File size: 23,904 Bytes
326c952 |
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 11:08:01,482 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:01,483 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 11:08:01,483 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:01,483 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 11:08:01,483 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:01,483 Train: 966 sentences
2023-10-17 11:08:01,483 (train_with_dev=False, train_with_test=False)
2023-10-17 11:08:01,483 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:01,483 Training Params:
2023-10-17 11:08:01,483 - learning_rate: "5e-05"
2023-10-17 11:08:01,483 - mini_batch_size: "8"
2023-10-17 11:08:01,483 - max_epochs: "10"
2023-10-17 11:08:01,483 - shuffle: "True"
2023-10-17 11:08:01,483 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:01,483 Plugins:
2023-10-17 11:08:01,483 - TensorboardLogger
2023-10-17 11:08:01,483 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 11:08:01,483 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:01,483 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 11:08:01,483 - metric: "('micro avg', 'f1-score')"
2023-10-17 11:08:01,483 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:01,483 Computation:
2023-10-17 11:08:01,483 - compute on device: cuda:0
2023-10-17 11:08:01,484 - embedding storage: none
2023-10-17 11:08:01,484 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:01,484 Model training base path: "hmbench-ajmc/fr-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-17 11:08:01,484 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:01,484 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:01,484 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 11:08:02,318 epoch 1 - iter 12/121 - loss 4.64640638 - time (sec): 0.83 - samples/sec: 3140.70 - lr: 0.000005 - momentum: 0.000000
2023-10-17 11:08:03,081 epoch 1 - iter 24/121 - loss 3.99522762 - time (sec): 1.60 - samples/sec: 3122.73 - lr: 0.000010 - momentum: 0.000000
2023-10-17 11:08:03,830 epoch 1 - iter 36/121 - loss 3.16441520 - time (sec): 2.35 - samples/sec: 3155.90 - lr: 0.000014 - momentum: 0.000000
2023-10-17 11:08:04,553 epoch 1 - iter 48/121 - loss 2.57702603 - time (sec): 3.07 - samples/sec: 3184.33 - lr: 0.000019 - momentum: 0.000000
2023-10-17 11:08:05,318 epoch 1 - iter 60/121 - loss 2.14240888 - time (sec): 3.83 - samples/sec: 3232.40 - lr: 0.000024 - momentum: 0.000000
2023-10-17 11:08:06,077 epoch 1 - iter 72/121 - loss 1.85994774 - time (sec): 4.59 - samples/sec: 3241.62 - lr: 0.000029 - momentum: 0.000000
2023-10-17 11:08:06,769 epoch 1 - iter 84/121 - loss 1.66472101 - time (sec): 5.28 - samples/sec: 3253.63 - lr: 0.000034 - momentum: 0.000000
2023-10-17 11:08:07,534 epoch 1 - iter 96/121 - loss 1.50200708 - time (sec): 6.05 - samples/sec: 3245.44 - lr: 0.000039 - momentum: 0.000000
2023-10-17 11:08:08,293 epoch 1 - iter 108/121 - loss 1.35604792 - time (sec): 6.81 - samples/sec: 3266.64 - lr: 0.000044 - momentum: 0.000000
2023-10-17 11:08:09,024 epoch 1 - iter 120/121 - loss 1.25714367 - time (sec): 7.54 - samples/sec: 3268.13 - lr: 0.000049 - momentum: 0.000000
2023-10-17 11:08:09,079 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:09,079 EPOCH 1 done: loss 1.2531 - lr: 0.000049
2023-10-17 11:08:09,987 DEV : loss 0.2125282734632492 - f1-score (micro avg) 0.5735
2023-10-17 11:08:09,994 saving best model
2023-10-17 11:08:10,387 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:11,148 epoch 2 - iter 12/121 - loss 0.21764231 - time (sec): 0.76 - samples/sec: 3515.23 - lr: 0.000049 - momentum: 0.000000
2023-10-17 11:08:11,940 epoch 2 - iter 24/121 - loss 0.21403451 - time (sec): 1.55 - samples/sec: 3305.25 - lr: 0.000049 - momentum: 0.000000
2023-10-17 11:08:12,715 epoch 2 - iter 36/121 - loss 0.20849589 - time (sec): 2.33 - samples/sec: 3163.91 - lr: 0.000048 - momentum: 0.000000
2023-10-17 11:08:13,518 epoch 2 - iter 48/121 - loss 0.21445341 - time (sec): 3.13 - samples/sec: 3181.97 - lr: 0.000048 - momentum: 0.000000
2023-10-17 11:08:14,201 epoch 2 - iter 60/121 - loss 0.21111530 - time (sec): 3.81 - samples/sec: 3192.03 - lr: 0.000047 - momentum: 0.000000
2023-10-17 11:08:14,970 epoch 2 - iter 72/121 - loss 0.20856125 - time (sec): 4.58 - samples/sec: 3197.45 - lr: 0.000047 - momentum: 0.000000
2023-10-17 11:08:15,747 epoch 2 - iter 84/121 - loss 0.20056727 - time (sec): 5.36 - samples/sec: 3229.91 - lr: 0.000046 - momentum: 0.000000
2023-10-17 11:08:16,510 epoch 2 - iter 96/121 - loss 0.19028230 - time (sec): 6.12 - samples/sec: 3241.23 - lr: 0.000046 - momentum: 0.000000
2023-10-17 11:08:17,257 epoch 2 - iter 108/121 - loss 0.18897016 - time (sec): 6.87 - samples/sec: 3246.99 - lr: 0.000045 - momentum: 0.000000
2023-10-17 11:08:18,016 epoch 2 - iter 120/121 - loss 0.18464719 - time (sec): 7.63 - samples/sec: 3230.93 - lr: 0.000045 - momentum: 0.000000
2023-10-17 11:08:18,062 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:18,063 EPOCH 2 done: loss 0.1849 - lr: 0.000045
2023-10-17 11:08:18,872 DEV : loss 0.1422046422958374 - f1-score (micro avg) 0.7617
2023-10-17 11:08:18,878 saving best model
2023-10-17 11:08:19,429 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:20,275 epoch 3 - iter 12/121 - loss 0.10190167 - time (sec): 0.84 - samples/sec: 2692.70 - lr: 0.000044 - momentum: 0.000000
2023-10-17 11:08:21,108 epoch 3 - iter 24/121 - loss 0.10884942 - time (sec): 1.68 - samples/sec: 2845.97 - lr: 0.000043 - momentum: 0.000000
2023-10-17 11:08:21,986 epoch 3 - iter 36/121 - loss 0.10662259 - time (sec): 2.55 - samples/sec: 2876.54 - lr: 0.000043 - momentum: 0.000000
2023-10-17 11:08:22,893 epoch 3 - iter 48/121 - loss 0.10825608 - time (sec): 3.46 - samples/sec: 2835.33 - lr: 0.000042 - momentum: 0.000000
2023-10-17 11:08:23,661 epoch 3 - iter 60/121 - loss 0.11156383 - time (sec): 4.23 - samples/sec: 2893.76 - lr: 0.000042 - momentum: 0.000000
2023-10-17 11:08:24,420 epoch 3 - iter 72/121 - loss 0.11438574 - time (sec): 4.99 - samples/sec: 2921.92 - lr: 0.000041 - momentum: 0.000000
2023-10-17 11:08:25,109 epoch 3 - iter 84/121 - loss 0.11338348 - time (sec): 5.68 - samples/sec: 2963.65 - lr: 0.000041 - momentum: 0.000000
2023-10-17 11:08:25,833 epoch 3 - iter 96/121 - loss 0.10924170 - time (sec): 6.40 - samples/sec: 3006.45 - lr: 0.000040 - momentum: 0.000000
2023-10-17 11:08:26,557 epoch 3 - iter 108/121 - loss 0.10890897 - time (sec): 7.13 - samples/sec: 3068.47 - lr: 0.000040 - momentum: 0.000000
2023-10-17 11:08:27,364 epoch 3 - iter 120/121 - loss 0.10887373 - time (sec): 7.93 - samples/sec: 3095.94 - lr: 0.000039 - momentum: 0.000000
2023-10-17 11:08:27,416 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:27,416 EPOCH 3 done: loss 0.1081 - lr: 0.000039
2023-10-17 11:08:28,182 DEV : loss 0.1380065679550171 - f1-score (micro avg) 0.8312
2023-10-17 11:08:28,187 saving best model
2023-10-17 11:08:28,750 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:29,522 epoch 4 - iter 12/121 - loss 0.07019196 - time (sec): 0.77 - samples/sec: 3201.37 - lr: 0.000038 - momentum: 0.000000
2023-10-17 11:08:30,263 epoch 4 - iter 24/121 - loss 0.08708994 - time (sec): 1.51 - samples/sec: 3150.58 - lr: 0.000038 - momentum: 0.000000
2023-10-17 11:08:31,016 epoch 4 - iter 36/121 - loss 0.07707486 - time (sec): 2.26 - samples/sec: 3236.51 - lr: 0.000037 - momentum: 0.000000
2023-10-17 11:08:31,780 epoch 4 - iter 48/121 - loss 0.08053888 - time (sec): 3.03 - samples/sec: 3197.85 - lr: 0.000037 - momentum: 0.000000
2023-10-17 11:08:32,554 epoch 4 - iter 60/121 - loss 0.07902996 - time (sec): 3.80 - samples/sec: 3187.67 - lr: 0.000036 - momentum: 0.000000
2023-10-17 11:08:33,392 epoch 4 - iter 72/121 - loss 0.07645033 - time (sec): 4.64 - samples/sec: 3185.72 - lr: 0.000036 - momentum: 0.000000
2023-10-17 11:08:34,119 epoch 4 - iter 84/121 - loss 0.07914731 - time (sec): 5.37 - samples/sec: 3177.40 - lr: 0.000035 - momentum: 0.000000
2023-10-17 11:08:34,864 epoch 4 - iter 96/121 - loss 0.08010322 - time (sec): 6.11 - samples/sec: 3219.41 - lr: 0.000035 - momentum: 0.000000
2023-10-17 11:08:35,606 epoch 4 - iter 108/121 - loss 0.07826876 - time (sec): 6.85 - samples/sec: 3227.24 - lr: 0.000034 - momentum: 0.000000
2023-10-17 11:08:36,361 epoch 4 - iter 120/121 - loss 0.07554766 - time (sec): 7.61 - samples/sec: 3226.33 - lr: 0.000034 - momentum: 0.000000
2023-10-17 11:08:36,413 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:36,413 EPOCH 4 done: loss 0.0750 - lr: 0.000034
2023-10-17 11:08:37,162 DEV : loss 0.15543414652347565 - f1-score (micro avg) 0.8178
2023-10-17 11:08:37,167 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:37,826 epoch 5 - iter 12/121 - loss 0.04766896 - time (sec): 0.66 - samples/sec: 3563.90 - lr: 0.000033 - momentum: 0.000000
2023-10-17 11:08:38,573 epoch 5 - iter 24/121 - loss 0.04508118 - time (sec): 1.40 - samples/sec: 3377.50 - lr: 0.000032 - momentum: 0.000000
2023-10-17 11:08:39,321 epoch 5 - iter 36/121 - loss 0.06730774 - time (sec): 2.15 - samples/sec: 3350.74 - lr: 0.000032 - momentum: 0.000000
2023-10-17 11:08:40,116 epoch 5 - iter 48/121 - loss 0.07034756 - time (sec): 2.95 - samples/sec: 3343.74 - lr: 0.000031 - momentum: 0.000000
2023-10-17 11:08:40,891 epoch 5 - iter 60/121 - loss 0.06704156 - time (sec): 3.72 - samples/sec: 3284.32 - lr: 0.000031 - momentum: 0.000000
2023-10-17 11:08:41,677 epoch 5 - iter 72/121 - loss 0.06279205 - time (sec): 4.51 - samples/sec: 3306.89 - lr: 0.000030 - momentum: 0.000000
2023-10-17 11:08:42,381 epoch 5 - iter 84/121 - loss 0.06617666 - time (sec): 5.21 - samples/sec: 3279.85 - lr: 0.000030 - momentum: 0.000000
2023-10-17 11:08:43,202 epoch 5 - iter 96/121 - loss 0.06529293 - time (sec): 6.03 - samples/sec: 3252.32 - lr: 0.000029 - momentum: 0.000000
2023-10-17 11:08:43,941 epoch 5 - iter 108/121 - loss 0.06344772 - time (sec): 6.77 - samples/sec: 3268.41 - lr: 0.000029 - momentum: 0.000000
2023-10-17 11:08:44,647 epoch 5 - iter 120/121 - loss 0.06037539 - time (sec): 7.48 - samples/sec: 3285.74 - lr: 0.000028 - momentum: 0.000000
2023-10-17 11:08:44,701 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:44,701 EPOCH 5 done: loss 0.0600 - lr: 0.000028
2023-10-17 11:08:45,476 DEV : loss 0.16548091173171997 - f1-score (micro avg) 0.8331
2023-10-17 11:08:45,481 saving best model
2023-10-17 11:08:46,020 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:46,755 epoch 6 - iter 12/121 - loss 0.03259283 - time (sec): 0.73 - samples/sec: 3292.80 - lr: 0.000027 - momentum: 0.000000
2023-10-17 11:08:47,480 epoch 6 - iter 24/121 - loss 0.03577944 - time (sec): 1.46 - samples/sec: 3333.04 - lr: 0.000027 - momentum: 0.000000
2023-10-17 11:08:48,318 epoch 6 - iter 36/121 - loss 0.03375939 - time (sec): 2.29 - samples/sec: 3286.49 - lr: 0.000026 - momentum: 0.000000
2023-10-17 11:08:49,095 epoch 6 - iter 48/121 - loss 0.03529970 - time (sec): 3.07 - samples/sec: 3239.23 - lr: 0.000026 - momentum: 0.000000
2023-10-17 11:08:49,861 epoch 6 - iter 60/121 - loss 0.03304757 - time (sec): 3.84 - samples/sec: 3201.69 - lr: 0.000025 - momentum: 0.000000
2023-10-17 11:08:50,619 epoch 6 - iter 72/121 - loss 0.03748447 - time (sec): 4.60 - samples/sec: 3193.26 - lr: 0.000025 - momentum: 0.000000
2023-10-17 11:08:51,383 epoch 6 - iter 84/121 - loss 0.03924613 - time (sec): 5.36 - samples/sec: 3226.65 - lr: 0.000024 - momentum: 0.000000
2023-10-17 11:08:52,169 epoch 6 - iter 96/121 - loss 0.03747713 - time (sec): 6.14 - samples/sec: 3243.99 - lr: 0.000024 - momentum: 0.000000
2023-10-17 11:08:52,902 epoch 6 - iter 108/121 - loss 0.03732658 - time (sec): 6.88 - samples/sec: 3240.78 - lr: 0.000023 - momentum: 0.000000
2023-10-17 11:08:53,620 epoch 6 - iter 120/121 - loss 0.03850823 - time (sec): 7.60 - samples/sec: 3240.53 - lr: 0.000022 - momentum: 0.000000
2023-10-17 11:08:53,667 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:53,667 EPOCH 6 done: loss 0.0383 - lr: 0.000022
2023-10-17 11:08:54,432 DEV : loss 0.19360701739788055 - f1-score (micro avg) 0.8263
2023-10-17 11:08:54,437 ----------------------------------------------------------------------------------------------------
2023-10-17 11:08:55,126 epoch 7 - iter 12/121 - loss 0.05617147 - time (sec): 0.69 - samples/sec: 3192.31 - lr: 0.000022 - momentum: 0.000000
2023-10-17 11:08:55,900 epoch 7 - iter 24/121 - loss 0.04013345 - time (sec): 1.46 - samples/sec: 3077.29 - lr: 0.000021 - momentum: 0.000000
2023-10-17 11:08:56,654 epoch 7 - iter 36/121 - loss 0.03935435 - time (sec): 2.22 - samples/sec: 3213.72 - lr: 0.000021 - momentum: 0.000000
2023-10-17 11:08:57,417 epoch 7 - iter 48/121 - loss 0.03899974 - time (sec): 2.98 - samples/sec: 3237.49 - lr: 0.000020 - momentum: 0.000000
2023-10-17 11:08:58,156 epoch 7 - iter 60/121 - loss 0.03729358 - time (sec): 3.72 - samples/sec: 3253.20 - lr: 0.000020 - momentum: 0.000000
2023-10-17 11:08:58,955 epoch 7 - iter 72/121 - loss 0.03709520 - time (sec): 4.52 - samples/sec: 3304.23 - lr: 0.000019 - momentum: 0.000000
2023-10-17 11:08:59,758 epoch 7 - iter 84/121 - loss 0.03509845 - time (sec): 5.32 - samples/sec: 3294.34 - lr: 0.000019 - momentum: 0.000000
2023-10-17 11:09:00,540 epoch 7 - iter 96/121 - loss 0.03302967 - time (sec): 6.10 - samples/sec: 3281.68 - lr: 0.000018 - momentum: 0.000000
2023-10-17 11:09:01,317 epoch 7 - iter 108/121 - loss 0.03142229 - time (sec): 6.88 - samples/sec: 3253.78 - lr: 0.000017 - momentum: 0.000000
2023-10-17 11:09:02,084 epoch 7 - iter 120/121 - loss 0.03049685 - time (sec): 7.65 - samples/sec: 3219.99 - lr: 0.000017 - momentum: 0.000000
2023-10-17 11:09:02,130 ----------------------------------------------------------------------------------------------------
2023-10-17 11:09:02,130 EPOCH 7 done: loss 0.0306 - lr: 0.000017
2023-10-17 11:09:02,900 DEV : loss 0.1948603093624115 - f1-score (micro avg) 0.8331
2023-10-17 11:09:02,905 ----------------------------------------------------------------------------------------------------
2023-10-17 11:09:03,630 epoch 8 - iter 12/121 - loss 0.01798295 - time (sec): 0.72 - samples/sec: 3173.02 - lr: 0.000016 - momentum: 0.000000
2023-10-17 11:09:04,365 epoch 8 - iter 24/121 - loss 0.01653155 - time (sec): 1.46 - samples/sec: 3229.96 - lr: 0.000016 - momentum: 0.000000
2023-10-17 11:09:05,147 epoch 8 - iter 36/121 - loss 0.02239749 - time (sec): 2.24 - samples/sec: 3246.69 - lr: 0.000015 - momentum: 0.000000
2023-10-17 11:09:05,835 epoch 8 - iter 48/121 - loss 0.02224355 - time (sec): 2.93 - samples/sec: 3179.90 - lr: 0.000015 - momentum: 0.000000
2023-10-17 11:09:06,568 epoch 8 - iter 60/121 - loss 0.02209139 - time (sec): 3.66 - samples/sec: 3271.64 - lr: 0.000014 - momentum: 0.000000
2023-10-17 11:09:07,344 epoch 8 - iter 72/121 - loss 0.02267916 - time (sec): 4.44 - samples/sec: 3268.05 - lr: 0.000014 - momentum: 0.000000
2023-10-17 11:09:08,155 epoch 8 - iter 84/121 - loss 0.02026633 - time (sec): 5.25 - samples/sec: 3227.66 - lr: 0.000013 - momentum: 0.000000
2023-10-17 11:09:08,924 epoch 8 - iter 96/121 - loss 0.02139598 - time (sec): 6.02 - samples/sec: 3241.90 - lr: 0.000013 - momentum: 0.000000
2023-10-17 11:09:09,703 epoch 8 - iter 108/121 - loss 0.02001586 - time (sec): 6.80 - samples/sec: 3249.84 - lr: 0.000012 - momentum: 0.000000
2023-10-17 11:09:10,428 epoch 8 - iter 120/121 - loss 0.02118325 - time (sec): 7.52 - samples/sec: 3270.15 - lr: 0.000011 - momentum: 0.000000
2023-10-17 11:09:10,476 ----------------------------------------------------------------------------------------------------
2023-10-17 11:09:10,477 EPOCH 8 done: loss 0.0215 - lr: 0.000011
2023-10-17 11:09:11,407 DEV : loss 0.21927376091480255 - f1-score (micro avg) 0.8323
2023-10-17 11:09:11,412 ----------------------------------------------------------------------------------------------------
2023-10-17 11:09:12,122 epoch 9 - iter 12/121 - loss 0.02132408 - time (sec): 0.71 - samples/sec: 3315.28 - lr: 0.000011 - momentum: 0.000000
2023-10-17 11:09:12,847 epoch 9 - iter 24/121 - loss 0.01773470 - time (sec): 1.43 - samples/sec: 3278.97 - lr: 0.000010 - momentum: 0.000000
2023-10-17 11:09:13,566 epoch 9 - iter 36/121 - loss 0.02313221 - time (sec): 2.15 - samples/sec: 3188.50 - lr: 0.000010 - momentum: 0.000000
2023-10-17 11:09:14,319 epoch 9 - iter 48/121 - loss 0.01886954 - time (sec): 2.91 - samples/sec: 3251.85 - lr: 0.000009 - momentum: 0.000000
2023-10-17 11:09:15,023 epoch 9 - iter 60/121 - loss 0.01968467 - time (sec): 3.61 - samples/sec: 3228.26 - lr: 0.000009 - momentum: 0.000000
2023-10-17 11:09:15,732 epoch 9 - iter 72/121 - loss 0.02178143 - time (sec): 4.32 - samples/sec: 3249.54 - lr: 0.000008 - momentum: 0.000000
2023-10-17 11:09:16,520 epoch 9 - iter 84/121 - loss 0.02212080 - time (sec): 5.11 - samples/sec: 3265.04 - lr: 0.000008 - momentum: 0.000000
2023-10-17 11:09:17,344 epoch 9 - iter 96/121 - loss 0.02053703 - time (sec): 5.93 - samples/sec: 3266.72 - lr: 0.000007 - momentum: 0.000000
2023-10-17 11:09:18,120 epoch 9 - iter 108/121 - loss 0.01868999 - time (sec): 6.71 - samples/sec: 3277.35 - lr: 0.000006 - momentum: 0.000000
2023-10-17 11:09:18,877 epoch 9 - iter 120/121 - loss 0.01827614 - time (sec): 7.46 - samples/sec: 3284.29 - lr: 0.000006 - momentum: 0.000000
2023-10-17 11:09:18,939 ----------------------------------------------------------------------------------------------------
2023-10-17 11:09:18,939 EPOCH 9 done: loss 0.0183 - lr: 0.000006
2023-10-17 11:09:19,740 DEV : loss 0.22246934473514557 - f1-score (micro avg) 0.8319
2023-10-17 11:09:19,747 ----------------------------------------------------------------------------------------------------
2023-10-17 11:09:20,492 epoch 10 - iter 12/121 - loss 0.00591043 - time (sec): 0.74 - samples/sec: 3213.24 - lr: 0.000005 - momentum: 0.000000
2023-10-17 11:09:21,241 epoch 10 - iter 24/121 - loss 0.01457823 - time (sec): 1.49 - samples/sec: 3338.82 - lr: 0.000005 - momentum: 0.000000
2023-10-17 11:09:22,020 epoch 10 - iter 36/121 - loss 0.01630667 - time (sec): 2.27 - samples/sec: 3205.55 - lr: 0.000004 - momentum: 0.000000
2023-10-17 11:09:22,738 epoch 10 - iter 48/121 - loss 0.02028258 - time (sec): 2.99 - samples/sec: 3259.08 - lr: 0.000004 - momentum: 0.000000
2023-10-17 11:09:23,503 epoch 10 - iter 60/121 - loss 0.01793374 - time (sec): 3.76 - samples/sec: 3264.73 - lr: 0.000003 - momentum: 0.000000
2023-10-17 11:09:24,276 epoch 10 - iter 72/121 - loss 0.01632167 - time (sec): 4.53 - samples/sec: 3244.65 - lr: 0.000003 - momentum: 0.000000
2023-10-17 11:09:25,096 epoch 10 - iter 84/121 - loss 0.01470658 - time (sec): 5.35 - samples/sec: 3185.43 - lr: 0.000002 - momentum: 0.000000
2023-10-17 11:09:25,921 epoch 10 - iter 96/121 - loss 0.01418145 - time (sec): 6.17 - samples/sec: 3171.54 - lr: 0.000001 - momentum: 0.000000
2023-10-17 11:09:26,658 epoch 10 - iter 108/121 - loss 0.01327139 - time (sec): 6.91 - samples/sec: 3204.74 - lr: 0.000001 - momentum: 0.000000
2023-10-17 11:09:27,373 epoch 10 - iter 120/121 - loss 0.01330474 - time (sec): 7.62 - samples/sec: 3225.61 - lr: 0.000000 - momentum: 0.000000
2023-10-17 11:09:27,427 ----------------------------------------------------------------------------------------------------
2023-10-17 11:09:27,427 EPOCH 10 done: loss 0.0132 - lr: 0.000000
2023-10-17 11:09:28,181 DEV : loss 0.22670365869998932 - f1-score (micro avg) 0.8413
2023-10-17 11:09:28,186 saving best model
2023-10-17 11:09:29,078 ----------------------------------------------------------------------------------------------------
2023-10-17 11:09:29,079 Loading model from best epoch ...
2023-10-17 11:09:30,511 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 11:09:31,188
Results:
- F-score (micro) 0.8092
- F-score (macro) 0.5501
- Accuracy 0.6986
By class:
precision recall f1-score support
pers 0.8652 0.8777 0.8714 139
scope 0.8085 0.8837 0.8444 129
work 0.6489 0.7625 0.7011 80
loc 0.6667 0.2222 0.3333 9
date 0.0000 0.0000 0.0000 3
micro avg 0.7889 0.8306 0.8092 360
macro avg 0.5979 0.5492 0.5501 360
weighted avg 0.7847 0.8306 0.8032 360
2023-10-17 11:09:31,188 ----------------------------------------------------------------------------------------------------
|