File size: 24,030 Bytes
ad21869 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 |
2023-10-13 08:26:08,641 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:08,642 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(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): BertEncoder(
(layer): ModuleList(
(0-11): 12 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(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): BertSelfOutput(
(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): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(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)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=768, out_features=768, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-13 08:26:08,643 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:08,643 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
2023-10-13 08:26:08,643 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:08,643 Train: 1100 sentences
2023-10-13 08:26:08,643 (train_with_dev=False, train_with_test=False)
2023-10-13 08:26:08,643 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:08,643 Training Params:
2023-10-13 08:26:08,643 - learning_rate: "3e-05"
2023-10-13 08:26:08,643 - mini_batch_size: "8"
2023-10-13 08:26:08,643 - max_epochs: "10"
2023-10-13 08:26:08,643 - shuffle: "True"
2023-10-13 08:26:08,643 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:08,643 Plugins:
2023-10-13 08:26:08,643 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 08:26:08,643 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:08,643 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 08:26:08,643 - metric: "('micro avg', 'f1-score')"
2023-10-13 08:26:08,643 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:08,643 Computation:
2023-10-13 08:26:08,644 - compute on device: cuda:0
2023-10-13 08:26:08,644 - embedding storage: none
2023-10-13 08:26:08,644 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:08,644 Model training base path: "hmbench-ajmc/de-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr3e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-13 08:26:08,644 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:08,644 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:09,422 epoch 1 - iter 13/138 - loss 3.69385232 - time (sec): 0.78 - samples/sec: 3038.15 - lr: 0.000003 - momentum: 0.000000
2023-10-13 08:26:10,209 epoch 1 - iter 26/138 - loss 3.48751350 - time (sec): 1.56 - samples/sec: 2794.15 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:26:10,915 epoch 1 - iter 39/138 - loss 3.14421571 - time (sec): 2.27 - samples/sec: 2861.72 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:26:11,700 epoch 1 - iter 52/138 - loss 2.55104060 - time (sec): 3.06 - samples/sec: 2958.09 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:26:12,383 epoch 1 - iter 65/138 - loss 2.27203720 - time (sec): 3.74 - samples/sec: 2925.20 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:26:13,075 epoch 1 - iter 78/138 - loss 2.05008068 - time (sec): 4.43 - samples/sec: 2937.45 - lr: 0.000017 - momentum: 0.000000
2023-10-13 08:26:13,765 epoch 1 - iter 91/138 - loss 1.90458869 - time (sec): 5.12 - samples/sec: 2929.88 - lr: 0.000020 - momentum: 0.000000
2023-10-13 08:26:14,469 epoch 1 - iter 104/138 - loss 1.74598608 - time (sec): 5.82 - samples/sec: 3002.95 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:26:15,175 epoch 1 - iter 117/138 - loss 1.62592389 - time (sec): 6.53 - samples/sec: 2999.69 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:26:15,861 epoch 1 - iter 130/138 - loss 1.53245007 - time (sec): 7.22 - samples/sec: 2988.19 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:26:16,273 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:16,273 EPOCH 1 done: loss 1.4877 - lr: 0.000028
2023-10-13 08:26:17,009 DEV : loss 0.4509204924106598 - f1-score (micro avg) 0.252
2023-10-13 08:26:17,014 saving best model
2023-10-13 08:26:17,324 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:18,094 epoch 2 - iter 13/138 - loss 0.42619151 - time (sec): 0.77 - samples/sec: 3139.00 - lr: 0.000030 - momentum: 0.000000
2023-10-13 08:26:18,813 epoch 2 - iter 26/138 - loss 0.42325569 - time (sec): 1.49 - samples/sec: 2842.22 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:26:19,574 epoch 2 - iter 39/138 - loss 0.40997183 - time (sec): 2.25 - samples/sec: 2873.00 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:26:20,308 epoch 2 - iter 52/138 - loss 0.38690384 - time (sec): 2.98 - samples/sec: 2847.84 - lr: 0.000029 - momentum: 0.000000
2023-10-13 08:26:21,088 epoch 2 - iter 65/138 - loss 0.35398854 - time (sec): 3.76 - samples/sec: 2828.03 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:26:21,829 epoch 2 - iter 78/138 - loss 0.34914707 - time (sec): 4.50 - samples/sec: 2839.82 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:26:22,619 epoch 2 - iter 91/138 - loss 0.33145623 - time (sec): 5.29 - samples/sec: 2804.89 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:26:23,354 epoch 2 - iter 104/138 - loss 0.32408274 - time (sec): 6.03 - samples/sec: 2836.09 - lr: 0.000028 - momentum: 0.000000
2023-10-13 08:26:24,107 epoch 2 - iter 117/138 - loss 0.31812163 - time (sec): 6.78 - samples/sec: 2840.83 - lr: 0.000027 - momentum: 0.000000
2023-10-13 08:26:24,862 epoch 2 - iter 130/138 - loss 0.30813736 - time (sec): 7.54 - samples/sec: 2864.21 - lr: 0.000027 - momentum: 0.000000
2023-10-13 08:26:25,257 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:25,257 EPOCH 2 done: loss 0.3027 - lr: 0.000027
2023-10-13 08:26:25,943 DEV : loss 0.17636282742023468 - f1-score (micro avg) 0.778
2023-10-13 08:26:25,948 saving best model
2023-10-13 08:26:26,398 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:27,114 epoch 3 - iter 13/138 - loss 0.19209402 - time (sec): 0.71 - samples/sec: 2745.27 - lr: 0.000026 - momentum: 0.000000
2023-10-13 08:26:27,844 epoch 3 - iter 26/138 - loss 0.17352554 - time (sec): 1.44 - samples/sec: 2823.15 - lr: 0.000026 - momentum: 0.000000
2023-10-13 08:26:28,634 epoch 3 - iter 39/138 - loss 0.15390368 - time (sec): 2.23 - samples/sec: 2883.66 - lr: 0.000026 - momentum: 0.000000
2023-10-13 08:26:29,333 epoch 3 - iter 52/138 - loss 0.15118087 - time (sec): 2.93 - samples/sec: 2943.74 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:26:30,113 epoch 3 - iter 65/138 - loss 0.14765733 - time (sec): 3.71 - samples/sec: 2924.73 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:26:30,808 epoch 3 - iter 78/138 - loss 0.14673277 - time (sec): 4.41 - samples/sec: 2912.56 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:26:31,557 epoch 3 - iter 91/138 - loss 0.14147220 - time (sec): 5.16 - samples/sec: 2928.81 - lr: 0.000025 - momentum: 0.000000
2023-10-13 08:26:32,243 epoch 3 - iter 104/138 - loss 0.14096598 - time (sec): 5.84 - samples/sec: 2920.42 - lr: 0.000024 - momentum: 0.000000
2023-10-13 08:26:32,960 epoch 3 - iter 117/138 - loss 0.14195431 - time (sec): 6.56 - samples/sec: 2918.85 - lr: 0.000024 - momentum: 0.000000
2023-10-13 08:26:33,700 epoch 3 - iter 130/138 - loss 0.14176228 - time (sec): 7.30 - samples/sec: 2926.93 - lr: 0.000024 - momentum: 0.000000
2023-10-13 08:26:34,155 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:34,156 EPOCH 3 done: loss 0.1380 - lr: 0.000024
2023-10-13 08:26:34,849 DEV : loss 0.143646240234375 - f1-score (micro avg) 0.8205
2023-10-13 08:26:34,855 saving best model
2023-10-13 08:26:35,284 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:35,970 epoch 4 - iter 13/138 - loss 0.08454532 - time (sec): 0.68 - samples/sec: 3075.26 - lr: 0.000023 - momentum: 0.000000
2023-10-13 08:26:36,679 epoch 4 - iter 26/138 - loss 0.08622317 - time (sec): 1.39 - samples/sec: 3188.79 - lr: 0.000023 - momentum: 0.000000
2023-10-13 08:26:37,419 epoch 4 - iter 39/138 - loss 0.08311552 - time (sec): 2.13 - samples/sec: 3020.59 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:26:38,143 epoch 4 - iter 52/138 - loss 0.08780046 - time (sec): 2.86 - samples/sec: 3015.55 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:26:38,891 epoch 4 - iter 65/138 - loss 0.09046928 - time (sec): 3.61 - samples/sec: 2989.97 - lr: 0.000022 - momentum: 0.000000
2023-10-13 08:26:39,603 epoch 4 - iter 78/138 - loss 0.08806674 - time (sec): 4.32 - samples/sec: 2967.54 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:26:40,326 epoch 4 - iter 91/138 - loss 0.09086554 - time (sec): 5.04 - samples/sec: 2956.46 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:26:41,041 epoch 4 - iter 104/138 - loss 0.09148364 - time (sec): 5.75 - samples/sec: 2937.77 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:26:41,774 epoch 4 - iter 117/138 - loss 0.08793837 - time (sec): 6.49 - samples/sec: 2965.76 - lr: 0.000021 - momentum: 0.000000
2023-10-13 08:26:42,559 epoch 4 - iter 130/138 - loss 0.08716024 - time (sec): 7.27 - samples/sec: 2967.65 - lr: 0.000020 - momentum: 0.000000
2023-10-13 08:26:42,991 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:42,991 EPOCH 4 done: loss 0.0879 - lr: 0.000020
2023-10-13 08:26:43,743 DEV : loss 0.11688197404146194 - f1-score (micro avg) 0.8363
2023-10-13 08:26:43,751 saving best model
2023-10-13 08:26:44,217 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:44,914 epoch 5 - iter 13/138 - loss 0.06018758 - time (sec): 0.70 - samples/sec: 3152.46 - lr: 0.000020 - momentum: 0.000000
2023-10-13 08:26:45,613 epoch 5 - iter 26/138 - loss 0.05888432 - time (sec): 1.40 - samples/sec: 3154.60 - lr: 0.000019 - momentum: 0.000000
2023-10-13 08:26:46,320 epoch 5 - iter 39/138 - loss 0.07909944 - time (sec): 2.10 - samples/sec: 3081.62 - lr: 0.000019 - momentum: 0.000000
2023-10-13 08:26:47,012 epoch 5 - iter 52/138 - loss 0.07517801 - time (sec): 2.79 - samples/sec: 3044.85 - lr: 0.000019 - momentum: 0.000000
2023-10-13 08:26:47,695 epoch 5 - iter 65/138 - loss 0.07414495 - time (sec): 3.48 - samples/sec: 3081.70 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:26:48,384 epoch 5 - iter 78/138 - loss 0.06957352 - time (sec): 4.17 - samples/sec: 3091.23 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:26:49,177 epoch 5 - iter 91/138 - loss 0.07023772 - time (sec): 4.96 - samples/sec: 3029.09 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:26:49,892 epoch 5 - iter 104/138 - loss 0.06663683 - time (sec): 5.67 - samples/sec: 3001.18 - lr: 0.000018 - momentum: 0.000000
2023-10-13 08:26:50,631 epoch 5 - iter 117/138 - loss 0.06142101 - time (sec): 6.41 - samples/sec: 2987.60 - lr: 0.000017 - momentum: 0.000000
2023-10-13 08:26:51,381 epoch 5 - iter 130/138 - loss 0.06264685 - time (sec): 7.16 - samples/sec: 2981.96 - lr: 0.000017 - momentum: 0.000000
2023-10-13 08:26:51,831 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:51,831 EPOCH 5 done: loss 0.0616 - lr: 0.000017
2023-10-13 08:26:52,525 DEV : loss 0.13779088854789734 - f1-score (micro avg) 0.8491
2023-10-13 08:26:52,529 saving best model
2023-10-13 08:26:52,961 ----------------------------------------------------------------------------------------------------
2023-10-13 08:26:53,703 epoch 6 - iter 13/138 - loss 0.04866097 - time (sec): 0.74 - samples/sec: 2926.90 - lr: 0.000016 - momentum: 0.000000
2023-10-13 08:26:54,438 epoch 6 - iter 26/138 - loss 0.04983253 - time (sec): 1.47 - samples/sec: 3002.73 - lr: 0.000016 - momentum: 0.000000
2023-10-13 08:26:55,179 epoch 6 - iter 39/138 - loss 0.04671730 - time (sec): 2.21 - samples/sec: 2963.62 - lr: 0.000016 - momentum: 0.000000
2023-10-13 08:26:55,881 epoch 6 - iter 52/138 - loss 0.04256161 - time (sec): 2.91 - samples/sec: 2936.14 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:26:56,642 epoch 6 - iter 65/138 - loss 0.04915792 - time (sec): 3.68 - samples/sec: 2898.60 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:26:57,370 epoch 6 - iter 78/138 - loss 0.05052720 - time (sec): 4.40 - samples/sec: 2902.03 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:26:58,105 epoch 6 - iter 91/138 - loss 0.04712163 - time (sec): 5.14 - samples/sec: 2912.36 - lr: 0.000015 - momentum: 0.000000
2023-10-13 08:26:58,791 epoch 6 - iter 104/138 - loss 0.04683642 - time (sec): 5.82 - samples/sec: 2920.37 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:26:59,545 epoch 6 - iter 117/138 - loss 0.04608847 - time (sec): 6.58 - samples/sec: 2921.76 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:27:00,304 epoch 6 - iter 130/138 - loss 0.04480660 - time (sec): 7.34 - samples/sec: 2928.18 - lr: 0.000014 - momentum: 0.000000
2023-10-13 08:27:00,745 ----------------------------------------------------------------------------------------------------
2023-10-13 08:27:00,745 EPOCH 6 done: loss 0.0445 - lr: 0.000014
2023-10-13 08:27:01,427 DEV : loss 0.15820646286010742 - f1-score (micro avg) 0.8578
2023-10-13 08:27:01,432 saving best model
2023-10-13 08:27:01,866 ----------------------------------------------------------------------------------------------------
2023-10-13 08:27:02,596 epoch 7 - iter 13/138 - loss 0.01760101 - time (sec): 0.73 - samples/sec: 2835.15 - lr: 0.000013 - momentum: 0.000000
2023-10-13 08:27:03,364 epoch 7 - iter 26/138 - loss 0.01763718 - time (sec): 1.50 - samples/sec: 2876.71 - lr: 0.000013 - momentum: 0.000000
2023-10-13 08:27:04,079 epoch 7 - iter 39/138 - loss 0.01604980 - time (sec): 2.21 - samples/sec: 2822.15 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:27:04,825 epoch 7 - iter 52/138 - loss 0.02286619 - time (sec): 2.96 - samples/sec: 2914.57 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:27:05,547 epoch 7 - iter 65/138 - loss 0.02495223 - time (sec): 3.68 - samples/sec: 2919.90 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:27:06,256 epoch 7 - iter 78/138 - loss 0.03142391 - time (sec): 4.39 - samples/sec: 2942.85 - lr: 0.000012 - momentum: 0.000000
2023-10-13 08:27:06,968 epoch 7 - iter 91/138 - loss 0.03459948 - time (sec): 5.10 - samples/sec: 2956.26 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:27:07,691 epoch 7 - iter 104/138 - loss 0.04089677 - time (sec): 5.82 - samples/sec: 2977.62 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:27:08,418 epoch 7 - iter 117/138 - loss 0.03890265 - time (sec): 6.55 - samples/sec: 2961.50 - lr: 0.000011 - momentum: 0.000000
2023-10-13 08:27:09,142 epoch 7 - iter 130/138 - loss 0.03775231 - time (sec): 7.27 - samples/sec: 2945.83 - lr: 0.000010 - momentum: 0.000000
2023-10-13 08:27:09,593 ----------------------------------------------------------------------------------------------------
2023-10-13 08:27:09,593 EPOCH 7 done: loss 0.0364 - lr: 0.000010
2023-10-13 08:27:10,332 DEV : loss 0.16149461269378662 - f1-score (micro avg) 0.8694
2023-10-13 08:27:10,341 saving best model
2023-10-13 08:27:10,872 ----------------------------------------------------------------------------------------------------
2023-10-13 08:27:11,670 epoch 8 - iter 13/138 - loss 0.02451766 - time (sec): 0.78 - samples/sec: 2718.28 - lr: 0.000010 - momentum: 0.000000
2023-10-13 08:27:12,481 epoch 8 - iter 26/138 - loss 0.01900098 - time (sec): 1.59 - samples/sec: 2654.23 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:27:13,274 epoch 8 - iter 39/138 - loss 0.03325250 - time (sec): 2.39 - samples/sec: 2719.54 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:27:14,039 epoch 8 - iter 52/138 - loss 0.03077508 - time (sec): 3.15 - samples/sec: 2655.42 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:27:14,872 epoch 8 - iter 65/138 - loss 0.02914906 - time (sec): 3.99 - samples/sec: 2673.98 - lr: 0.000009 - momentum: 0.000000
2023-10-13 08:27:15,649 epoch 8 - iter 78/138 - loss 0.02905655 - time (sec): 4.76 - samples/sec: 2691.84 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:27:16,390 epoch 8 - iter 91/138 - loss 0.03090136 - time (sec): 5.50 - samples/sec: 2750.11 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:27:17,116 epoch 8 - iter 104/138 - loss 0.02860076 - time (sec): 6.23 - samples/sec: 2779.83 - lr: 0.000008 - momentum: 0.000000
2023-10-13 08:27:17,831 epoch 8 - iter 117/138 - loss 0.02699396 - time (sec): 6.95 - samples/sec: 2788.13 - lr: 0.000007 - momentum: 0.000000
2023-10-13 08:27:18,668 epoch 8 - iter 130/138 - loss 0.02432561 - time (sec): 7.78 - samples/sec: 2791.72 - lr: 0.000007 - momentum: 0.000000
2023-10-13 08:27:19,073 ----------------------------------------------------------------------------------------------------
2023-10-13 08:27:19,073 EPOCH 8 done: loss 0.0255 - lr: 0.000007
2023-10-13 08:27:19,745 DEV : loss 0.16744698584079742 - f1-score (micro avg) 0.8701
2023-10-13 08:27:19,751 saving best model
2023-10-13 08:27:20,202 ----------------------------------------------------------------------------------------------------
2023-10-13 08:27:20,944 epoch 9 - iter 13/138 - loss 0.00538376 - time (sec): 0.73 - samples/sec: 3011.35 - lr: 0.000006 - momentum: 0.000000
2023-10-13 08:27:21,667 epoch 9 - iter 26/138 - loss 0.02331938 - time (sec): 1.45 - samples/sec: 2964.44 - lr: 0.000006 - momentum: 0.000000
2023-10-13 08:27:22,408 epoch 9 - iter 39/138 - loss 0.01893327 - time (sec): 2.19 - samples/sec: 2859.27 - lr: 0.000006 - momentum: 0.000000
2023-10-13 08:27:23,198 epoch 9 - iter 52/138 - loss 0.01939908 - time (sec): 2.98 - samples/sec: 2919.64 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:27:23,945 epoch 9 - iter 65/138 - loss 0.02382819 - time (sec): 3.73 - samples/sec: 2912.31 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:27:24,724 epoch 9 - iter 78/138 - loss 0.02378557 - time (sec): 4.51 - samples/sec: 2905.29 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:27:25,481 epoch 9 - iter 91/138 - loss 0.02458101 - time (sec): 5.27 - samples/sec: 2872.17 - lr: 0.000005 - momentum: 0.000000
2023-10-13 08:27:26,237 epoch 9 - iter 104/138 - loss 0.02381904 - time (sec): 6.02 - samples/sec: 2883.72 - lr: 0.000004 - momentum: 0.000000
2023-10-13 08:27:26,959 epoch 9 - iter 117/138 - loss 0.02435136 - time (sec): 6.74 - samples/sec: 2889.26 - lr: 0.000004 - momentum: 0.000000
2023-10-13 08:27:27,656 epoch 9 - iter 130/138 - loss 0.02457447 - time (sec): 7.44 - samples/sec: 2874.68 - lr: 0.000004 - momentum: 0.000000
2023-10-13 08:27:28,130 ----------------------------------------------------------------------------------------------------
2023-10-13 08:27:28,130 EPOCH 9 done: loss 0.0244 - lr: 0.000004
2023-10-13 08:27:28,844 DEV : loss 0.16501910984516144 - f1-score (micro avg) 0.8741
2023-10-13 08:27:28,850 saving best model
2023-10-13 08:27:29,280 ----------------------------------------------------------------------------------------------------
2023-10-13 08:27:30,048 epoch 10 - iter 13/138 - loss 0.02703412 - time (sec): 0.77 - samples/sec: 3082.15 - lr: 0.000003 - momentum: 0.000000
2023-10-13 08:27:30,794 epoch 10 - iter 26/138 - loss 0.01688742 - time (sec): 1.51 - samples/sec: 2890.78 - lr: 0.000003 - momentum: 0.000000
2023-10-13 08:27:31,538 epoch 10 - iter 39/138 - loss 0.01465273 - time (sec): 2.26 - samples/sec: 2855.54 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:27:32,259 epoch 10 - iter 52/138 - loss 0.01768204 - time (sec): 2.98 - samples/sec: 2879.42 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:27:32,978 epoch 10 - iter 65/138 - loss 0.01594072 - time (sec): 3.70 - samples/sec: 2851.87 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:27:33,760 epoch 10 - iter 78/138 - loss 0.01490758 - time (sec): 4.48 - samples/sec: 2852.30 - lr: 0.000002 - momentum: 0.000000
2023-10-13 08:27:34,525 epoch 10 - iter 91/138 - loss 0.01513863 - time (sec): 5.24 - samples/sec: 2876.63 - lr: 0.000001 - momentum: 0.000000
2023-10-13 08:27:35,251 epoch 10 - iter 104/138 - loss 0.01506427 - time (sec): 5.97 - samples/sec: 2910.48 - lr: 0.000001 - momentum: 0.000000
2023-10-13 08:27:35,944 epoch 10 - iter 117/138 - loss 0.01598269 - time (sec): 6.66 - samples/sec: 2900.63 - lr: 0.000001 - momentum: 0.000000
2023-10-13 08:27:36,673 epoch 10 - iter 130/138 - loss 0.01793335 - time (sec): 7.39 - samples/sec: 2895.22 - lr: 0.000000 - momentum: 0.000000
2023-10-13 08:27:37,116 ----------------------------------------------------------------------------------------------------
2023-10-13 08:27:37,116 EPOCH 10 done: loss 0.0179 - lr: 0.000000
2023-10-13 08:27:37,847 DEV : loss 0.16583986580371857 - f1-score (micro avg) 0.8694
2023-10-13 08:27:38,184 ----------------------------------------------------------------------------------------------------
2023-10-13 08:27:38,185 Loading model from best epoch ...
2023-10-13 08:27:39,730 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
2023-10-13 08:27:40,442
Results:
- F-score (micro) 0.89
- F-score (macro) 0.6905
- Accuracy 0.819
By class:
precision recall f1-score support
scope 0.8649 0.9091 0.8864 176
pers 0.9380 0.9453 0.9416 128
work 0.8243 0.8243 0.8243 74
loc 0.6667 1.0000 0.8000 2
object 0.0000 0.0000 0.0000 2
micro avg 0.8798 0.9005 0.8900 382
macro avg 0.6588 0.7357 0.6905 382
weighted avg 0.8759 0.9005 0.8878 382
2023-10-13 08:27:40,442 ----------------------------------------------------------------------------------------------------
|