File size: 24,037 Bytes
545d882 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 |
2023-10-17 08:25:45,131 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:45,132 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 08:25:45,132 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:45,132 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 08:25:45,132 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:45,132 Train: 1100 sentences
2023-10-17 08:25:45,133 (train_with_dev=False, train_with_test=False)
2023-10-17 08:25:45,133 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:45,133 Training Params:
2023-10-17 08:25:45,133 - learning_rate: "5e-05"
2023-10-17 08:25:45,133 - mini_batch_size: "8"
2023-10-17 08:25:45,133 - max_epochs: "10"
2023-10-17 08:25:45,133 - shuffle: "True"
2023-10-17 08:25:45,133 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:45,133 Plugins:
2023-10-17 08:25:45,133 - TensorboardLogger
2023-10-17 08:25:45,133 - LinearScheduler | warmup_fraction: '0.1'
2023-10-17 08:25:45,133 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:45,133 Final evaluation on model from best epoch (best-model.pt)
2023-10-17 08:25:45,133 - metric: "('micro avg', 'f1-score')"
2023-10-17 08:25:45,133 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:45,133 Computation:
2023-10-17 08:25:45,133 - compute on device: cuda:0
2023-10-17 08:25:45,133 - embedding storage: none
2023-10-17 08:25:45,133 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:45,133 Model training base path: "hmbench-ajmc/de-hmteams/teams-base-historic-multilingual-discriminator-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-17 08:25:45,133 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:45,133 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:45,133 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-17 08:25:45,848 epoch 1 - iter 13/138 - loss 3.43157067 - time (sec): 0.71 - samples/sec: 2929.63 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:25:46,565 epoch 1 - iter 26/138 - loss 2.94341200 - time (sec): 1.43 - samples/sec: 2874.16 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:25:47,299 epoch 1 - iter 39/138 - loss 2.44187050 - time (sec): 2.17 - samples/sec: 2856.13 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:25:48,033 epoch 1 - iter 52/138 - loss 1.99375963 - time (sec): 2.90 - samples/sec: 2884.28 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:25:48,817 epoch 1 - iter 65/138 - loss 1.67766635 - time (sec): 3.68 - samples/sec: 2904.62 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:25:49,568 epoch 1 - iter 78/138 - loss 1.47320680 - time (sec): 4.43 - samples/sec: 2945.81 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:25:50,281 epoch 1 - iter 91/138 - loss 1.34446452 - time (sec): 5.15 - samples/sec: 2939.89 - lr: 0.000033 - momentum: 0.000000
2023-10-17 08:25:51,011 epoch 1 - iter 104/138 - loss 1.21983278 - time (sec): 5.88 - samples/sec: 2920.15 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:25:51,739 epoch 1 - iter 117/138 - loss 1.12117619 - time (sec): 6.60 - samples/sec: 2922.93 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:25:52,492 epoch 1 - iter 130/138 - loss 1.04463454 - time (sec): 7.36 - samples/sec: 2913.32 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:25:52,958 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:52,959 EPOCH 1 done: loss 1.0032 - lr: 0.000047
2023-10-17 08:25:53,496 DEV : loss 0.24205686151981354 - f1-score (micro avg) 0.6482
2023-10-17 08:25:53,501 saving best model
2023-10-17 08:25:53,850 ----------------------------------------------------------------------------------------------------
2023-10-17 08:25:54,620 epoch 2 - iter 13/138 - loss 0.27043098 - time (sec): 0.77 - samples/sec: 3000.22 - lr: 0.000050 - momentum: 0.000000
2023-10-17 08:25:55,375 epoch 2 - iter 26/138 - loss 0.24273143 - time (sec): 1.52 - samples/sec: 2867.69 - lr: 0.000049 - momentum: 0.000000
2023-10-17 08:25:56,141 epoch 2 - iter 39/138 - loss 0.25085266 - time (sec): 2.29 - samples/sec: 2912.01 - lr: 0.000048 - momentum: 0.000000
2023-10-17 08:25:56,839 epoch 2 - iter 52/138 - loss 0.24140883 - time (sec): 2.99 - samples/sec: 2964.32 - lr: 0.000048 - momentum: 0.000000
2023-10-17 08:25:57,552 epoch 2 - iter 65/138 - loss 0.22835302 - time (sec): 3.70 - samples/sec: 2914.71 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:25:58,287 epoch 2 - iter 78/138 - loss 0.22322428 - time (sec): 4.43 - samples/sec: 2942.73 - lr: 0.000047 - momentum: 0.000000
2023-10-17 08:25:59,003 epoch 2 - iter 91/138 - loss 0.21610097 - time (sec): 5.15 - samples/sec: 2928.90 - lr: 0.000046 - momentum: 0.000000
2023-10-17 08:25:59,721 epoch 2 - iter 104/138 - loss 0.20830588 - time (sec): 5.87 - samples/sec: 2900.25 - lr: 0.000046 - momentum: 0.000000
2023-10-17 08:26:00,475 epoch 2 - iter 117/138 - loss 0.20383405 - time (sec): 6.62 - samples/sec: 2896.09 - lr: 0.000045 - momentum: 0.000000
2023-10-17 08:26:01,219 epoch 2 - iter 130/138 - loss 0.19579670 - time (sec): 7.37 - samples/sec: 2918.89 - lr: 0.000045 - momentum: 0.000000
2023-10-17 08:26:01,632 ----------------------------------------------------------------------------------------------------
2023-10-17 08:26:01,632 EPOCH 2 done: loss 0.1884 - lr: 0.000045
2023-10-17 08:26:02,266 DEV : loss 0.1432858556509018 - f1-score (micro avg) 0.81
2023-10-17 08:26:02,270 saving best model
2023-10-17 08:26:02,711 ----------------------------------------------------------------------------------------------------
2023-10-17 08:26:03,446 epoch 3 - iter 13/138 - loss 0.09662708 - time (sec): 0.73 - samples/sec: 2790.06 - lr: 0.000044 - momentum: 0.000000
2023-10-17 08:26:04,204 epoch 3 - iter 26/138 - loss 0.09741620 - time (sec): 1.49 - samples/sec: 2984.74 - lr: 0.000043 - momentum: 0.000000
2023-10-17 08:26:04,938 epoch 3 - iter 39/138 - loss 0.08527103 - time (sec): 2.22 - samples/sec: 2877.89 - lr: 0.000043 - momentum: 0.000000
2023-10-17 08:26:05,629 epoch 3 - iter 52/138 - loss 0.08554042 - time (sec): 2.91 - samples/sec: 2862.09 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:26:06,459 epoch 3 - iter 65/138 - loss 0.08704104 - time (sec): 3.74 - samples/sec: 2854.23 - lr: 0.000042 - momentum: 0.000000
2023-10-17 08:26:07,158 epoch 3 - iter 78/138 - loss 0.08765933 - time (sec): 4.44 - samples/sec: 2849.52 - lr: 0.000041 - momentum: 0.000000
2023-10-17 08:26:07,904 epoch 3 - iter 91/138 - loss 0.08846124 - time (sec): 5.19 - samples/sec: 2895.29 - lr: 0.000041 - momentum: 0.000000
2023-10-17 08:26:08,635 epoch 3 - iter 104/138 - loss 0.09460890 - time (sec): 5.92 - samples/sec: 2910.53 - lr: 0.000040 - momentum: 0.000000
2023-10-17 08:26:09,355 epoch 3 - iter 117/138 - loss 0.09452451 - time (sec): 6.64 - samples/sec: 2913.27 - lr: 0.000040 - momentum: 0.000000
2023-10-17 08:26:10,070 epoch 3 - iter 130/138 - loss 0.10364161 - time (sec): 7.35 - samples/sec: 2935.57 - lr: 0.000039 - momentum: 0.000000
2023-10-17 08:26:10,523 ----------------------------------------------------------------------------------------------------
2023-10-17 08:26:10,524 EPOCH 3 done: loss 0.1024 - lr: 0.000039
2023-10-17 08:26:11,150 DEV : loss 0.12657766044139862 - f1-score (micro avg) 0.8361
2023-10-17 08:26:11,155 saving best model
2023-10-17 08:26:11,582 ----------------------------------------------------------------------------------------------------
2023-10-17 08:26:12,355 epoch 4 - iter 13/138 - loss 0.06164321 - time (sec): 0.77 - samples/sec: 3073.48 - lr: 0.000038 - momentum: 0.000000
2023-10-17 08:26:13,056 epoch 4 - iter 26/138 - loss 0.06900307 - time (sec): 1.47 - samples/sec: 2954.64 - lr: 0.000038 - momentum: 0.000000
2023-10-17 08:26:13,791 epoch 4 - iter 39/138 - loss 0.06737054 - time (sec): 2.21 - samples/sec: 2937.82 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:26:14,545 epoch 4 - iter 52/138 - loss 0.07217805 - time (sec): 2.96 - samples/sec: 2878.16 - lr: 0.000037 - momentum: 0.000000
2023-10-17 08:26:15,299 epoch 4 - iter 65/138 - loss 0.08353830 - time (sec): 3.72 - samples/sec: 2937.82 - lr: 0.000036 - momentum: 0.000000
2023-10-17 08:26:16,032 epoch 4 - iter 78/138 - loss 0.08015663 - time (sec): 4.45 - samples/sec: 2898.27 - lr: 0.000036 - momentum: 0.000000
2023-10-17 08:26:16,795 epoch 4 - iter 91/138 - loss 0.08026446 - time (sec): 5.21 - samples/sec: 2882.55 - lr: 0.000035 - momentum: 0.000000
2023-10-17 08:26:17,553 epoch 4 - iter 104/138 - loss 0.07646328 - time (sec): 5.97 - samples/sec: 2892.23 - lr: 0.000035 - momentum: 0.000000
2023-10-17 08:26:18,316 epoch 4 - iter 117/138 - loss 0.07885204 - time (sec): 6.73 - samples/sec: 2889.55 - lr: 0.000034 - momentum: 0.000000
2023-10-17 08:26:19,036 epoch 4 - iter 130/138 - loss 0.07560251 - time (sec): 7.45 - samples/sec: 2876.21 - lr: 0.000034 - momentum: 0.000000
2023-10-17 08:26:19,522 ----------------------------------------------------------------------------------------------------
2023-10-17 08:26:19,522 EPOCH 4 done: loss 0.0749 - lr: 0.000034
2023-10-17 08:26:20,300 DEV : loss 0.1538553535938263 - f1-score (micro avg) 0.8603
2023-10-17 08:26:20,304 saving best model
2023-10-17 08:26:20,751 ----------------------------------------------------------------------------------------------------
2023-10-17 08:26:21,519 epoch 5 - iter 13/138 - loss 0.08129829 - time (sec): 0.76 - samples/sec: 2871.69 - lr: 0.000033 - momentum: 0.000000
2023-10-17 08:26:22,261 epoch 5 - iter 26/138 - loss 0.07047005 - time (sec): 1.51 - samples/sec: 2959.86 - lr: 0.000032 - momentum: 0.000000
2023-10-17 08:26:23,009 epoch 5 - iter 39/138 - loss 0.05637991 - time (sec): 2.25 - samples/sec: 2912.22 - lr: 0.000032 - momentum: 0.000000
2023-10-17 08:26:23,713 epoch 5 - iter 52/138 - loss 0.06461053 - time (sec): 2.96 - samples/sec: 2932.41 - lr: 0.000031 - momentum: 0.000000
2023-10-17 08:26:24,440 epoch 5 - iter 65/138 - loss 0.05885473 - time (sec): 3.68 - samples/sec: 2890.02 - lr: 0.000031 - momentum: 0.000000
2023-10-17 08:26:25,168 epoch 5 - iter 78/138 - loss 0.05439754 - time (sec): 4.41 - samples/sec: 2858.89 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:26:25,890 epoch 5 - iter 91/138 - loss 0.05002054 - time (sec): 5.13 - samples/sec: 2894.96 - lr: 0.000030 - momentum: 0.000000
2023-10-17 08:26:26,677 epoch 5 - iter 104/138 - loss 0.04951814 - time (sec): 5.92 - samples/sec: 2896.48 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:26:27,422 epoch 5 - iter 117/138 - loss 0.05822267 - time (sec): 6.67 - samples/sec: 2926.58 - lr: 0.000029 - momentum: 0.000000
2023-10-17 08:26:28,160 epoch 5 - iter 130/138 - loss 0.05859372 - time (sec): 7.40 - samples/sec: 2912.45 - lr: 0.000028 - momentum: 0.000000
2023-10-17 08:26:28,641 ----------------------------------------------------------------------------------------------------
2023-10-17 08:26:28,641 EPOCH 5 done: loss 0.0567 - lr: 0.000028
2023-10-17 08:26:29,293 DEV : loss 0.1780824512243271 - f1-score (micro avg) 0.8585
2023-10-17 08:26:29,298 ----------------------------------------------------------------------------------------------------
2023-10-17 08:26:30,026 epoch 6 - iter 13/138 - loss 0.01633802 - time (sec): 0.73 - samples/sec: 2852.60 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:26:30,810 epoch 6 - iter 26/138 - loss 0.04230440 - time (sec): 1.51 - samples/sec: 3087.62 - lr: 0.000027 - momentum: 0.000000
2023-10-17 08:26:31,517 epoch 6 - iter 39/138 - loss 0.03934797 - time (sec): 2.22 - samples/sec: 3064.13 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:26:32,220 epoch 6 - iter 52/138 - loss 0.03589065 - time (sec): 2.92 - samples/sec: 3006.27 - lr: 0.000026 - momentum: 0.000000
2023-10-17 08:26:32,987 epoch 6 - iter 65/138 - loss 0.03127002 - time (sec): 3.69 - samples/sec: 2954.78 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:26:33,761 epoch 6 - iter 78/138 - loss 0.03359051 - time (sec): 4.46 - samples/sec: 2937.72 - lr: 0.000025 - momentum: 0.000000
2023-10-17 08:26:34,495 epoch 6 - iter 91/138 - loss 0.03844091 - time (sec): 5.20 - samples/sec: 2929.76 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:26:35,251 epoch 6 - iter 104/138 - loss 0.04316186 - time (sec): 5.95 - samples/sec: 2900.75 - lr: 0.000024 - momentum: 0.000000
2023-10-17 08:26:35,972 epoch 6 - iter 117/138 - loss 0.04013628 - time (sec): 6.67 - samples/sec: 2892.73 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:26:36,732 epoch 6 - iter 130/138 - loss 0.03952624 - time (sec): 7.43 - samples/sec: 2887.53 - lr: 0.000023 - momentum: 0.000000
2023-10-17 08:26:37,182 ----------------------------------------------------------------------------------------------------
2023-10-17 08:26:37,182 EPOCH 6 done: loss 0.0465 - lr: 0.000023
2023-10-17 08:26:37,817 DEV : loss 0.15540580451488495 - f1-score (micro avg) 0.878
2023-10-17 08:26:37,821 saving best model
2023-10-17 08:26:38,247 ----------------------------------------------------------------------------------------------------
2023-10-17 08:26:38,961 epoch 7 - iter 13/138 - loss 0.01397906 - time (sec): 0.71 - samples/sec: 2701.78 - lr: 0.000022 - momentum: 0.000000
2023-10-17 08:26:39,684 epoch 7 - iter 26/138 - loss 0.02628629 - time (sec): 1.43 - samples/sec: 2811.79 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:26:40,426 epoch 7 - iter 39/138 - loss 0.04371837 - time (sec): 2.18 - samples/sec: 2824.49 - lr: 0.000021 - momentum: 0.000000
2023-10-17 08:26:41,229 epoch 7 - iter 52/138 - loss 0.03915947 - time (sec): 2.98 - samples/sec: 2767.46 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:26:41,999 epoch 7 - iter 65/138 - loss 0.04067551 - time (sec): 3.75 - samples/sec: 2832.11 - lr: 0.000020 - momentum: 0.000000
2023-10-17 08:26:42,768 epoch 7 - iter 78/138 - loss 0.03483673 - time (sec): 4.52 - samples/sec: 2803.84 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:26:43,487 epoch 7 - iter 91/138 - loss 0.03369841 - time (sec): 5.24 - samples/sec: 2815.66 - lr: 0.000019 - momentum: 0.000000
2023-10-17 08:26:44,217 epoch 7 - iter 104/138 - loss 0.03193150 - time (sec): 5.97 - samples/sec: 2838.02 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:26:45,044 epoch 7 - iter 117/138 - loss 0.03238749 - time (sec): 6.79 - samples/sec: 2837.96 - lr: 0.000018 - momentum: 0.000000
2023-10-17 08:26:45,778 epoch 7 - iter 130/138 - loss 0.03219471 - time (sec): 7.53 - samples/sec: 2849.44 - lr: 0.000017 - momentum: 0.000000
2023-10-17 08:26:46,304 ----------------------------------------------------------------------------------------------------
2023-10-17 08:26:46,305 EPOCH 7 done: loss 0.0327 - lr: 0.000017
2023-10-17 08:26:46,939 DEV : loss 0.1837640106678009 - f1-score (micro avg) 0.8723
2023-10-17 08:26:46,943 ----------------------------------------------------------------------------------------------------
2023-10-17 08:26:47,694 epoch 8 - iter 13/138 - loss 0.03074508 - time (sec): 0.75 - samples/sec: 2800.68 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:26:48,456 epoch 8 - iter 26/138 - loss 0.03711762 - time (sec): 1.51 - samples/sec: 2885.28 - lr: 0.000016 - momentum: 0.000000
2023-10-17 08:26:49,183 epoch 8 - iter 39/138 - loss 0.03737419 - time (sec): 2.24 - samples/sec: 2923.03 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:26:49,928 epoch 8 - iter 52/138 - loss 0.03129664 - time (sec): 2.98 - samples/sec: 2923.94 - lr: 0.000015 - momentum: 0.000000
2023-10-17 08:26:50,719 epoch 8 - iter 65/138 - loss 0.02973725 - time (sec): 3.78 - samples/sec: 2904.04 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:26:51,464 epoch 8 - iter 78/138 - loss 0.02665196 - time (sec): 4.52 - samples/sec: 2865.84 - lr: 0.000014 - momentum: 0.000000
2023-10-17 08:26:52,233 epoch 8 - iter 91/138 - loss 0.02400992 - time (sec): 5.29 - samples/sec: 2882.20 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:26:52,997 epoch 8 - iter 104/138 - loss 0.02313313 - time (sec): 6.05 - samples/sec: 2852.01 - lr: 0.000013 - momentum: 0.000000
2023-10-17 08:26:53,736 epoch 8 - iter 117/138 - loss 0.02525509 - time (sec): 6.79 - samples/sec: 2864.69 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:26:54,476 epoch 8 - iter 130/138 - loss 0.02647068 - time (sec): 7.53 - samples/sec: 2872.09 - lr: 0.000012 - momentum: 0.000000
2023-10-17 08:26:54,891 ----------------------------------------------------------------------------------------------------
2023-10-17 08:26:54,891 EPOCH 8 done: loss 0.0254 - lr: 0.000012
2023-10-17 08:26:55,522 DEV : loss 0.1877746433019638 - f1-score (micro avg) 0.8886
2023-10-17 08:26:55,527 saving best model
2023-10-17 08:26:55,982 ----------------------------------------------------------------------------------------------------
2023-10-17 08:26:56,735 epoch 9 - iter 13/138 - loss 0.03739812 - time (sec): 0.75 - samples/sec: 2778.28 - lr: 0.000011 - momentum: 0.000000
2023-10-17 08:26:57,511 epoch 9 - iter 26/138 - loss 0.03150704 - time (sec): 1.52 - samples/sec: 2646.10 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:26:58,254 epoch 9 - iter 39/138 - loss 0.02277717 - time (sec): 2.27 - samples/sec: 2725.40 - lr: 0.000010 - momentum: 0.000000
2023-10-17 08:26:59,043 epoch 9 - iter 52/138 - loss 0.02272279 - time (sec): 3.06 - samples/sec: 2739.97 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:26:59,809 epoch 9 - iter 65/138 - loss 0.02141050 - time (sec): 3.82 - samples/sec: 2755.69 - lr: 0.000009 - momentum: 0.000000
2023-10-17 08:27:00,593 epoch 9 - iter 78/138 - loss 0.01807415 - time (sec): 4.61 - samples/sec: 2814.59 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:27:01,320 epoch 9 - iter 91/138 - loss 0.01752412 - time (sec): 5.33 - samples/sec: 2842.18 - lr: 0.000008 - momentum: 0.000000
2023-10-17 08:27:02,012 epoch 9 - iter 104/138 - loss 0.01795085 - time (sec): 6.02 - samples/sec: 2815.64 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:27:02,736 epoch 9 - iter 117/138 - loss 0.01734781 - time (sec): 6.75 - samples/sec: 2805.11 - lr: 0.000007 - momentum: 0.000000
2023-10-17 08:27:03,488 epoch 9 - iter 130/138 - loss 0.01818842 - time (sec): 7.50 - samples/sec: 2821.79 - lr: 0.000006 - momentum: 0.000000
2023-10-17 08:27:03,951 ----------------------------------------------------------------------------------------------------
2023-10-17 08:27:03,951 EPOCH 9 done: loss 0.0195 - lr: 0.000006
2023-10-17 08:27:04,597 DEV : loss 0.1937594711780548 - f1-score (micro avg) 0.891
2023-10-17 08:27:04,602 saving best model
2023-10-17 08:27:05,034 ----------------------------------------------------------------------------------------------------
2023-10-17 08:27:05,765 epoch 10 - iter 13/138 - loss 0.01784379 - time (sec): 0.73 - samples/sec: 2849.89 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:27:06,512 epoch 10 - iter 26/138 - loss 0.00971714 - time (sec): 1.48 - samples/sec: 2742.12 - lr: 0.000005 - momentum: 0.000000
2023-10-17 08:27:07,254 epoch 10 - iter 39/138 - loss 0.01334067 - time (sec): 2.22 - samples/sec: 2773.75 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:27:08,064 epoch 10 - iter 52/138 - loss 0.01038007 - time (sec): 3.03 - samples/sec: 2718.39 - lr: 0.000004 - momentum: 0.000000
2023-10-17 08:27:08,829 epoch 10 - iter 65/138 - loss 0.00834679 - time (sec): 3.79 - samples/sec: 2764.06 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:27:09,585 epoch 10 - iter 78/138 - loss 0.01155802 - time (sec): 4.55 - samples/sec: 2772.08 - lr: 0.000003 - momentum: 0.000000
2023-10-17 08:27:10,365 epoch 10 - iter 91/138 - loss 0.01175069 - time (sec): 5.33 - samples/sec: 2802.48 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:27:11,119 epoch 10 - iter 104/138 - loss 0.01383309 - time (sec): 6.08 - samples/sec: 2813.68 - lr: 0.000002 - momentum: 0.000000
2023-10-17 08:27:11,847 epoch 10 - iter 117/138 - loss 0.01436897 - time (sec): 6.81 - samples/sec: 2843.50 - lr: 0.000001 - momentum: 0.000000
2023-10-17 08:27:12,582 epoch 10 - iter 130/138 - loss 0.01313075 - time (sec): 7.55 - samples/sec: 2871.93 - lr: 0.000000 - momentum: 0.000000
2023-10-17 08:27:13,015 ----------------------------------------------------------------------------------------------------
2023-10-17 08:27:13,016 EPOCH 10 done: loss 0.0135 - lr: 0.000000
2023-10-17 08:27:13,658 DEV : loss 0.19636619091033936 - f1-score (micro avg) 0.8948
2023-10-17 08:27:13,663 saving best model
2023-10-17 08:27:14,448 ----------------------------------------------------------------------------------------------------
2023-10-17 08:27:14,449 Loading model from best epoch ...
2023-10-17 08:27:15,783 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 08:27:16,610
Results:
- F-score (micro) 0.9146
- F-score (macro) 0.8801
- Accuracy 0.8509
By class:
precision recall f1-score support
scope 0.8920 0.8920 0.8920 176
pers 0.9690 0.9766 0.9728 128
work 0.8873 0.8514 0.8690 74
object 1.0000 1.0000 1.0000 2
loc 1.0000 0.5000 0.6667 2
micro avg 0.9182 0.9110 0.9146 382
macro avg 0.9497 0.8440 0.8801 382
weighted avg 0.9180 0.9110 0.9140 382
2023-10-17 08:27:16,611 ----------------------------------------------------------------------------------------------------
|