File size: 23,904 Bytes
93a8540 |
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 10:59:11,153 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:11,154 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 10:59:11,154 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:11,155 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 10:59:11,155 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:11,155 Train: 966 sentences
2023-10-17 10:59:11,155 (train_with_dev=False, train_with_test=False)
2023-10-17 10:59:11,155 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:11,155 Training Params:
2023-10-17 10:59:11,155 - learning_rate: "5e-05"
2023-10-17 10:59:11,155 - mini_batch_size: "8"
2023-10-17 10:59:11,155 - max_epochs: "10"
2023-10-17 10:59:11,155 - shuffle: "True"
2023-10-17 10:59:11,155 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:11,155 Plugins:
2023-10-17 10:59:11,155 - TensorboardLogger
2023-10-17 10:59:11,155 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 10:59:11,155 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:11,155 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 10:59:11,155 - metric: "('micro avg', 'f1-score')"
2023-10-17 10:59:11,155 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:11,155 Computation:
2023-10-17 10:59:11,155 - compute on device: cuda:0
2023-10-17 10:59:11,155 - embedding storage: none
2023-10-17 10:59:11,155 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:11,155 Model training base path: "hmbench-ajmc/fr-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-17 10:59:11,155 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:11,155 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:11,156 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 10:59:11,922 epoch 1 - iter 12/121 - loss 4.03984930 - time (sec): 0.77 - samples/sec: 3147.08 - lr: 0.000005 - momentum: 0.000000
2023-10-17 10:59:12,693 epoch 1 - iter 24/121 - loss 3.48446989 - time (sec): 1.54 - samples/sec: 3092.14 - lr: 0.000010 - momentum: 0.000000
2023-10-17 10:59:13,455 epoch 1 - iter 36/121 - loss 2.67791785 - time (sec): 2.30 - samples/sec: 3076.89 - lr: 0.000014 - momentum: 0.000000
2023-10-17 10:59:14,250 epoch 1 - iter 48/121 - loss 2.12343318 - time (sec): 3.09 - samples/sec: 3155.57 - lr: 0.000019 - momentum: 0.000000
2023-10-17 10:59:14,963 epoch 1 - iter 60/121 - loss 1.83639393 - time (sec): 3.81 - samples/sec: 3164.40 - lr: 0.000024 - momentum: 0.000000
2023-10-17 10:59:15,774 epoch 1 - iter 72/121 - loss 1.59197735 - time (sec): 4.62 - samples/sec: 3166.88 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:59:16,559 epoch 1 - iter 84/121 - loss 1.38890453 - time (sec): 5.40 - samples/sec: 3203.74 - lr: 0.000034 - momentum: 0.000000
2023-10-17 10:59:17,384 epoch 1 - iter 96/121 - loss 1.23617388 - time (sec): 6.23 - samples/sec: 3230.26 - lr: 0.000039 - momentum: 0.000000
2023-10-17 10:59:18,103 epoch 1 - iter 108/121 - loss 1.14774163 - time (sec): 6.95 - samples/sec: 3218.01 - lr: 0.000044 - momentum: 0.000000
2023-10-17 10:59:18,839 epoch 1 - iter 120/121 - loss 1.06464850 - time (sec): 7.68 - samples/sec: 3203.69 - lr: 0.000049 - momentum: 0.000000
2023-10-17 10:59:18,889 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:18,889 EPOCH 1 done: loss 1.0617 - lr: 0.000049
2023-10-17 10:59:19,489 DEV : loss 0.1961672157049179 - f1-score (micro avg) 0.6743
2023-10-17 10:59:19,496 saving best model
2023-10-17 10:59:19,852 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:20,572 epoch 2 - iter 12/121 - loss 0.15827254 - time (sec): 0.72 - samples/sec: 3192.69 - lr: 0.000049 - momentum: 0.000000
2023-10-17 10:59:21,293 epoch 2 - iter 24/121 - loss 0.17582748 - time (sec): 1.44 - samples/sec: 3244.43 - lr: 0.000049 - momentum: 0.000000
2023-10-17 10:59:22,031 epoch 2 - iter 36/121 - loss 0.17655963 - time (sec): 2.18 - samples/sec: 3262.17 - lr: 0.000048 - momentum: 0.000000
2023-10-17 10:59:22,756 epoch 2 - iter 48/121 - loss 0.17537386 - time (sec): 2.90 - samples/sec: 3279.04 - lr: 0.000048 - momentum: 0.000000
2023-10-17 10:59:23,488 epoch 2 - iter 60/121 - loss 0.17240823 - time (sec): 3.63 - samples/sec: 3367.13 - lr: 0.000047 - momentum: 0.000000
2023-10-17 10:59:24,179 epoch 2 - iter 72/121 - loss 0.16952158 - time (sec): 4.33 - samples/sec: 3361.95 - lr: 0.000047 - momentum: 0.000000
2023-10-17 10:59:24,925 epoch 2 - iter 84/121 - loss 0.17046832 - time (sec): 5.07 - samples/sec: 3379.50 - lr: 0.000046 - momentum: 0.000000
2023-10-17 10:59:25,682 epoch 2 - iter 96/121 - loss 0.16917879 - time (sec): 5.83 - samples/sec: 3376.28 - lr: 0.000046 - momentum: 0.000000
2023-10-17 10:59:26,409 epoch 2 - iter 108/121 - loss 0.17347387 - time (sec): 6.56 - samples/sec: 3375.46 - lr: 0.000045 - momentum: 0.000000
2023-10-17 10:59:27,143 epoch 2 - iter 120/121 - loss 0.16878218 - time (sec): 7.29 - samples/sec: 3369.89 - lr: 0.000045 - momentum: 0.000000
2023-10-17 10:59:27,212 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:27,212 EPOCH 2 done: loss 0.1677 - lr: 0.000045
2023-10-17 10:59:27,952 DEV : loss 0.1310185343027115 - f1-score (micro avg) 0.7902
2023-10-17 10:59:27,957 saving best model
2023-10-17 10:59:28,419 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:29,178 epoch 3 - iter 12/121 - loss 0.10826741 - time (sec): 0.76 - samples/sec: 3195.99 - lr: 0.000044 - momentum: 0.000000
2023-10-17 10:59:29,975 epoch 3 - iter 24/121 - loss 0.10014281 - time (sec): 1.55 - samples/sec: 3198.69 - lr: 0.000043 - momentum: 0.000000
2023-10-17 10:59:30,789 epoch 3 - iter 36/121 - loss 0.08852980 - time (sec): 2.37 - samples/sec: 3253.47 - lr: 0.000043 - momentum: 0.000000
2023-10-17 10:59:31,523 epoch 3 - iter 48/121 - loss 0.09054614 - time (sec): 3.10 - samples/sec: 3269.07 - lr: 0.000042 - momentum: 0.000000
2023-10-17 10:59:32,308 epoch 3 - iter 60/121 - loss 0.09464846 - time (sec): 3.89 - samples/sec: 3230.95 - lr: 0.000042 - momentum: 0.000000
2023-10-17 10:59:33,063 epoch 3 - iter 72/121 - loss 0.09297356 - time (sec): 4.64 - samples/sec: 3236.86 - lr: 0.000041 - momentum: 0.000000
2023-10-17 10:59:33,835 epoch 3 - iter 84/121 - loss 0.09444515 - time (sec): 5.41 - samples/sec: 3252.57 - lr: 0.000041 - momentum: 0.000000
2023-10-17 10:59:34,496 epoch 3 - iter 96/121 - loss 0.09242019 - time (sec): 6.07 - samples/sec: 3222.44 - lr: 0.000040 - momentum: 0.000000
2023-10-17 10:59:35,321 epoch 3 - iter 108/121 - loss 0.09432936 - time (sec): 6.90 - samples/sec: 3234.89 - lr: 0.000040 - momentum: 0.000000
2023-10-17 10:59:36,015 epoch 3 - iter 120/121 - loss 0.09658269 - time (sec): 7.59 - samples/sec: 3233.86 - lr: 0.000039 - momentum: 0.000000
2023-10-17 10:59:36,063 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:36,063 EPOCH 3 done: loss 0.0974 - lr: 0.000039
2023-10-17 10:59:36,968 DEV : loss 0.1393728405237198 - f1-score (micro avg) 0.7876
2023-10-17 10:59:36,973 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:37,680 epoch 4 - iter 12/121 - loss 0.08865884 - time (sec): 0.71 - samples/sec: 2936.04 - lr: 0.000038 - momentum: 0.000000
2023-10-17 10:59:38,402 epoch 4 - iter 24/121 - loss 0.08219893 - time (sec): 1.43 - samples/sec: 3193.06 - lr: 0.000038 - momentum: 0.000000
2023-10-17 10:59:39,132 epoch 4 - iter 36/121 - loss 0.07077621 - time (sec): 2.16 - samples/sec: 3224.72 - lr: 0.000037 - momentum: 0.000000
2023-10-17 10:59:39,924 epoch 4 - iter 48/121 - loss 0.07382872 - time (sec): 2.95 - samples/sec: 3239.62 - lr: 0.000037 - momentum: 0.000000
2023-10-17 10:59:40,673 epoch 4 - iter 60/121 - loss 0.07391026 - time (sec): 3.70 - samples/sec: 3303.16 - lr: 0.000036 - momentum: 0.000000
2023-10-17 10:59:41,351 epoch 4 - iter 72/121 - loss 0.07389219 - time (sec): 4.38 - samples/sec: 3292.30 - lr: 0.000036 - momentum: 0.000000
2023-10-17 10:59:42,120 epoch 4 - iter 84/121 - loss 0.07656404 - time (sec): 5.15 - samples/sec: 3267.91 - lr: 0.000035 - momentum: 0.000000
2023-10-17 10:59:42,924 epoch 4 - iter 96/121 - loss 0.07249895 - time (sec): 5.95 - samples/sec: 3252.09 - lr: 0.000035 - momentum: 0.000000
2023-10-17 10:59:43,744 epoch 4 - iter 108/121 - loss 0.07277711 - time (sec): 6.77 - samples/sec: 3235.49 - lr: 0.000034 - momentum: 0.000000
2023-10-17 10:59:44,483 epoch 4 - iter 120/121 - loss 0.07000959 - time (sec): 7.51 - samples/sec: 3279.76 - lr: 0.000034 - momentum: 0.000000
2023-10-17 10:59:44,536 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:44,537 EPOCH 4 done: loss 0.0697 - lr: 0.000034
2023-10-17 10:59:45,299 DEV : loss 0.14459875226020813 - f1-score (micro avg) 0.8271
2023-10-17 10:59:45,304 saving best model
2023-10-17 10:59:45,866 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:46,629 epoch 5 - iter 12/121 - loss 0.02887476 - time (sec): 0.76 - samples/sec: 2855.60 - lr: 0.000033 - momentum: 0.000000
2023-10-17 10:59:47,382 epoch 5 - iter 24/121 - loss 0.04306695 - time (sec): 1.51 - samples/sec: 2936.78 - lr: 0.000032 - momentum: 0.000000
2023-10-17 10:59:48,156 epoch 5 - iter 36/121 - loss 0.04116509 - time (sec): 2.29 - samples/sec: 3078.29 - lr: 0.000032 - momentum: 0.000000
2023-10-17 10:59:48,932 epoch 5 - iter 48/121 - loss 0.04444591 - time (sec): 3.06 - samples/sec: 3059.29 - lr: 0.000031 - momentum: 0.000000
2023-10-17 10:59:49,642 epoch 5 - iter 60/121 - loss 0.04296924 - time (sec): 3.77 - samples/sec: 3119.81 - lr: 0.000031 - momentum: 0.000000
2023-10-17 10:59:50,439 epoch 5 - iter 72/121 - loss 0.04428424 - time (sec): 4.57 - samples/sec: 3190.03 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:59:51,138 epoch 5 - iter 84/121 - loss 0.04849569 - time (sec): 5.27 - samples/sec: 3198.35 - lr: 0.000030 - momentum: 0.000000
2023-10-17 10:59:51,883 epoch 5 - iter 96/121 - loss 0.04770451 - time (sec): 6.01 - samples/sec: 3216.01 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:59:52,654 epoch 5 - iter 108/121 - loss 0.04792742 - time (sec): 6.78 - samples/sec: 3223.52 - lr: 0.000029 - momentum: 0.000000
2023-10-17 10:59:53,460 epoch 5 - iter 120/121 - loss 0.04620569 - time (sec): 7.59 - samples/sec: 3247.60 - lr: 0.000028 - momentum: 0.000000
2023-10-17 10:59:53,506 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:53,506 EPOCH 5 done: loss 0.0463 - lr: 0.000028
2023-10-17 10:59:54,274 DEV : loss 0.17430146038532257 - f1-score (micro avg) 0.8213
2023-10-17 10:59:54,279 ----------------------------------------------------------------------------------------------------
2023-10-17 10:59:55,044 epoch 6 - iter 12/121 - loss 0.01847800 - time (sec): 0.76 - samples/sec: 3208.68 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:59:55,824 epoch 6 - iter 24/121 - loss 0.03217283 - time (sec): 1.54 - samples/sec: 3204.69 - lr: 0.000027 - momentum: 0.000000
2023-10-17 10:59:56,586 epoch 6 - iter 36/121 - loss 0.02784672 - time (sec): 2.31 - samples/sec: 3258.56 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:59:57,301 epoch 6 - iter 48/121 - loss 0.03009212 - time (sec): 3.02 - samples/sec: 3219.73 - lr: 0.000026 - momentum: 0.000000
2023-10-17 10:59:58,064 epoch 6 - iter 60/121 - loss 0.02992772 - time (sec): 3.78 - samples/sec: 3228.02 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:59:58,730 epoch 6 - iter 72/121 - loss 0.03460899 - time (sec): 4.45 - samples/sec: 3206.50 - lr: 0.000025 - momentum: 0.000000
2023-10-17 10:59:59,551 epoch 6 - iter 84/121 - loss 0.03594070 - time (sec): 5.27 - samples/sec: 3236.97 - lr: 0.000024 - momentum: 0.000000
2023-10-17 11:00:00,296 epoch 6 - iter 96/121 - loss 0.03447924 - time (sec): 6.02 - samples/sec: 3248.21 - lr: 0.000024 - momentum: 0.000000
2023-10-17 11:00:01,069 epoch 6 - iter 108/121 - loss 0.03462584 - time (sec): 6.79 - samples/sec: 3230.49 - lr: 0.000023 - momentum: 0.000000
2023-10-17 11:00:01,843 epoch 6 - iter 120/121 - loss 0.03552862 - time (sec): 7.56 - samples/sec: 3248.07 - lr: 0.000022 - momentum: 0.000000
2023-10-17 11:00:01,896 ----------------------------------------------------------------------------------------------------
2023-10-17 11:00:01,896 EPOCH 6 done: loss 0.0353 - lr: 0.000022
2023-10-17 11:00:02,659 DEV : loss 0.18103647232055664 - f1-score (micro avg) 0.8313
2023-10-17 11:00:02,665 saving best model
2023-10-17 11:00:03,141 ----------------------------------------------------------------------------------------------------
2023-10-17 11:00:03,919 epoch 7 - iter 12/121 - loss 0.00538440 - time (sec): 0.78 - samples/sec: 3084.81 - lr: 0.000022 - momentum: 0.000000
2023-10-17 11:00:04,636 epoch 7 - iter 24/121 - loss 0.01679636 - time (sec): 1.49 - samples/sec: 3120.00 - lr: 0.000021 - momentum: 0.000000
2023-10-17 11:00:05,357 epoch 7 - iter 36/121 - loss 0.01730532 - time (sec): 2.21 - samples/sec: 3266.30 - lr: 0.000021 - momentum: 0.000000
2023-10-17 11:00:06,130 epoch 7 - iter 48/121 - loss 0.01620835 - time (sec): 2.99 - samples/sec: 3235.73 - lr: 0.000020 - momentum: 0.000000
2023-10-17 11:00:06,936 epoch 7 - iter 60/121 - loss 0.01692553 - time (sec): 3.79 - samples/sec: 3232.05 - lr: 0.000020 - momentum: 0.000000
2023-10-17 11:00:07,684 epoch 7 - iter 72/121 - loss 0.01923634 - time (sec): 4.54 - samples/sec: 3210.04 - lr: 0.000019 - momentum: 0.000000
2023-10-17 11:00:08,439 epoch 7 - iter 84/121 - loss 0.01934951 - time (sec): 5.30 - samples/sec: 3195.12 - lr: 0.000019 - momentum: 0.000000
2023-10-17 11:00:09,176 epoch 7 - iter 96/121 - loss 0.01857928 - time (sec): 6.03 - samples/sec: 3222.90 - lr: 0.000018 - momentum: 0.000000
2023-10-17 11:00:09,963 epoch 7 - iter 108/121 - loss 0.02199051 - time (sec): 6.82 - samples/sec: 3232.32 - lr: 0.000017 - momentum: 0.000000
2023-10-17 11:00:10,726 epoch 7 - iter 120/121 - loss 0.02166981 - time (sec): 7.58 - samples/sec: 3249.58 - lr: 0.000017 - momentum: 0.000000
2023-10-17 11:00:10,778 ----------------------------------------------------------------------------------------------------
2023-10-17 11:00:10,779 EPOCH 7 done: loss 0.0216 - lr: 0.000017
2023-10-17 11:00:11,528 DEV : loss 0.21778880059719086 - f1-score (micro avg) 0.8219
2023-10-17 11:00:11,533 ----------------------------------------------------------------------------------------------------
2023-10-17 11:00:12,264 epoch 8 - iter 12/121 - loss 0.02496822 - time (sec): 0.73 - samples/sec: 2937.07 - lr: 0.000016 - momentum: 0.000000
2023-10-17 11:00:13,009 epoch 8 - iter 24/121 - loss 0.02109802 - time (sec): 1.48 - samples/sec: 3251.41 - lr: 0.000016 - momentum: 0.000000
2023-10-17 11:00:13,747 epoch 8 - iter 36/121 - loss 0.01936223 - time (sec): 2.21 - samples/sec: 3235.36 - lr: 0.000015 - momentum: 0.000000
2023-10-17 11:00:14,553 epoch 8 - iter 48/121 - loss 0.01994877 - time (sec): 3.02 - samples/sec: 3318.25 - lr: 0.000015 - momentum: 0.000000
2023-10-17 11:00:15,315 epoch 8 - iter 60/121 - loss 0.01807436 - time (sec): 3.78 - samples/sec: 3314.06 - lr: 0.000014 - momentum: 0.000000
2023-10-17 11:00:16,041 epoch 8 - iter 72/121 - loss 0.01570418 - time (sec): 4.51 - samples/sec: 3324.75 - lr: 0.000014 - momentum: 0.000000
2023-10-17 11:00:16,740 epoch 8 - iter 84/121 - loss 0.01662327 - time (sec): 5.21 - samples/sec: 3297.61 - lr: 0.000013 - momentum: 0.000000
2023-10-17 11:00:17,484 epoch 8 - iter 96/121 - loss 0.01642530 - time (sec): 5.95 - samples/sec: 3299.09 - lr: 0.000013 - momentum: 0.000000
2023-10-17 11:00:18,233 epoch 8 - iter 108/121 - loss 0.01566260 - time (sec): 6.70 - samples/sec: 3321.79 - lr: 0.000012 - momentum: 0.000000
2023-10-17 11:00:19,000 epoch 8 - iter 120/121 - loss 0.01682453 - time (sec): 7.47 - samples/sec: 3296.10 - lr: 0.000011 - momentum: 0.000000
2023-10-17 11:00:19,049 ----------------------------------------------------------------------------------------------------
2023-10-17 11:00:19,049 EPOCH 8 done: loss 0.0168 - lr: 0.000011
2023-10-17 11:00:19,793 DEV : loss 0.20616522431373596 - f1-score (micro avg) 0.8471
2023-10-17 11:00:19,798 saving best model
2023-10-17 11:00:20,275 ----------------------------------------------------------------------------------------------------
2023-10-17 11:00:20,996 epoch 9 - iter 12/121 - loss 0.00858267 - time (sec): 0.72 - samples/sec: 3485.09 - lr: 0.000011 - momentum: 0.000000
2023-10-17 11:00:21,740 epoch 9 - iter 24/121 - loss 0.00850359 - time (sec): 1.46 - samples/sec: 3284.45 - lr: 0.000010 - momentum: 0.000000
2023-10-17 11:00:22,497 epoch 9 - iter 36/121 - loss 0.00987893 - time (sec): 2.22 - samples/sec: 3161.64 - lr: 0.000010 - momentum: 0.000000
2023-10-17 11:00:23,238 epoch 9 - iter 48/121 - loss 0.01135571 - time (sec): 2.96 - samples/sec: 3149.06 - lr: 0.000009 - momentum: 0.000000
2023-10-17 11:00:23,944 epoch 9 - iter 60/121 - loss 0.01031270 - time (sec): 3.66 - samples/sec: 3157.12 - lr: 0.000009 - momentum: 0.000000
2023-10-17 11:00:24,698 epoch 9 - iter 72/121 - loss 0.01012589 - time (sec): 4.42 - samples/sec: 3221.68 - lr: 0.000008 - momentum: 0.000000
2023-10-17 11:00:25,457 epoch 9 - iter 84/121 - loss 0.01006348 - time (sec): 5.18 - samples/sec: 3249.12 - lr: 0.000008 - momentum: 0.000000
2023-10-17 11:00:26,209 epoch 9 - iter 96/121 - loss 0.00959124 - time (sec): 5.93 - samples/sec: 3289.50 - lr: 0.000007 - momentum: 0.000000
2023-10-17 11:00:27,012 epoch 9 - iter 108/121 - loss 0.01110604 - time (sec): 6.73 - samples/sec: 3297.45 - lr: 0.000006 - momentum: 0.000000
2023-10-17 11:00:27,767 epoch 9 - iter 120/121 - loss 0.01135045 - time (sec): 7.49 - samples/sec: 3285.16 - lr: 0.000006 - momentum: 0.000000
2023-10-17 11:00:27,818 ----------------------------------------------------------------------------------------------------
2023-10-17 11:00:27,818 EPOCH 9 done: loss 0.0113 - lr: 0.000006
2023-10-17 11:00:28,558 DEV : loss 0.22430478036403656 - f1-score (micro avg) 0.8344
2023-10-17 11:00:28,563 ----------------------------------------------------------------------------------------------------
2023-10-17 11:00:29,271 epoch 10 - iter 12/121 - loss 0.00114275 - time (sec): 0.71 - samples/sec: 3349.89 - lr: 0.000005 - momentum: 0.000000
2023-10-17 11:00:30,014 epoch 10 - iter 24/121 - loss 0.00352745 - time (sec): 1.45 - samples/sec: 3148.72 - lr: 0.000005 - momentum: 0.000000
2023-10-17 11:00:30,762 epoch 10 - iter 36/121 - loss 0.00508865 - time (sec): 2.20 - samples/sec: 3258.73 - lr: 0.000004 - momentum: 0.000000
2023-10-17 11:00:31,488 epoch 10 - iter 48/121 - loss 0.00574434 - time (sec): 2.92 - samples/sec: 3315.20 - lr: 0.000004 - momentum: 0.000000
2023-10-17 11:00:32,294 epoch 10 - iter 60/121 - loss 0.01005490 - time (sec): 3.73 - samples/sec: 3408.82 - lr: 0.000003 - momentum: 0.000000
2023-10-17 11:00:33,000 epoch 10 - iter 72/121 - loss 0.00878139 - time (sec): 4.44 - samples/sec: 3351.85 - lr: 0.000003 - momentum: 0.000000
2023-10-17 11:00:33,751 epoch 10 - iter 84/121 - loss 0.00826920 - time (sec): 5.19 - samples/sec: 3290.92 - lr: 0.000002 - momentum: 0.000000
2023-10-17 11:00:34,538 epoch 10 - iter 96/121 - loss 0.00749286 - time (sec): 5.97 - samples/sec: 3298.47 - lr: 0.000001 - momentum: 0.000000
2023-10-17 11:00:35,317 epoch 10 - iter 108/121 - loss 0.00742557 - time (sec): 6.75 - samples/sec: 3321.00 - lr: 0.000001 - momentum: 0.000000
2023-10-17 11:00:36,004 epoch 10 - iter 120/121 - loss 0.00724469 - time (sec): 7.44 - samples/sec: 3303.38 - lr: 0.000000 - momentum: 0.000000
2023-10-17 11:00:36,061 ----------------------------------------------------------------------------------------------------
2023-10-17 11:00:36,062 EPOCH 10 done: loss 0.0072 - lr: 0.000000
2023-10-17 11:00:36,815 DEV : loss 0.2359585165977478 - f1-score (micro avg) 0.8375
2023-10-17 11:00:37,205 ----------------------------------------------------------------------------------------------------
2023-10-17 11:00:37,206 Loading model from best epoch ...
2023-10-17 11:00:38,619 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 11:00:39,465
Results:
- F-score (micro) 0.8174
- F-score (macro) 0.5799
- Accuracy 0.7075
By class:
precision recall f1-score support
pers 0.8652 0.8777 0.8714 139
scope 0.8321 0.8837 0.8571 129
work 0.6630 0.7625 0.7093 80
loc 0.7500 0.3333 0.4615 9
date 0.0000 0.0000 0.0000 3
micro avg 0.8021 0.8333 0.8174 360
macro avg 0.6221 0.5715 0.5799 360
weighted avg 0.7984 0.8333 0.8128 360
2023-10-17 11:00:39,466 ----------------------------------------------------------------------------------------------------
|