File size: 23,868 Bytes
113d5a6 |
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 |
2023-10-17 08:44:19,028 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:19,028 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 08:44:19,029 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:19,029 MultiCorpus: 1100 train + 206 dev + 240 test sentences
- NER_HIPE_2022 Corpus: 1100 train + 206 dev + 240 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/de/with_doc_seperator
2023-10-17 08:44:19,029 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:19,029 Train: 1100 sentences
2023-10-17 08:44:19,029 (train_with_dev=False, train_with_test=False)
2023-10-17 08:44:19,029 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:19,029 Training Params:
2023-10-17 08:44:19,029 - learning_rate: "5e-05"
2023-10-17 08:44:19,029 - mini_batch_size: "8"
2023-10-17 08:44:19,029 - max_epochs: "10"
2023-10-17 08:44:19,029 - shuffle: "True"
2023-10-17 08:44:19,029 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:19,029 Plugins:
2023-10-17 08:44:19,029 - TensorboardLogger
2023-10-17 08:44:19,029 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 08:44:19,029 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:19,029 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 08:44:19,029 - metric: "('micro avg', 'f1-score')"
2023-10-17 08:44:19,029 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:19,029 Computation:
2023-10-17 08:44:19,029 - compute on device: cuda:0
2023-10-17 08:44:19,029 - embedding storage: none
2023-10-17 08:44:19,029 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:19,029 Model training base path: "hmbench-ajmc/de-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-3"
2023-10-17 08:44:19,029 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:19,029 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:19,030 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 08:44:19,768 epoch 1 - iter 13/138 - loss 4.23689924 - time (sec): 0.74 - samples/sec: 2774.07 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:44:20,495 epoch 1 - iter 26/138 - loss 3.72475472 - time (sec): 1.46 - samples/sec: 2913.13 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:44:21,251 epoch 1 - iter 39/138 - loss 2.98979238 - time (sec): 2.22 - samples/sec: 2921.26 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:44:21,935 epoch 1 - iter 52/138 - loss 2.54874337 - time (sec): 2.90 - samples/sec: 2897.74 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:44:22,680 epoch 1 - iter 65/138 - loss 2.15097506 - time (sec): 3.65 - samples/sec: 2930.58 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:44:23,413 epoch 1 - iter 78/138 - loss 1.87984736 - time (sec): 4.38 - samples/sec: 2955.37 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:44:24,111 epoch 1 - iter 91/138 - loss 1.68323414 - time (sec): 5.08 - samples/sec: 2946.32 - lr: 0.000033 - momentum: 0.000000
2023-10-17 08:44:24,831 epoch 1 - iter 104/138 - loss 1.52893736 - time (sec): 5.80 - samples/sec: 2944.10 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:44:25,580 epoch 1 - iter 117/138 - loss 1.38071873 - time (sec): 6.55 - samples/sec: 2965.22 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:44:26,331 epoch 1 - iter 130/138 - loss 1.27710721 - time (sec): 7.30 - samples/sec: 2942.46 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:44:26,778 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:26,779 EPOCH 1 done: loss 1.2205 - lr: 0.000047
2023-10-17 08:44:27,298 DEV : loss 0.2180010974407196 - f1-score (micro avg) 0.6165
2023-10-17 08:44:27,302 saving best model
2023-10-17 08:44:27,635 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:28,352 epoch 2 - iter 13/138 - loss 0.32735430 - time (sec): 0.72 - samples/sec: 2955.79 - lr: 0.000050 - momentum: 0.000000
2023-10-17 08:44:29,091 epoch 2 - iter 26/138 - loss 0.25572315 - time (sec): 1.45 - samples/sec: 2994.77 - lr: 0.000049 - momentum: 0.000000
2023-10-17 08:44:29,834 epoch 2 - iter 39/138 - loss 0.22680767 - time (sec): 2.20 - samples/sec: 3035.05 - lr: 0.000048 - momentum: 0.000000
2023-10-17 08:44:30,583 epoch 2 - iter 52/138 - loss 0.21737319 - time (sec): 2.95 - samples/sec: 2983.02 - lr: 0.000048 - momentum: 0.000000
2023-10-17 08:44:31,379 epoch 2 - iter 65/138 - loss 0.22002365 - time (sec): 3.74 - samples/sec: 2953.68 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:44:32,138 epoch 2 - iter 78/138 - loss 0.21253262 - time (sec): 4.50 - samples/sec: 2911.39 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:44:32,878 epoch 2 - iter 91/138 - loss 0.20313431 - time (sec): 5.24 - samples/sec: 2935.34 - lr: 0.000046 - momentum: 0.000000
2023-10-17 08:44:33,599 epoch 2 - iter 104/138 - loss 0.19280458 - time (sec): 5.96 - samples/sec: 2929.87 - lr: 0.000046 - momentum: 0.000000
2023-10-17 08:44:34,321 epoch 2 - iter 117/138 - loss 0.18675924 - time (sec): 6.69 - samples/sec: 2907.25 - lr: 0.000045 - momentum: 0.000000
2023-10-17 08:44:35,032 epoch 2 - iter 130/138 - loss 0.18564163 - time (sec): 7.40 - samples/sec: 2925.93 - lr: 0.000045 - momentum: 0.000000
2023-10-17 08:44:35,445 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:35,445 EPOCH 2 done: loss 0.1811 - lr: 0.000045
2023-10-17 08:44:36,075 DEV : loss 0.1245567575097084 - f1-score (micro avg) 0.8305
2023-10-17 08:44:36,081 saving best model
2023-10-17 08:44:36,514 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:37,267 epoch 3 - iter 13/138 - loss 0.09322958 - time (sec): 0.75 - samples/sec: 3137.87 - lr: 0.000044 - momentum: 0.000000
2023-10-17 08:44:38,045 epoch 3 - iter 26/138 - loss 0.09112179 - time (sec): 1.53 - samples/sec: 3017.03 - lr: 0.000043 - momentum: 0.000000
2023-10-17 08:44:38,773 epoch 3 - iter 39/138 - loss 0.08652228 - time (sec): 2.26 - samples/sec: 2969.66 - lr: 0.000043 - momentum: 0.000000
2023-10-17 08:44:39,577 epoch 3 - iter 52/138 - loss 0.08449509 - time (sec): 3.06 - samples/sec: 2976.32 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:44:40,288 epoch 3 - iter 65/138 - loss 0.08587226 - time (sec): 3.77 - samples/sec: 2933.58 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:44:40,994 epoch 3 - iter 78/138 - loss 0.08699599 - time (sec): 4.48 - samples/sec: 2945.29 - lr: 0.000041 - momentum: 0.000000
2023-10-17 08:44:41,741 epoch 3 - iter 91/138 - loss 0.08855977 - time (sec): 5.22 - samples/sec: 2925.78 - lr: 0.000041 - momentum: 0.000000
2023-10-17 08:44:42,515 epoch 3 - iter 104/138 - loss 0.09546851 - time (sec): 6.00 - samples/sec: 2930.81 - lr: 0.000040 - momentum: 0.000000
2023-10-17 08:44:43,229 epoch 3 - iter 117/138 - loss 0.09882774 - time (sec): 6.71 - samples/sec: 2931.36 - lr: 0.000040 - momentum: 0.000000
2023-10-17 08:44:43,923 epoch 3 - iter 130/138 - loss 0.10085115 - time (sec): 7.41 - samples/sec: 2918.81 - lr: 0.000039 - momentum: 0.000000
2023-10-17 08:44:44,378 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:44,378 EPOCH 3 done: loss 0.1007 - lr: 0.000039
2023-10-17 08:44:45,057 DEV : loss 0.12144241482019424 - f1-score (micro avg) 0.8712
2023-10-17 08:44:45,062 saving best model
2023-10-17 08:44:45,506 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:46,236 epoch 4 - iter 13/138 - loss 0.04907389 - time (sec): 0.73 - samples/sec: 2851.54 - lr: 0.000038 - momentum: 0.000000
2023-10-17 08:44:46,974 epoch 4 - iter 26/138 - loss 0.05122455 - time (sec): 1.46 - samples/sec: 2921.88 - lr: 0.000038 - momentum: 0.000000
2023-10-17 08:44:47,691 epoch 4 - iter 39/138 - loss 0.04522759 - time (sec): 2.18 - samples/sec: 2934.94 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:44:48,379 epoch 4 - iter 52/138 - loss 0.05104925 - time (sec): 2.87 - samples/sec: 2931.29 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:44:49,118 epoch 4 - iter 65/138 - loss 0.05515405 - time (sec): 3.61 - samples/sec: 2912.28 - lr: 0.000036 - momentum: 0.000000
2023-10-17 08:44:49,867 epoch 4 - iter 78/138 - loss 0.05772425 - time (sec): 4.36 - samples/sec: 2916.87 - lr: 0.000036 - momentum: 0.000000
2023-10-17 08:44:50,808 epoch 4 - iter 91/138 - loss 0.06243765 - time (sec): 5.30 - samples/sec: 2776.13 - lr: 0.000035 - momentum: 0.000000
2023-10-17 08:44:51,582 epoch 4 - iter 104/138 - loss 0.06726186 - time (sec): 6.07 - samples/sec: 2789.80 - lr: 0.000035 - momentum: 0.000000
2023-10-17 08:44:52,345 epoch 4 - iter 117/138 - loss 0.07264164 - time (sec): 6.83 - samples/sec: 2814.01 - lr: 0.000034 - momentum: 0.000000
2023-10-17 08:44:53,114 epoch 4 - iter 130/138 - loss 0.07138648 - time (sec): 7.60 - samples/sec: 2816.62 - lr: 0.000034 - momentum: 0.000000
2023-10-17 08:44:53,545 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:53,546 EPOCH 4 done: loss 0.0710 - lr: 0.000034
2023-10-17 08:44:54,237 DEV : loss 0.14281411468982697 - f1-score (micro avg) 0.862
2023-10-17 08:44:54,241 ----------------------------------------------------------------------------------------------------
2023-10-17 08:44:54,974 epoch 5 - iter 13/138 - loss 0.08915125 - time (sec): 0.73 - samples/sec: 3058.74 - lr: 0.000033 - momentum: 0.000000
2023-10-17 08:44:55,741 epoch 5 - iter 26/138 - loss 0.08538151 - time (sec): 1.50 - samples/sec: 2962.06 - lr: 0.000032 - momentum: 0.000000
2023-10-17 08:44:56,460 epoch 5 - iter 39/138 - loss 0.07451282 - time (sec): 2.22 - samples/sec: 2981.90 - lr: 0.000032 - momentum: 0.000000
2023-10-17 08:44:57,151 epoch 5 - iter 52/138 - loss 0.07480996 - time (sec): 2.91 - samples/sec: 2949.67 - lr: 0.000031 - momentum: 0.000000
2023-10-17 08:44:57,881 epoch 5 - iter 65/138 - loss 0.07246926 - time (sec): 3.64 - samples/sec: 2991.74 - lr: 0.000031 - momentum: 0.000000
2023-10-17 08:44:58,615 epoch 5 - iter 78/138 - loss 0.07599919 - time (sec): 4.37 - samples/sec: 2985.62 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:44:59,411 epoch 5 - iter 91/138 - loss 0.06939059 - time (sec): 5.17 - samples/sec: 2936.72 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:45:00,192 epoch 5 - iter 104/138 - loss 0.06497114 - time (sec): 5.95 - samples/sec: 2929.39 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:45:00,968 epoch 5 - iter 117/138 - loss 0.06053147 - time (sec): 6.73 - samples/sec: 2901.84 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:45:01,711 epoch 5 - iter 130/138 - loss 0.05852661 - time (sec): 7.47 - samples/sec: 2896.28 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:45:02,146 ----------------------------------------------------------------------------------------------------
2023-10-17 08:45:02,147 EPOCH 5 done: loss 0.0586 - lr: 0.000028
2023-10-17 08:45:02,908 DEV : loss 0.1629737764596939 - f1-score (micro avg) 0.8708
2023-10-17 08:45:02,913 ----------------------------------------------------------------------------------------------------
2023-10-17 08:45:03,685 epoch 6 - iter 13/138 - loss 0.04102128 - time (sec): 0.77 - samples/sec: 2812.89 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:45:04,460 epoch 6 - iter 26/138 - loss 0.04110212 - time (sec): 1.55 - samples/sec: 2806.20 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:45:05,203 epoch 6 - iter 39/138 - loss 0.05651555 - time (sec): 2.29 - samples/sec: 2772.29 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:45:05,948 epoch 6 - iter 52/138 - loss 0.06576880 - time (sec): 3.03 - samples/sec: 2757.34 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:45:06,784 epoch 6 - iter 65/138 - loss 0.06388934 - time (sec): 3.87 - samples/sec: 2730.08 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:45:07,534 epoch 6 - iter 78/138 - loss 0.06510905 - time (sec): 4.62 - samples/sec: 2741.94 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:45:08,290 epoch 6 - iter 91/138 - loss 0.06051985 - time (sec): 5.38 - samples/sec: 2776.01 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:45:09,031 epoch 6 - iter 104/138 - loss 0.05874561 - time (sec): 6.12 - samples/sec: 2790.54 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:45:09,785 epoch 6 - iter 117/138 - loss 0.05433949 - time (sec): 6.87 - samples/sec: 2798.80 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:45:10,511 epoch 6 - iter 130/138 - loss 0.05085432 - time (sec): 7.60 - samples/sec: 2812.65 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:45:10,968 ----------------------------------------------------------------------------------------------------
2023-10-17 08:45:10,968 EPOCH 6 done: loss 0.0483 - lr: 0.000023
2023-10-17 08:45:11,693 DEV : loss 0.170999675989151 - f1-score (micro avg) 0.8633
2023-10-17 08:45:11,698 ----------------------------------------------------------------------------------------------------
2023-10-17 08:45:12,423 epoch 7 - iter 13/138 - loss 0.03378651 - time (sec): 0.72 - samples/sec: 3015.17 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:45:13,108 epoch 7 - iter 26/138 - loss 0.02358591 - time (sec): 1.41 - samples/sec: 3105.48 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:45:13,838 epoch 7 - iter 39/138 - loss 0.02146202 - time (sec): 2.14 - samples/sec: 2952.00 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:45:14,582 epoch 7 - iter 52/138 - loss 0.02877528 - time (sec): 2.88 - samples/sec: 2903.41 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:45:15,433 epoch 7 - iter 65/138 - loss 0.03053761 - time (sec): 3.73 - samples/sec: 2887.15 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:45:16,172 epoch 7 - iter 78/138 - loss 0.03095385 - time (sec): 4.47 - samples/sec: 2900.40 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:45:16,946 epoch 7 - iter 91/138 - loss 0.03425468 - time (sec): 5.25 - samples/sec: 2880.13 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:45:17,724 epoch 7 - iter 104/138 - loss 0.03297833 - time (sec): 6.02 - samples/sec: 2879.17 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:45:18,528 epoch 7 - iter 117/138 - loss 0.03274128 - time (sec): 6.83 - samples/sec: 2851.13 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:45:19,270 epoch 7 - iter 130/138 - loss 0.02995608 - time (sec): 7.57 - samples/sec: 2854.86 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:45:19,722 ----------------------------------------------------------------------------------------------------
2023-10-17 08:45:19,722 EPOCH 7 done: loss 0.0314 - lr: 0.000017
2023-10-17 08:45:20,357 DEV : loss 0.17921938002109528 - f1-score (micro avg) 0.87
2023-10-17 08:45:20,361 ----------------------------------------------------------------------------------------------------
2023-10-17 08:45:21,112 epoch 8 - iter 13/138 - loss 0.02216531 - time (sec): 0.75 - samples/sec: 2828.84 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:45:21,865 epoch 8 - iter 26/138 - loss 0.02828258 - time (sec): 1.50 - samples/sec: 2755.00 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:45:22,673 epoch 8 - iter 39/138 - loss 0.02250268 - time (sec): 2.31 - samples/sec: 2815.84 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:45:23,452 epoch 8 - iter 52/138 - loss 0.02828756 - time (sec): 3.09 - samples/sec: 2851.32 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:45:24,194 epoch 8 - iter 65/138 - loss 0.02593428 - time (sec): 3.83 - samples/sec: 2859.14 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:45:24,899 epoch 8 - iter 78/138 - loss 0.02247652 - time (sec): 4.54 - samples/sec: 2816.39 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:45:25,713 epoch 8 - iter 91/138 - loss 0.02261621 - time (sec): 5.35 - samples/sec: 2810.12 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:45:26,434 epoch 8 - iter 104/138 - loss 0.02328122 - time (sec): 6.07 - samples/sec: 2832.31 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:45:27,199 epoch 8 - iter 117/138 - loss 0.02375440 - time (sec): 6.84 - samples/sec: 2830.24 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:45:27,931 epoch 8 - iter 130/138 - loss 0.02207057 - time (sec): 7.57 - samples/sec: 2852.00 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:45:28,389 ----------------------------------------------------------------------------------------------------
2023-10-17 08:45:28,389 EPOCH 8 done: loss 0.0274 - lr: 0.000012
2023-10-17 08:45:29,033 DEV : loss 0.17290453612804413 - f1-score (micro avg) 0.8766
2023-10-17 08:45:29,038 saving best model
2023-10-17 08:45:29,506 ----------------------------------------------------------------------------------------------------
2023-10-17 08:45:30,198 epoch 9 - iter 13/138 - loss 0.01479278 - time (sec): 0.69 - samples/sec: 3106.28 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:45:30,964 epoch 9 - iter 26/138 - loss 0.02037863 - time (sec): 1.46 - samples/sec: 3100.06 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:45:31,702 epoch 9 - iter 39/138 - loss 0.01773260 - time (sec): 2.19 - samples/sec: 3000.70 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:45:32,385 epoch 9 - iter 52/138 - loss 0.01953734 - time (sec): 2.88 - samples/sec: 2969.58 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:45:33,085 epoch 9 - iter 65/138 - loss 0.01974078 - time (sec): 3.58 - samples/sec: 3023.44 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:45:33,826 epoch 9 - iter 78/138 - loss 0.01741408 - time (sec): 4.32 - samples/sec: 3019.24 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:45:34,528 epoch 9 - iter 91/138 - loss 0.02365520 - time (sec): 5.02 - samples/sec: 2999.66 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:45:35,271 epoch 9 - iter 104/138 - loss 0.02358021 - time (sec): 5.76 - samples/sec: 2981.94 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:45:35,978 epoch 9 - iter 117/138 - loss 0.02217472 - time (sec): 6.47 - samples/sec: 2980.06 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:45:36,677 epoch 9 - iter 130/138 - loss 0.02094984 - time (sec): 7.17 - samples/sec: 2990.54 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:45:37,134 ----------------------------------------------------------------------------------------------------
2023-10-17 08:45:37,134 EPOCH 9 done: loss 0.0202 - lr: 0.000006
2023-10-17 08:45:37,790 DEV : loss 0.1823827624320984 - f1-score (micro avg) 0.872
2023-10-17 08:45:37,795 ----------------------------------------------------------------------------------------------------
2023-10-17 08:45:38,585 epoch 10 - iter 13/138 - loss 0.08784586 - time (sec): 0.79 - samples/sec: 3159.87 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:45:39,345 epoch 10 - iter 26/138 - loss 0.04966499 - time (sec): 1.55 - samples/sec: 3082.74 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:45:40,068 epoch 10 - iter 39/138 - loss 0.03598407 - time (sec): 2.27 - samples/sec: 3108.91 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:45:40,765 epoch 10 - iter 52/138 - loss 0.02807500 - time (sec): 2.97 - samples/sec: 3070.41 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:45:41,488 epoch 10 - iter 65/138 - loss 0.02399162 - time (sec): 3.69 - samples/sec: 3046.98 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:45:42,166 epoch 10 - iter 78/138 - loss 0.02165534 - time (sec): 4.37 - samples/sec: 3027.74 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:45:42,927 epoch 10 - iter 91/138 - loss 0.01912312 - time (sec): 5.13 - samples/sec: 2985.88 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:45:43,591 epoch 10 - iter 104/138 - loss 0.01886851 - time (sec): 5.79 - samples/sec: 2971.80 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:45:44,320 epoch 10 - iter 117/138 - loss 0.01778748 - time (sec): 6.52 - samples/sec: 2975.48 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:45:45,028 epoch 10 - iter 130/138 - loss 0.01786949 - time (sec): 7.23 - samples/sec: 2985.97 - lr: 0.000000 - momentum: 0.000000
2023-10-17 08:45:45,509 ----------------------------------------------------------------------------------------------------
2023-10-17 08:45:45,510 EPOCH 10 done: loss 0.0176 - lr: 0.000000
2023-10-17 08:45:46,146 DEV : loss 0.18640285730361938 - f1-score (micro avg) 0.872
2023-10-17 08:45:46,499 ----------------------------------------------------------------------------------------------------
2023-10-17 08:45:46,500 Loading model from best epoch ...
2023-10-17 08:45:47,856 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 08:45:48,650
Results:
- F-score (micro) 0.9067
- F-score (macro) 0.9372
- Accuracy 0.8413
By class:
precision recall f1-score support
scope 0.8895 0.9148 0.9020 176
pers 0.9683 0.9531 0.9606 128
work 0.7975 0.8514 0.8235 74
object 1.0000 1.0000 1.0000 2
loc 1.0000 1.0000 1.0000 2
micro avg 0.8974 0.9162 0.9067 382
macro avg 0.9310 0.9438 0.9372 382
weighted avg 0.8992 0.9162 0.9075 382
2023-10-17 08:45:48,650 ----------------------------------------------------------------------------------------------------
|