/sec: 2844.22 - lr: 0.000026 - momentum: 0.000000
2023-10-16 09:07:54,708 epoch 6 - iter 890/1786 - loss 0.03981327 - time (sec): 44.35 - samples/sec: 2841.28 - lr: 0.000025 - momentum: 0.000000
2023-10-16 09:08:03,319 epoch 6 - iter 1068/1786 - loss 0.04083548 - time (sec): 52.96 - samples/sec: 2847.09 - lr: 0.000024 - momentum: 0.000000
2023-10-16 09:08:12,027 epoch 6 - iter 1246/1786 - loss 0.03919229 - time (sec): 61.67 - samples/sec: 2864.30 - lr: 0.000024 - momentum: 0.000000
2023-10-16 09:08:20,916 epoch 6 - iter 1424/1786 - loss 0.03877892 - time (sec): 70.56 - samples/sec: 2859.52 - lr: 0.000023 - momentum: 0.000000
2023-10-16 09:08:29,628 epoch 6 - iter 1602/1786 - loss 0.03867605 - time (sec): 79.27 - samples/sec: 2826.83 - lr: 0.000023 - momentum: 0.000000
2023-10-16 09:08:38,418 epoch 6 - iter 1780/1786 - loss 0.03727970 - time (sec): 88.06 - samples/sec: 2818.19 - lr: 0.000022 - momentum: 0.000000
2023-10-16 09:08:38,691 ----------------------------------------------------------------------------------------------------
2023-10-16 09:08:38,692 EPOCH 6 done: loss 0.0374 - lr: 0.000022
2023-10-16 09:08:42,752 DEV : loss 0.17757560312747955 - f1-score (micro avg) 0.7965
2023-10-16 09:08:42,768 saving best model
2023-10-16 09:08:43,226 ----------------------------------------------------------------------------------------------------
2023-10-16 09:08:52,089 epoch 7 - iter 178/1786 - loss 0.02636464 - time (sec): 8.86 - samples/sec: 2939.18 - lr: 0.000022 - momentum: 0.000000
2023-10-16 09:09:00,968 epoch 7 - iter 356/1786 - loss 0.02439338 - time (sec): 17.74 - samples/sec: 2863.45 - lr: 0.000021 - momentum: 0.000000
2023-10-16 09:09:10,107 epoch 7 - iter 534/1786 - loss 0.02868364 - time (sec): 26.88 - samples/sec: 2854.02 - lr: 0.000021 - momentum: 0.000000
2023-10-16 09:09:19,060 epoch 7 - iter 712/1786 - loss 0.02820433 - time (sec): 35.83 - samples/sec: 2819.85 - lr: 0.000020 - momentum: 0.000000
2023-10-16 09:09:27,628 epoch 7 - iter 890/1786 - loss 0.02740708 - time (sec): 44.40 - samples/sec: 2785.70 - lr: 0.000019 - momentum: 0.000000
2023-10-16 09:09:36,262 epoch 7 - iter 1068/1786 - loss 0.02931505 - time (sec): 53.03 - samples/sec: 2806.18 - lr: 0.000019 - momentum: 0.000000
2023-10-16 09:09:45,216 epoch 7 - iter 1246/1786 - loss 0.03044116 - time (sec): 61.98 - samples/sec: 2800.05 - lr: 0.000018 - momentum: 0.000000
2023-10-16 09:09:54,031 epoch 7 - iter 1424/1786 - loss 0.02978701 - time (sec): 70.80 - samples/sec: 2799.97 - lr: 0.000018 - momentum: 0.000000
2023-10-16 09:10:02,873 epoch 7 - iter 1602/1786 - loss 0.02989961 - time (sec): 79.64 - samples/sec: 2806.68 - lr: 0.000017 - momentum: 0.000000
2023-10-16 09:10:11,596 epoch 7 - iter 1780/1786 - loss 0.02875032 - time (sec): 88.37 - samples/sec: 2808.03 - lr: 0.000017 - momentum: 0.000000
2023-10-16 09:10:11,893 ----------------------------------------------------------------------------------------------------
2023-10-16 09:10:11,893 EPOCH 7 done: loss 0.0287 - lr: 0.000017
2023-10-16 09:10:16,471 DEV : loss 0.18708880245685577 - f1-score (micro avg) 0.7881
2023-10-16 09:10:16,487 ----------------------------------------------------------------------------------------------------
2023-10-16 09:10:25,496 epoch 8 - iter 178/1786 - loss 0.02251176 - time (sec): 9.01 - samples/sec: 2740.85 - lr: 0.000016 - momentum: 0.000000
2023-10-16 09:10:34,741 epoch 8 - iter 356/1786 - loss 0.02431309 - time (sec): 18.25 - samples/sec: 2821.65 - lr: 0.000016 - momentum: 0.000000
2023-10-16 09:10:43,326 epoch 8 - iter 534/1786 - loss 0.02293182 - time (sec): 26.84 - samples/sec: 2799.53 - lr: 0.000015 - momentum: 0.000000
2023-10-16 09:10:52,052 epoch 8 - iter 712/1786 - loss 0.02282164 - time (sec): 35.56 - samples/sec: 2771.58 - lr: 0.000014 - momentum: 0.000000
2023-10-16 09:11:00,770 epoch 8 - iter 890/1786 - loss 0.02376710 - time (sec): 44.28 - samples/sec: 2737.99 - lr: 0.000014 - momentum: 0.000000
2023-10-16 09:11:10,034 epoch 8 - iter 1068/1786 - loss 0.02298881 - time (sec): 53.55 - samples/sec: 2757.59 - lr: 0.000013 - momentum: 0.000000
2023-10-16 09:11:18,807 epoch 8 - iter 1246/1786 - loss 0.02269837 - time (sec): 62.32 - samples/sec: 2768.66 - lr: 0.000013 - momentum: 0.000000
2023-10-16 09:11:27,722 epoch 8 - iter 1424/1786 - loss 0.02263600 - time (sec): 71.23 - samples/sec: 2798.75 - lr: 0.000012 - momentum: 0.000000
2023-10-16 09:11:36,520 epoch 8 - iter 1602/1786 - loss 0.02277881 - time (sec): 80.03 - samples/sec: 2808.31 - lr: 0.000012 - momentum: 0.000000
2023-10-16 09:11:44,784 epoch 8 - iter 1780/1786 - loss 0.02229267 - time (sec): 88.30 - samples/sec: 2811.57 - lr: 0.000011 - momentum: 0.000000
2023-10-16 09:11:45,035 ----------------------------------------------------------------------------------------------------
2023-10-16 09:11:45,036 EPOCH 8 done: loss 0.0224 - lr: 0.000011
2023-10-16 09:11:49,077 DEV : loss 0.2031172513961792 - f1-score (micro avg) 0.7976
2023-10-16 09:11:49,093 saving best model
2023-10-16 09:11:49,555 ----------------------------------------------------------------------------------------------------
2023-10-16 09:11:58,992 epoch 9 - iter 178/1786 - loss 0.02189545 - time (sec): 9.43 - samples/sec: 2629.64 - lr: 0.000011 - momentum: 0.000000
2023-10-16 09:12:07,517 epoch 9 - iter 356/1786 - loss 0.01747601 - time (sec): 17.96 - samples/sec: 2762.19 - lr: 0.000010 - momentum: 0.000000
2023-10-16 09:12:16,272 epoch 9 - iter 534/1786 - loss 0.01786279 - time (sec): 26.71 - samples/sec: 2828.56 - lr: 0.000009 - momentum: 0.000000
2023-10-16 09:12:25,209 epoch 9 - iter 712/1786 - loss 0.01550724 - time (sec): 35.65 - samples/sec: 2797.98 - lr: 0.000009 - momentum: 0.000000
2023-10-16 09:12:33,988 epoch 9 - iter 890/1786 - loss 0.01545366 - time (sec): 44.43 - samples/sec: 2794.14 - lr: 0.000008 - momentum: 0.000000
2023-10-16 09:12:42,895 epoch 9 - iter 1068/1786 - loss 0.01489816 - time (sec): 53.34 - samples/sec: 2812.03 - lr: 0.000008 - momentum: 0.000000
2023-10-16 09:12:51,587 epoch 9 - iter 1246/1786 - loss 0.01528282 - time (sec): 62.03 - samples/sec: 2832.17 - lr: 0.000007 - momentum: 0.000000
2023-10-16 09:13:00,247 epoch 9 - iter 1424/1786 - loss 0.01453460 - time (sec): 70.69 - samples/sec: 2840.93 - lr: 0.000007 - momentum: 0.000000
2023-10-16 09:13:08,659 epoch 9 - iter 1602/1786 - loss 0.01464221 - time (sec): 79.10 - samples/sec: 2832.27 - lr: 0.000006 - momentum: 0.000000
2023-10-16 09:13:17,327 epoch 9 - iter 1780/1786 - loss 0.01504874 - time (sec): 87.77 - samples/sec: 2825.63 - lr: 0.000006 - momentum: 0.000000
2023-10-16 09:13:17,625 ----------------------------------------------------------------------------------------------------
2023-10-16 09:13:17,625 EPOCH 9 done: loss 0.0150 - lr: 0.000006
2023-10-16 09:13:21,657 DEV : loss 0.18724995851516724 - f1-score (micro avg) 0.8027
2023-10-16 09:13:21,673 saving best model
2023-10-16 09:13:22,121 ----------------------------------------------------------------------------------------------------
2023-10-16 09:13:31,015 epoch 10 - iter 178/1786 - loss 0.00571772 - time (sec): 8.89 - samples/sec: 2914.90 - lr: 0.000005 - momentum: 0.000000
2023-10-16 09:13:39,977 epoch 10 - iter 356/1786 - loss 0.00917164 - time (sec): 17.85 - samples/sec: 2941.77 - lr: 0.000004 - momentum: 0.000000
2023-10-16 09:13:49,087 epoch 10 - iter 534/1786 - loss 0.00931697 - time (sec): 26.96 - samples/sec: 2873.81 - lr: 0.000004 - momentum: 0.000000
2023-10-16 09:13:57,666 epoch 10 - iter 712/1786 - loss 0.00909777 - time (sec): 35.54 - samples/sec: 2879.62 - lr: 0.000003 - momentum: 0.000000
2023-10-16 09:14:06,392 epoch 10 - iter 890/1786 - loss 0.00911553 - time (sec): 44.27 - samples/sec: 2859.28 - lr: 0.000003 - momentum: 0.000000
2023-10-16 09:14:14,995 epoch 10 - iter 1068/1786 - loss 0.00852015 - time (sec): 52.87 - samples/sec: 2838.41 - lr: 0.000002 - momentum: 0.000000
2023-10-16 09:14:23,665 epoch 10 - iter 1246/1786 - loss 0.00864843 - time (sec): 61.54 - samples/sec: 2817.84 - lr: 0.000002 - momentum: 0.000000
2023-10-16 09:14:32,625 epoch 10 - iter 1424/1786 - loss 0.00874912 - time (sec): 70.50 - samples/sec: 2815.35 - lr: 0.000001 - momentum: 0.000000
2023-10-16 09:14:41,364 epoch 10 - iter 1602/1786 - loss 0.00888492 - time (sec): 79.24 - samples/sec: 2821.53 - lr: 0.000001 - momentum: 0.000000
2023-10-16 09:14:50,107 epoch 10 - iter 1780/1786 - loss 0.00921080 - time (sec): 87.98 - samples/sec: 2818.57 - lr: 0.000000 - momentum: 0.000000
2023-10-16 09:14:50,383 ----------------------------------------------------------------------------------------------------
2023-10-16 09:14:50,383 EPOCH 10 done: loss 0.0092 - lr: 0.000000
2023-10-16 09:14:55,010 DEV : loss 0.2006855458021164 - f1-score (micro avg) 0.8054
2023-10-16 09:14:55,026 saving best model
2023-10-16 09:14:55,864 ----------------------------------------------------------------------------------------------------
2023-10-16 09:14:55,865 Loading model from best epoch ...
2023-10-16 09:14:57,658 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
2023-10-16 09:15:06,928
Results:
- F-score (micro) 0.6898
- F-score (macro) 0.6075
- Accuracy 0.5436
By class:
precision recall f1-score support
LOC 0.7223 0.6721 0.6963 1095
PER 0.7563 0.7668 0.7615 1012
ORG 0.4680 0.5742 0.5157 357
HumanProd 0.3559 0.6364 0.4565 33
micro avg 0.6837 0.6960 0.6898 2497
macro avg 0.5756 0.6624 0.6075 2497
weighted avg 0.6949 0.6960 0.6938 2497
2023-10-16 09:15:06,928 ----------------------------------------------------------------------------------------------------
|