File size: 24,113 Bytes
11e83cf |
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 |
2023-10-17 22:29:04,133 ----------------------------------------------------------------------------------------------------
2023-10-17 22:29:04,134 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=21, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-17 22:29:04,134 ----------------------------------------------------------------------------------------------------
2023-10-17 22:29:04,135 MultiCorpus: 5901 train + 1287 dev + 1505 test sentences
- NER_HIPE_2022 Corpus: 5901 train + 1287 dev + 1505 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/hipe2020/fr/with_doc_seperator
2023-10-17 22:29:04,135 ----------------------------------------------------------------------------------------------------
2023-10-17 22:29:04,135 Train: 5901 sentences
2023-10-17 22:29:04,135 (train_with_dev=False, train_with_test=False)
2023-10-17 22:29:04,135 ----------------------------------------------------------------------------------------------------
2023-10-17 22:29:04,135 Training Params:
2023-10-17 22:29:04,135 - learning_rate: "5e-05"
2023-10-17 22:29:04,135 - mini_batch_size: "8"
2023-10-17 22:29:04,135 - max_epochs: "10"
2023-10-17 22:29:04,135 - shuffle: "True"
2023-10-17 22:29:04,135 ----------------------------------------------------------------------------------------------------
2023-10-17 22:29:04,135 Plugins:
2023-10-17 22:29:04,135 - TensorboardLogger
2023-10-17 22:29:04,135 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 22:29:04,135 ----------------------------------------------------------------------------------------------------
2023-10-17 22:29:04,135 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 22:29:04,135 - metric: "('micro avg', 'f1-score')"
2023-10-17 22:29:04,135 ----------------------------------------------------------------------------------------------------
2023-10-17 22:29:04,135 Computation:
2023-10-17 22:29:04,135 - compute on device: cuda:0
2023-10-17 22:29:04,135 - embedding storage: none
2023-10-17 22:29:04,135 ----------------------------------------------------------------------------------------------------
2023-10-17 22:29:04,135 Model training base path: "hmbench-hipe2020/fr-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-4"
2023-10-17 22:29:04,135 ----------------------------------------------------------------------------------------------------
2023-10-17 22:29:04,135 ----------------------------------------------------------------------------------------------------
2023-10-17 22:29:04,135 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 22:29:09,478 epoch 1 - iter 73/738 - loss 2.99837415 - time (sec): 5.34 - samples/sec: 3166.80 - lr: 0.000005 - momentum: 0.000000
2023-10-17 22:29:13,992 epoch 1 - iter 146/738 - loss 1.93510546 - time (sec): 9.86 - samples/sec: 3213.05 - lr: 0.000010 - momentum: 0.000000
2023-10-17 22:29:18,605 epoch 1 - iter 219/738 - loss 1.45592106 - time (sec): 14.47 - samples/sec: 3237.90 - lr: 0.000015 - momentum: 0.000000
2023-10-17 22:29:23,729 epoch 1 - iter 292/738 - loss 1.16902073 - time (sec): 19.59 - samples/sec: 3260.44 - lr: 0.000020 - momentum: 0.000000
2023-10-17 22:29:29,137 epoch 1 - iter 365/738 - loss 0.97478204 - time (sec): 25.00 - samples/sec: 3288.96 - lr: 0.000025 - momentum: 0.000000
2023-10-17 22:29:34,394 epoch 1 - iter 438/738 - loss 0.85064874 - time (sec): 30.26 - samples/sec: 3282.58 - lr: 0.000030 - momentum: 0.000000
2023-10-17 22:29:38,954 epoch 1 - iter 511/738 - loss 0.76809817 - time (sec): 34.82 - samples/sec: 3284.37 - lr: 0.000035 - momentum: 0.000000
2023-10-17 22:29:43,774 epoch 1 - iter 584/738 - loss 0.69380777 - time (sec): 39.64 - samples/sec: 3297.78 - lr: 0.000039 - momentum: 0.000000
2023-10-17 22:29:48,885 epoch 1 - iter 657/738 - loss 0.63363005 - time (sec): 44.75 - samples/sec: 3296.16 - lr: 0.000044 - momentum: 0.000000
2023-10-17 22:29:53,841 epoch 1 - iter 730/738 - loss 0.58383020 - time (sec): 49.70 - samples/sec: 3320.47 - lr: 0.000049 - momentum: 0.000000
2023-10-17 22:29:54,280 ----------------------------------------------------------------------------------------------------
2023-10-17 22:29:54,280 EPOCH 1 done: loss 0.5810 - lr: 0.000049
2023-10-17 22:30:00,666 DEV : loss 0.11302945762872696 - f1-score (micro avg) 0.7588
2023-10-17 22:30:00,696 saving best model
2023-10-17 22:30:01,082 ----------------------------------------------------------------------------------------------------
2023-10-17 22:30:06,731 epoch 2 - iter 73/738 - loss 0.13556365 - time (sec): 5.65 - samples/sec: 3264.65 - lr: 0.000049 - momentum: 0.000000
2023-10-17 22:30:11,564 epoch 2 - iter 146/738 - loss 0.12924036 - time (sec): 10.48 - samples/sec: 3272.73 - lr: 0.000049 - momentum: 0.000000
2023-10-17 22:30:16,659 epoch 2 - iter 219/738 - loss 0.12904174 - time (sec): 15.58 - samples/sec: 3216.43 - lr: 0.000048 - momentum: 0.000000
2023-10-17 22:30:21,829 epoch 2 - iter 292/738 - loss 0.12726226 - time (sec): 20.75 - samples/sec: 3213.20 - lr: 0.000048 - momentum: 0.000000
2023-10-17 22:30:26,397 epoch 2 - iter 365/738 - loss 0.12773492 - time (sec): 25.31 - samples/sec: 3210.16 - lr: 0.000047 - momentum: 0.000000
2023-10-17 22:30:31,342 epoch 2 - iter 438/738 - loss 0.12584037 - time (sec): 30.26 - samples/sec: 3214.18 - lr: 0.000047 - momentum: 0.000000
2023-10-17 22:30:35,893 epoch 2 - iter 511/738 - loss 0.12575633 - time (sec): 34.81 - samples/sec: 3231.66 - lr: 0.000046 - momentum: 0.000000
2023-10-17 22:30:40,717 epoch 2 - iter 584/738 - loss 0.12360271 - time (sec): 39.63 - samples/sec: 3238.30 - lr: 0.000046 - momentum: 0.000000
2023-10-17 22:30:45,964 epoch 2 - iter 657/738 - loss 0.12251002 - time (sec): 44.88 - samples/sec: 3232.19 - lr: 0.000045 - momentum: 0.000000
2023-10-17 22:30:51,263 epoch 2 - iter 730/738 - loss 0.12156812 - time (sec): 50.18 - samples/sec: 3246.27 - lr: 0.000045 - momentum: 0.000000
2023-10-17 22:30:52,229 ----------------------------------------------------------------------------------------------------
2023-10-17 22:30:52,230 EPOCH 2 done: loss 0.1202 - lr: 0.000045
2023-10-17 22:31:04,398 DEV : loss 0.10620676726102829 - f1-score (micro avg) 0.7916
2023-10-17 22:31:04,432 saving best model
2023-10-17 22:31:04,975 ----------------------------------------------------------------------------------------------------
2023-10-17 22:31:10,816 epoch 3 - iter 73/738 - loss 0.07909692 - time (sec): 5.83 - samples/sec: 3171.41 - lr: 0.000044 - momentum: 0.000000
2023-10-17 22:31:15,736 epoch 3 - iter 146/738 - loss 0.07174765 - time (sec): 10.76 - samples/sec: 3220.31 - lr: 0.000043 - momentum: 0.000000
2023-10-17 22:31:20,225 epoch 3 - iter 219/738 - loss 0.07516061 - time (sec): 15.24 - samples/sec: 3229.49 - lr: 0.000043 - momentum: 0.000000
2023-10-17 22:31:25,006 epoch 3 - iter 292/738 - loss 0.07592855 - time (sec): 20.02 - samples/sec: 3258.50 - lr: 0.000042 - momentum: 0.000000
2023-10-17 22:31:29,791 epoch 3 - iter 365/738 - loss 0.07330696 - time (sec): 24.81 - samples/sec: 3265.19 - lr: 0.000042 - momentum: 0.000000
2023-10-17 22:31:35,103 epoch 3 - iter 438/738 - loss 0.07170697 - time (sec): 30.12 - samples/sec: 3259.41 - lr: 0.000041 - momentum: 0.000000
2023-10-17 22:31:40,196 epoch 3 - iter 511/738 - loss 0.07160535 - time (sec): 35.21 - samples/sec: 3281.25 - lr: 0.000041 - momentum: 0.000000
2023-10-17 22:31:44,916 epoch 3 - iter 584/738 - loss 0.07165988 - time (sec): 39.93 - samples/sec: 3271.92 - lr: 0.000040 - momentum: 0.000000
2023-10-17 22:31:50,041 epoch 3 - iter 657/738 - loss 0.07097376 - time (sec): 45.06 - samples/sec: 3262.94 - lr: 0.000040 - momentum: 0.000000
2023-10-17 22:31:55,409 epoch 3 - iter 730/738 - loss 0.06940338 - time (sec): 50.43 - samples/sec: 3262.11 - lr: 0.000039 - momentum: 0.000000
2023-10-17 22:31:55,939 ----------------------------------------------------------------------------------------------------
2023-10-17 22:31:55,939 EPOCH 3 done: loss 0.0693 - lr: 0.000039
2023-10-17 22:32:07,887 DEV : loss 0.13108941912651062 - f1-score (micro avg) 0.8242
2023-10-17 22:32:07,925 saving best model
2023-10-17 22:32:08,507 ----------------------------------------------------------------------------------------------------
2023-10-17 22:32:13,135 epoch 4 - iter 73/738 - loss 0.03703471 - time (sec): 4.63 - samples/sec: 3340.02 - lr: 0.000038 - momentum: 0.000000
2023-10-17 22:32:18,045 epoch 4 - iter 146/738 - loss 0.04150633 - time (sec): 9.54 - samples/sec: 3270.43 - lr: 0.000038 - momentum: 0.000000
2023-10-17 22:32:23,041 epoch 4 - iter 219/738 - loss 0.04123992 - time (sec): 14.53 - samples/sec: 3241.31 - lr: 0.000037 - momentum: 0.000000
2023-10-17 22:32:29,054 epoch 4 - iter 292/738 - loss 0.04410608 - time (sec): 20.54 - samples/sec: 3232.21 - lr: 0.000037 - momentum: 0.000000
2023-10-17 22:32:34,235 epoch 4 - iter 365/738 - loss 0.04659042 - time (sec): 25.73 - samples/sec: 3244.52 - lr: 0.000036 - momentum: 0.000000
2023-10-17 22:32:38,717 epoch 4 - iter 438/738 - loss 0.04619151 - time (sec): 30.21 - samples/sec: 3250.55 - lr: 0.000036 - momentum: 0.000000
2023-10-17 22:32:43,578 epoch 4 - iter 511/738 - loss 0.04635101 - time (sec): 35.07 - samples/sec: 3264.21 - lr: 0.000035 - momentum: 0.000000
2023-10-17 22:32:48,576 epoch 4 - iter 584/738 - loss 0.04536034 - time (sec): 40.07 - samples/sec: 3252.74 - lr: 0.000035 - momentum: 0.000000
2023-10-17 22:32:53,408 epoch 4 - iter 657/738 - loss 0.04699631 - time (sec): 44.90 - samples/sec: 3255.70 - lr: 0.000034 - momentum: 0.000000
2023-10-17 22:32:59,545 epoch 4 - iter 730/738 - loss 0.04985825 - time (sec): 51.04 - samples/sec: 3227.83 - lr: 0.000033 - momentum: 0.000000
2023-10-17 22:33:00,140 ----------------------------------------------------------------------------------------------------
2023-10-17 22:33:00,141 EPOCH 4 done: loss 0.0499 - lr: 0.000033
2023-10-17 22:33:11,846 DEV : loss 0.14125068485736847 - f1-score (micro avg) 0.815
2023-10-17 22:33:11,880 ----------------------------------------------------------------------------------------------------
2023-10-17 22:33:16,573 epoch 5 - iter 73/738 - loss 0.02879190 - time (sec): 4.69 - samples/sec: 3375.41 - lr: 0.000033 - momentum: 0.000000
2023-10-17 22:33:21,534 epoch 5 - iter 146/738 - loss 0.03371840 - time (sec): 9.65 - samples/sec: 3230.98 - lr: 0.000032 - momentum: 0.000000
2023-10-17 22:33:26,804 epoch 5 - iter 219/738 - loss 0.02945090 - time (sec): 14.92 - samples/sec: 3219.82 - lr: 0.000032 - momentum: 0.000000
2023-10-17 22:33:32,072 epoch 5 - iter 292/738 - loss 0.03027760 - time (sec): 20.19 - samples/sec: 3232.69 - lr: 0.000031 - momentum: 0.000000
2023-10-17 22:33:37,069 epoch 5 - iter 365/738 - loss 0.03506228 - time (sec): 25.19 - samples/sec: 3268.76 - lr: 0.000031 - momentum: 0.000000
2023-10-17 22:33:42,047 epoch 5 - iter 438/738 - loss 0.03543225 - time (sec): 30.17 - samples/sec: 3282.24 - lr: 0.000030 - momentum: 0.000000
2023-10-17 22:33:46,992 epoch 5 - iter 511/738 - loss 0.03479354 - time (sec): 35.11 - samples/sec: 3270.66 - lr: 0.000030 - momentum: 0.000000
2023-10-17 22:33:52,994 epoch 5 - iter 584/738 - loss 0.03639606 - time (sec): 41.11 - samples/sec: 3261.04 - lr: 0.000029 - momentum: 0.000000
2023-10-17 22:33:57,710 epoch 5 - iter 657/738 - loss 0.03555295 - time (sec): 45.83 - samples/sec: 3261.10 - lr: 0.000028 - momentum: 0.000000
2023-10-17 22:34:02,438 epoch 5 - iter 730/738 - loss 0.03486226 - time (sec): 50.56 - samples/sec: 3265.54 - lr: 0.000028 - momentum: 0.000000
2023-10-17 22:34:02,901 ----------------------------------------------------------------------------------------------------
2023-10-17 22:34:02,901 EPOCH 5 done: loss 0.0347 - lr: 0.000028
2023-10-17 22:34:15,042 DEV : loss 0.19004222750663757 - f1-score (micro avg) 0.8312
2023-10-17 22:34:15,088 saving best model
2023-10-17 22:34:15,573 ----------------------------------------------------------------------------------------------------
2023-10-17 22:34:20,441 epoch 6 - iter 73/738 - loss 0.02813745 - time (sec): 4.86 - samples/sec: 3296.59 - lr: 0.000027 - momentum: 0.000000
2023-10-17 22:34:25,602 epoch 6 - iter 146/738 - loss 0.02518261 - time (sec): 10.03 - samples/sec: 3225.85 - lr: 0.000027 - momentum: 0.000000
2023-10-17 22:34:31,064 epoch 6 - iter 219/738 - loss 0.02167308 - time (sec): 15.49 - samples/sec: 3216.28 - lr: 0.000026 - momentum: 0.000000
2023-10-17 22:34:35,610 epoch 6 - iter 292/738 - loss 0.02310727 - time (sec): 20.03 - samples/sec: 3242.49 - lr: 0.000026 - momentum: 0.000000
2023-10-17 22:34:40,377 epoch 6 - iter 365/738 - loss 0.02350712 - time (sec): 24.80 - samples/sec: 3262.02 - lr: 0.000025 - momentum: 0.000000
2023-10-17 22:34:45,894 epoch 6 - iter 438/738 - loss 0.02568561 - time (sec): 30.32 - samples/sec: 3251.89 - lr: 0.000025 - momentum: 0.000000
2023-10-17 22:34:50,637 epoch 6 - iter 511/738 - loss 0.02533559 - time (sec): 35.06 - samples/sec: 3248.03 - lr: 0.000024 - momentum: 0.000000
2023-10-17 22:34:55,852 epoch 6 - iter 584/738 - loss 0.02483365 - time (sec): 40.28 - samples/sec: 3242.98 - lr: 0.000023 - momentum: 0.000000
2023-10-17 22:35:01,847 epoch 6 - iter 657/738 - loss 0.02444670 - time (sec): 46.27 - samples/sec: 3218.62 - lr: 0.000023 - momentum: 0.000000
2023-10-17 22:35:06,861 epoch 6 - iter 730/738 - loss 0.02362799 - time (sec): 51.28 - samples/sec: 3215.12 - lr: 0.000022 - momentum: 0.000000
2023-10-17 22:35:07,312 ----------------------------------------------------------------------------------------------------
2023-10-17 22:35:07,313 EPOCH 6 done: loss 0.0236 - lr: 0.000022
2023-10-17 22:35:19,340 DEV : loss 0.20369166135787964 - f1-score (micro avg) 0.8327
2023-10-17 22:35:19,377 saving best model
2023-10-17 22:35:19,893 ----------------------------------------------------------------------------------------------------
2023-10-17 22:35:25,515 epoch 7 - iter 73/738 - loss 0.01350406 - time (sec): 5.62 - samples/sec: 3361.02 - lr: 0.000022 - momentum: 0.000000
2023-10-17 22:35:30,665 epoch 7 - iter 146/738 - loss 0.01445533 - time (sec): 10.77 - samples/sec: 3283.82 - lr: 0.000021 - momentum: 0.000000
2023-10-17 22:35:35,388 epoch 7 - iter 219/738 - loss 0.01349065 - time (sec): 15.49 - samples/sec: 3296.44 - lr: 0.000021 - momentum: 0.000000
2023-10-17 22:35:40,791 epoch 7 - iter 292/738 - loss 0.01184519 - time (sec): 20.90 - samples/sec: 3258.88 - lr: 0.000020 - momentum: 0.000000
2023-10-17 22:35:45,268 epoch 7 - iter 365/738 - loss 0.01302031 - time (sec): 25.37 - samples/sec: 3278.92 - lr: 0.000020 - momentum: 0.000000
2023-10-17 22:35:50,359 epoch 7 - iter 438/738 - loss 0.01425148 - time (sec): 30.46 - samples/sec: 3268.83 - lr: 0.000019 - momentum: 0.000000
2023-10-17 22:35:55,534 epoch 7 - iter 511/738 - loss 0.01540854 - time (sec): 35.64 - samples/sec: 3248.78 - lr: 0.000018 - momentum: 0.000000
2023-10-17 22:36:00,485 epoch 7 - iter 584/738 - loss 0.01593386 - time (sec): 40.59 - samples/sec: 3255.73 - lr: 0.000018 - momentum: 0.000000
2023-10-17 22:36:05,813 epoch 7 - iter 657/738 - loss 0.01714758 - time (sec): 45.92 - samples/sec: 3233.90 - lr: 0.000017 - momentum: 0.000000
2023-10-17 22:36:11,021 epoch 7 - iter 730/738 - loss 0.01677898 - time (sec): 51.13 - samples/sec: 3219.64 - lr: 0.000017 - momentum: 0.000000
2023-10-17 22:36:11,589 ----------------------------------------------------------------------------------------------------
2023-10-17 22:36:11,590 EPOCH 7 done: loss 0.0167 - lr: 0.000017
2023-10-17 22:36:23,302 DEV : loss 0.20619168877601624 - f1-score (micro avg) 0.8309
2023-10-17 22:36:23,335 ----------------------------------------------------------------------------------------------------
2023-10-17 22:36:28,762 epoch 8 - iter 73/738 - loss 0.01436214 - time (sec): 5.43 - samples/sec: 3505.01 - lr: 0.000016 - momentum: 0.000000
2023-10-17 22:36:33,706 epoch 8 - iter 146/738 - loss 0.01406720 - time (sec): 10.37 - samples/sec: 3337.79 - lr: 0.000016 - momentum: 0.000000
2023-10-17 22:36:39,250 epoch 8 - iter 219/738 - loss 0.01357334 - time (sec): 15.91 - samples/sec: 3295.54 - lr: 0.000015 - momentum: 0.000000
2023-10-17 22:36:44,612 epoch 8 - iter 292/738 - loss 0.01362606 - time (sec): 21.28 - samples/sec: 3277.39 - lr: 0.000015 - momentum: 0.000000
2023-10-17 22:36:49,362 epoch 8 - iter 365/738 - loss 0.01258376 - time (sec): 26.03 - samples/sec: 3288.03 - lr: 0.000014 - momentum: 0.000000
2023-10-17 22:36:54,351 epoch 8 - iter 438/738 - loss 0.01195634 - time (sec): 31.01 - samples/sec: 3256.51 - lr: 0.000013 - momentum: 0.000000
2023-10-17 22:36:59,301 epoch 8 - iter 511/738 - loss 0.01131119 - time (sec): 35.97 - samples/sec: 3247.35 - lr: 0.000013 - momentum: 0.000000
2023-10-17 22:37:04,249 epoch 8 - iter 584/738 - loss 0.01151810 - time (sec): 40.91 - samples/sec: 3262.34 - lr: 0.000012 - momentum: 0.000000
2023-10-17 22:37:09,202 epoch 8 - iter 657/738 - loss 0.01136472 - time (sec): 45.87 - samples/sec: 3258.75 - lr: 0.000012 - momentum: 0.000000
2023-10-17 22:37:13,756 epoch 8 - iter 730/738 - loss 0.01129012 - time (sec): 50.42 - samples/sec: 3264.76 - lr: 0.000011 - momentum: 0.000000
2023-10-17 22:37:14,310 ----------------------------------------------------------------------------------------------------
2023-10-17 22:37:14,310 EPOCH 8 done: loss 0.0116 - lr: 0.000011
2023-10-17 22:37:25,955 DEV : loss 0.20618367195129395 - f1-score (micro avg) 0.8382
2023-10-17 22:37:25,994 saving best model
2023-10-17 22:37:26,535 ----------------------------------------------------------------------------------------------------
2023-10-17 22:37:31,040 epoch 9 - iter 73/738 - loss 0.00179149 - time (sec): 4.50 - samples/sec: 3318.58 - lr: 0.000011 - momentum: 0.000000
2023-10-17 22:37:35,954 epoch 9 - iter 146/738 - loss 0.00388670 - time (sec): 9.42 - samples/sec: 3296.02 - lr: 0.000010 - momentum: 0.000000
2023-10-17 22:37:42,573 epoch 9 - iter 219/738 - loss 0.00558100 - time (sec): 16.04 - samples/sec: 3216.48 - lr: 0.000010 - momentum: 0.000000
2023-10-17 22:37:46,905 epoch 9 - iter 292/738 - loss 0.00579759 - time (sec): 20.37 - samples/sec: 3248.07 - lr: 0.000009 - momentum: 0.000000
2023-10-17 22:37:51,672 epoch 9 - iter 365/738 - loss 0.00592703 - time (sec): 25.14 - samples/sec: 3250.17 - lr: 0.000008 - momentum: 0.000000
2023-10-17 22:37:57,350 epoch 9 - iter 438/738 - loss 0.00668828 - time (sec): 30.81 - samples/sec: 3240.67 - lr: 0.000008 - momentum: 0.000000
2023-10-17 22:38:02,024 epoch 9 - iter 511/738 - loss 0.00651091 - time (sec): 35.49 - samples/sec: 3235.47 - lr: 0.000007 - momentum: 0.000000
2023-10-17 22:38:07,184 epoch 9 - iter 584/738 - loss 0.00667925 - time (sec): 40.65 - samples/sec: 3233.36 - lr: 0.000007 - momentum: 0.000000
2023-10-17 22:38:12,714 epoch 9 - iter 657/738 - loss 0.00716367 - time (sec): 46.18 - samples/sec: 3221.77 - lr: 0.000006 - momentum: 0.000000
2023-10-17 22:38:17,563 epoch 9 - iter 730/738 - loss 0.00684617 - time (sec): 51.03 - samples/sec: 3226.78 - lr: 0.000006 - momentum: 0.000000
2023-10-17 22:38:18,091 ----------------------------------------------------------------------------------------------------
2023-10-17 22:38:18,091 EPOCH 9 done: loss 0.0068 - lr: 0.000006
2023-10-17 22:38:29,523 DEV : loss 0.20809388160705566 - f1-score (micro avg) 0.8473
2023-10-17 22:38:29,554 saving best model
2023-10-17 22:38:30,106 ----------------------------------------------------------------------------------------------------
2023-10-17 22:38:34,557 epoch 10 - iter 73/738 - loss 0.00157154 - time (sec): 4.45 - samples/sec: 3307.08 - lr: 0.000005 - momentum: 0.000000
2023-10-17 22:38:39,511 epoch 10 - iter 146/738 - loss 0.00473884 - time (sec): 9.40 - samples/sec: 3262.89 - lr: 0.000004 - momentum: 0.000000
2023-10-17 22:38:44,413 epoch 10 - iter 219/738 - loss 0.00435182 - time (sec): 14.30 - samples/sec: 3245.94 - lr: 0.000004 - momentum: 0.000000
2023-10-17 22:38:50,176 epoch 10 - iter 292/738 - loss 0.00461898 - time (sec): 20.06 - samples/sec: 3265.73 - lr: 0.000003 - momentum: 0.000000
2023-10-17 22:38:54,905 epoch 10 - iter 365/738 - loss 0.00437464 - time (sec): 24.79 - samples/sec: 3267.62 - lr: 0.000003 - momentum: 0.000000
2023-10-17 22:39:00,079 epoch 10 - iter 438/738 - loss 0.00585767 - time (sec): 29.97 - samples/sec: 3265.78 - lr: 0.000002 - momentum: 0.000000
2023-10-17 22:39:04,813 epoch 10 - iter 511/738 - loss 0.00504414 - time (sec): 34.70 - samples/sec: 3278.60 - lr: 0.000002 - momentum: 0.000000
2023-10-17 22:39:09,643 epoch 10 - iter 584/738 - loss 0.00486786 - time (sec): 39.53 - samples/sec: 3281.88 - lr: 0.000001 - momentum: 0.000000
2023-10-17 22:39:15,461 epoch 10 - iter 657/738 - loss 0.00498413 - time (sec): 45.35 - samples/sec: 3278.45 - lr: 0.000001 - momentum: 0.000000
2023-10-17 22:39:20,217 epoch 10 - iter 730/738 - loss 0.00471586 - time (sec): 50.11 - samples/sec: 3274.82 - lr: 0.000000 - momentum: 0.000000
2023-10-17 22:39:20,859 ----------------------------------------------------------------------------------------------------
2023-10-17 22:39:20,859 EPOCH 10 done: loss 0.0047 - lr: 0.000000
2023-10-17 22:39:32,442 DEV : loss 0.21777039766311646 - f1-score (micro avg) 0.8438
2023-10-17 22:39:32,871 ----------------------------------------------------------------------------------------------------
2023-10-17 22:39:32,873 Loading model from best epoch ...
2023-10-17 22:39:34,269 SequenceTagger predicts: Dictionary with 21 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org, S-time, B-time, E-time, I-time, S-prod, B-prod, E-prod, I-prod
2023-10-17 22:39:40,451
Results:
- F-score (micro) 0.8122
- F-score (macro) 0.7141
- Accuracy 0.7007
By class:
precision recall f1-score support
loc 0.8672 0.8904 0.8787 858
pers 0.7737 0.8212 0.7967 537
org 0.5714 0.6364 0.6022 132
prod 0.6667 0.6885 0.6774 61
time 0.5714 0.6667 0.6154 54
micro avg 0.7929 0.8325 0.8122 1642
macro avg 0.6901 0.7406 0.7141 1642
weighted avg 0.7957 0.8325 0.8135 1642
2023-10-17 22:39:40,451 ----------------------------------------------------------------------------------------------------
|