File size: 24,105 Bytes
afe905b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 |
2023-10-15 22:27:17,165 ----------------------------------------------------------------------------------------------------
2023-10-15 22:27:17,166 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=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-15 22:27:17,166 ----------------------------------------------------------------------------------------------------
2023-10-15 22:27:17,167 MultiCorpus: 20847 train + 1123 dev + 3350 test sentences
- NER_HIPE_2022 Corpus: 20847 train + 1123 dev + 3350 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/de/with_doc_seperator
2023-10-15 22:27:17,167 ----------------------------------------------------------------------------------------------------
2023-10-15 22:27:17,167 Train: 20847 sentences
2023-10-15 22:27:17,167 (train_with_dev=False, train_with_test=False)
2023-10-15 22:27:17,167 ----------------------------------------------------------------------------------------------------
2023-10-15 22:27:17,167 Training Params:
2023-10-15 22:27:17,167 - learning_rate: "5e-05"
2023-10-15 22:27:17,167 - mini_batch_size: "8"
2023-10-15 22:27:17,167 - max_epochs: "10"
2023-10-15 22:27:17,167 - shuffle: "True"
2023-10-15 22:27:17,167 ----------------------------------------------------------------------------------------------------
2023-10-15 22:27:17,167 Plugins:
2023-10-15 22:27:17,167 - LinearScheduler | warmup_fraction: '0.1'
2023-10-15 22:27:17,167 ----------------------------------------------------------------------------------------------------
2023-10-15 22:27:17,167 Final evaluation on model from best epoch (best-model.pt)
2023-10-15 22:27:17,167 - metric: "('micro avg', 'f1-score')"
2023-10-15 22:27:17,167 ----------------------------------------------------------------------------------------------------
2023-10-15 22:27:17,167 Computation:
2023-10-15 22:27:17,167 - compute on device: cuda:0
2023-10-15 22:27:17,167 - embedding storage: none
2023-10-15 22:27:17,167 ----------------------------------------------------------------------------------------------------
2023-10-15 22:27:17,167 Model training base path: "hmbench-newseye/de-dbmdz/bert-base-historic-multilingual-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-5"
2023-10-15 22:27:17,167 ----------------------------------------------------------------------------------------------------
2023-10-15 22:27:17,167 ----------------------------------------------------------------------------------------------------
2023-10-15 22:27:35,255 epoch 1 - iter 260/2606 - loss 1.55162857 - time (sec): 18.09 - samples/sec: 1967.99 - lr: 0.000005 - momentum: 0.000000
2023-10-15 22:27:54,274 epoch 1 - iter 520/2606 - loss 0.96412232 - time (sec): 37.11 - samples/sec: 1964.22 - lr: 0.000010 - momentum: 0.000000
2023-10-15 22:28:12,959 epoch 1 - iter 780/2606 - loss 0.73562218 - time (sec): 55.79 - samples/sec: 1943.27 - lr: 0.000015 - momentum: 0.000000
2023-10-15 22:28:31,659 epoch 1 - iter 1040/2606 - loss 0.61663296 - time (sec): 74.49 - samples/sec: 1932.82 - lr: 0.000020 - momentum: 0.000000
2023-10-15 22:28:50,994 epoch 1 - iter 1300/2606 - loss 0.53611543 - time (sec): 93.83 - samples/sec: 1924.35 - lr: 0.000025 - momentum: 0.000000
2023-10-15 22:29:09,993 epoch 1 - iter 1560/2606 - loss 0.47717725 - time (sec): 112.83 - samples/sec: 1932.36 - lr: 0.000030 - momentum: 0.000000
2023-10-15 22:29:28,212 epoch 1 - iter 1820/2606 - loss 0.43977687 - time (sec): 131.04 - samples/sec: 1946.37 - lr: 0.000035 - momentum: 0.000000
2023-10-15 22:29:47,039 epoch 1 - iter 2080/2606 - loss 0.40953012 - time (sec): 149.87 - samples/sec: 1941.48 - lr: 0.000040 - momentum: 0.000000
2023-10-15 22:30:05,821 epoch 1 - iter 2340/2606 - loss 0.38758796 - time (sec): 168.65 - samples/sec: 1935.49 - lr: 0.000045 - momentum: 0.000000
2023-10-15 22:30:25,851 epoch 1 - iter 2600/2606 - loss 0.36513327 - time (sec): 188.68 - samples/sec: 1941.95 - lr: 0.000050 - momentum: 0.000000
2023-10-15 22:30:26,350 ----------------------------------------------------------------------------------------------------
2023-10-15 22:30:26,350 EPOCH 1 done: loss 0.3646 - lr: 0.000050
2023-10-15 22:30:33,057 DEV : loss 0.12843316793441772 - f1-score (micro avg) 0.316
2023-10-15 22:30:33,086 saving best model
2023-10-15 22:30:33,466 ----------------------------------------------------------------------------------------------------
2023-10-15 22:30:53,080 epoch 2 - iter 260/2606 - loss 0.16328381 - time (sec): 19.61 - samples/sec: 1977.33 - lr: 0.000049 - momentum: 0.000000
2023-10-15 22:31:11,726 epoch 2 - iter 520/2606 - loss 0.15786652 - time (sec): 38.26 - samples/sec: 1960.14 - lr: 0.000049 - momentum: 0.000000
2023-10-15 22:31:30,523 epoch 2 - iter 780/2606 - loss 0.15096452 - time (sec): 57.06 - samples/sec: 1953.36 - lr: 0.000048 - momentum: 0.000000
2023-10-15 22:31:49,690 epoch 2 - iter 1040/2606 - loss 0.15162785 - time (sec): 76.22 - samples/sec: 1951.95 - lr: 0.000048 - momentum: 0.000000
2023-10-15 22:32:08,501 epoch 2 - iter 1300/2606 - loss 0.15521001 - time (sec): 95.03 - samples/sec: 1945.47 - lr: 0.000047 - momentum: 0.000000
2023-10-15 22:32:27,507 epoch 2 - iter 1560/2606 - loss 0.15206366 - time (sec): 114.04 - samples/sec: 1949.84 - lr: 0.000047 - momentum: 0.000000
2023-10-15 22:32:45,650 epoch 2 - iter 1820/2606 - loss 0.15321996 - time (sec): 132.18 - samples/sec: 1952.09 - lr: 0.000046 - momentum: 0.000000
2023-10-15 22:33:05,338 epoch 2 - iter 2080/2606 - loss 0.15196432 - time (sec): 151.87 - samples/sec: 1955.77 - lr: 0.000046 - momentum: 0.000000
2023-10-15 22:33:22,997 epoch 2 - iter 2340/2606 - loss 0.15199041 - time (sec): 169.53 - samples/sec: 1948.27 - lr: 0.000045 - momentum: 0.000000
2023-10-15 22:33:41,251 epoch 2 - iter 2600/2606 - loss 0.15242040 - time (sec): 187.78 - samples/sec: 1953.00 - lr: 0.000044 - momentum: 0.000000
2023-10-15 22:33:41,598 ----------------------------------------------------------------------------------------------------
2023-10-15 22:33:41,598 EPOCH 2 done: loss 0.1525 - lr: 0.000044
2023-10-15 22:33:49,877 DEV : loss 0.15268105268478394 - f1-score (micro avg) 0.3282
2023-10-15 22:33:49,905 saving best model
2023-10-15 22:33:51,198 ----------------------------------------------------------------------------------------------------
2023-10-15 22:34:09,869 epoch 3 - iter 260/2606 - loss 0.13841022 - time (sec): 18.67 - samples/sec: 1935.44 - lr: 0.000044 - momentum: 0.000000
2023-10-15 22:34:27,325 epoch 3 - iter 520/2606 - loss 0.12007742 - time (sec): 36.12 - samples/sec: 1904.26 - lr: 0.000043 - momentum: 0.000000
2023-10-15 22:34:45,146 epoch 3 - iter 780/2606 - loss 0.11917281 - time (sec): 53.94 - samples/sec: 1908.77 - lr: 0.000043 - momentum: 0.000000
2023-10-15 22:35:03,249 epoch 3 - iter 1040/2606 - loss 0.11929376 - time (sec): 72.05 - samples/sec: 1927.33 - lr: 0.000042 - momentum: 0.000000
2023-10-15 22:35:22,165 epoch 3 - iter 1300/2606 - loss 0.11370851 - time (sec): 90.96 - samples/sec: 1935.39 - lr: 0.000042 - momentum: 0.000000
2023-10-15 22:35:41,375 epoch 3 - iter 1560/2606 - loss 0.11243711 - time (sec): 110.17 - samples/sec: 1934.73 - lr: 0.000041 - momentum: 0.000000
2023-10-15 22:36:00,749 epoch 3 - iter 1820/2606 - loss 0.11154229 - time (sec): 129.55 - samples/sec: 1938.83 - lr: 0.000041 - momentum: 0.000000
2023-10-15 22:36:19,809 epoch 3 - iter 2080/2606 - loss 0.11088653 - time (sec): 148.61 - samples/sec: 1934.20 - lr: 0.000040 - momentum: 0.000000
2023-10-15 22:36:39,165 epoch 3 - iter 2340/2606 - loss 0.11020741 - time (sec): 167.96 - samples/sec: 1942.87 - lr: 0.000039 - momentum: 0.000000
2023-10-15 22:36:59,231 epoch 3 - iter 2600/2606 - loss 0.10875412 - time (sec): 188.03 - samples/sec: 1951.27 - lr: 0.000039 - momentum: 0.000000
2023-10-15 22:36:59,592 ----------------------------------------------------------------------------------------------------
2023-10-15 22:36:59,592 EPOCH 3 done: loss 0.1089 - lr: 0.000039
2023-10-15 22:37:07,822 DEV : loss 0.1973780244588852 - f1-score (micro avg) 0.3449
2023-10-15 22:37:07,851 saving best model
2023-10-15 22:37:08,459 ----------------------------------------------------------------------------------------------------
2023-10-15 22:37:26,324 epoch 4 - iter 260/2606 - loss 0.06966027 - time (sec): 17.86 - samples/sec: 1961.16 - lr: 0.000038 - momentum: 0.000000
2023-10-15 22:37:44,366 epoch 4 - iter 520/2606 - loss 0.07834214 - time (sec): 35.91 - samples/sec: 1997.72 - lr: 0.000038 - momentum: 0.000000
2023-10-15 22:38:04,320 epoch 4 - iter 780/2606 - loss 0.07562032 - time (sec): 55.86 - samples/sec: 1973.69 - lr: 0.000037 - momentum: 0.000000
2023-10-15 22:38:24,283 epoch 4 - iter 1040/2606 - loss 0.07734676 - time (sec): 75.82 - samples/sec: 1973.03 - lr: 0.000037 - momentum: 0.000000
2023-10-15 22:38:43,354 epoch 4 - iter 1300/2606 - loss 0.07949064 - time (sec): 94.89 - samples/sec: 1971.18 - lr: 0.000036 - momentum: 0.000000
2023-10-15 22:39:01,902 epoch 4 - iter 1560/2606 - loss 0.08030660 - time (sec): 113.44 - samples/sec: 1963.18 - lr: 0.000036 - momentum: 0.000000
2023-10-15 22:39:21,242 epoch 4 - iter 1820/2606 - loss 0.07936732 - time (sec): 132.78 - samples/sec: 1949.28 - lr: 0.000035 - momentum: 0.000000
2023-10-15 22:39:40,653 epoch 4 - iter 2080/2606 - loss 0.07794236 - time (sec): 152.19 - samples/sec: 1948.31 - lr: 0.000034 - momentum: 0.000000
2023-10-15 22:39:58,840 epoch 4 - iter 2340/2606 - loss 0.07686992 - time (sec): 170.38 - samples/sec: 1942.92 - lr: 0.000034 - momentum: 0.000000
2023-10-15 22:40:17,812 epoch 4 - iter 2600/2606 - loss 0.07640796 - time (sec): 189.35 - samples/sec: 1936.72 - lr: 0.000033 - momentum: 0.000000
2023-10-15 22:40:18,202 ----------------------------------------------------------------------------------------------------
2023-10-15 22:40:18,203 EPOCH 4 done: loss 0.0764 - lr: 0.000033
2023-10-15 22:40:26,731 DEV : loss 0.2757483124732971 - f1-score (micro avg) 0.3322
2023-10-15 22:40:26,762 ----------------------------------------------------------------------------------------------------
2023-10-15 22:40:46,807 epoch 5 - iter 260/2606 - loss 0.05178851 - time (sec): 20.04 - samples/sec: 1921.49 - lr: 0.000033 - momentum: 0.000000
2023-10-15 22:41:05,169 epoch 5 - iter 520/2606 - loss 0.05717423 - time (sec): 38.41 - samples/sec: 1904.14 - lr: 0.000032 - momentum: 0.000000
2023-10-15 22:41:23,526 epoch 5 - iter 780/2606 - loss 0.05742065 - time (sec): 56.76 - samples/sec: 1913.71 - lr: 0.000032 - momentum: 0.000000
2023-10-15 22:41:43,051 epoch 5 - iter 1040/2606 - loss 0.05696528 - time (sec): 76.29 - samples/sec: 1931.89 - lr: 0.000031 - momentum: 0.000000
2023-10-15 22:42:03,807 epoch 5 - iter 1300/2606 - loss 0.05634791 - time (sec): 97.04 - samples/sec: 1912.72 - lr: 0.000031 - momentum: 0.000000
2023-10-15 22:42:22,326 epoch 5 - iter 1560/2606 - loss 0.05592339 - time (sec): 115.56 - samples/sec: 1914.46 - lr: 0.000030 - momentum: 0.000000
2023-10-15 22:42:41,680 epoch 5 - iter 1820/2606 - loss 0.05667760 - time (sec): 134.92 - samples/sec: 1899.98 - lr: 0.000029 - momentum: 0.000000
2023-10-15 22:43:01,527 epoch 5 - iter 2080/2606 - loss 0.05618848 - time (sec): 154.76 - samples/sec: 1902.76 - lr: 0.000029 - momentum: 0.000000
2023-10-15 22:43:19,937 epoch 5 - iter 2340/2606 - loss 0.05550520 - time (sec): 173.17 - samples/sec: 1908.88 - lr: 0.000028 - momentum: 0.000000
2023-10-15 22:43:38,771 epoch 5 - iter 2600/2606 - loss 0.05582007 - time (sec): 192.01 - samples/sec: 1909.87 - lr: 0.000028 - momentum: 0.000000
2023-10-15 22:43:39,185 ----------------------------------------------------------------------------------------------------
2023-10-15 22:43:39,185 EPOCH 5 done: loss 0.0558 - lr: 0.000028
2023-10-15 22:43:47,656 DEV : loss 0.2588236331939697 - f1-score (micro avg) 0.3881
2023-10-15 22:43:47,689 saving best model
2023-10-15 22:43:48,321 ----------------------------------------------------------------------------------------------------
2023-10-15 22:44:07,112 epoch 6 - iter 260/2606 - loss 0.03581917 - time (sec): 18.79 - samples/sec: 1939.80 - lr: 0.000027 - momentum: 0.000000
2023-10-15 22:44:25,901 epoch 6 - iter 520/2606 - loss 0.03857227 - time (sec): 37.58 - samples/sec: 1955.95 - lr: 0.000027 - momentum: 0.000000
2023-10-15 22:44:45,197 epoch 6 - iter 780/2606 - loss 0.03718849 - time (sec): 56.87 - samples/sec: 1938.69 - lr: 0.000026 - momentum: 0.000000
2023-10-15 22:45:04,355 epoch 6 - iter 1040/2606 - loss 0.03795267 - time (sec): 76.03 - samples/sec: 1941.79 - lr: 0.000026 - momentum: 0.000000
2023-10-15 22:45:22,754 epoch 6 - iter 1300/2606 - loss 0.03927938 - time (sec): 94.43 - samples/sec: 1939.52 - lr: 0.000025 - momentum: 0.000000
2023-10-15 22:45:41,969 epoch 6 - iter 1560/2606 - loss 0.04094059 - time (sec): 113.64 - samples/sec: 1940.09 - lr: 0.000024 - momentum: 0.000000
2023-10-15 22:46:02,357 epoch 6 - iter 1820/2606 - loss 0.04042786 - time (sec): 134.03 - samples/sec: 1930.67 - lr: 0.000024 - momentum: 0.000000
2023-10-15 22:46:21,777 epoch 6 - iter 2080/2606 - loss 0.04023301 - time (sec): 153.45 - samples/sec: 1932.81 - lr: 0.000023 - momentum: 0.000000
2023-10-15 22:46:41,120 epoch 6 - iter 2340/2606 - loss 0.04049483 - time (sec): 172.80 - samples/sec: 1926.06 - lr: 0.000023 - momentum: 0.000000
2023-10-15 22:46:59,172 epoch 6 - iter 2600/2606 - loss 0.04034539 - time (sec): 190.85 - samples/sec: 1918.45 - lr: 0.000022 - momentum: 0.000000
2023-10-15 22:46:59,704 ----------------------------------------------------------------------------------------------------
2023-10-15 22:46:59,704 EPOCH 6 done: loss 0.0404 - lr: 0.000022
2023-10-15 22:47:07,922 DEV : loss 0.32239729166030884 - f1-score (micro avg) 0.372
2023-10-15 22:47:07,949 ----------------------------------------------------------------------------------------------------
2023-10-15 22:47:25,980 epoch 7 - iter 260/2606 - loss 0.03480989 - time (sec): 18.03 - samples/sec: 1933.48 - lr: 0.000022 - momentum: 0.000000
2023-10-15 22:47:44,510 epoch 7 - iter 520/2606 - loss 0.02997614 - time (sec): 36.56 - samples/sec: 1954.61 - lr: 0.000021 - momentum: 0.000000
2023-10-15 22:48:04,165 epoch 7 - iter 780/2606 - loss 0.03233103 - time (sec): 56.21 - samples/sec: 1925.02 - lr: 0.000021 - momentum: 0.000000
2023-10-15 22:48:24,145 epoch 7 - iter 1040/2606 - loss 0.03295900 - time (sec): 76.19 - samples/sec: 1934.66 - lr: 0.000020 - momentum: 0.000000
2023-10-15 22:48:43,108 epoch 7 - iter 1300/2606 - loss 0.03274939 - time (sec): 95.16 - samples/sec: 1926.80 - lr: 0.000019 - momentum: 0.000000
2023-10-15 22:49:02,094 epoch 7 - iter 1560/2606 - loss 0.03398439 - time (sec): 114.14 - samples/sec: 1930.83 - lr: 0.000019 - momentum: 0.000000
2023-10-15 22:49:21,082 epoch 7 - iter 1820/2606 - loss 0.03293724 - time (sec): 133.13 - samples/sec: 1940.00 - lr: 0.000018 - momentum: 0.000000
2023-10-15 22:49:40,117 epoch 7 - iter 2080/2606 - loss 0.03190338 - time (sec): 152.17 - samples/sec: 1943.14 - lr: 0.000018 - momentum: 0.000000
2023-10-15 22:49:58,402 epoch 7 - iter 2340/2606 - loss 0.03087084 - time (sec): 170.45 - samples/sec: 1932.03 - lr: 0.000017 - momentum: 0.000000
2023-10-15 22:50:18,500 epoch 7 - iter 2600/2606 - loss 0.03081740 - time (sec): 190.55 - samples/sec: 1922.45 - lr: 0.000017 - momentum: 0.000000
2023-10-15 22:50:19,063 ----------------------------------------------------------------------------------------------------
2023-10-15 22:50:19,063 EPOCH 7 done: loss 0.0308 - lr: 0.000017
2023-10-15 22:50:27,425 DEV : loss 0.4393313229084015 - f1-score (micro avg) 0.3524
2023-10-15 22:50:27,461 ----------------------------------------------------------------------------------------------------
2023-10-15 22:50:47,488 epoch 8 - iter 260/2606 - loss 0.02062903 - time (sec): 20.03 - samples/sec: 1948.84 - lr: 0.000016 - momentum: 0.000000
2023-10-15 22:51:06,776 epoch 8 - iter 520/2606 - loss 0.02059864 - time (sec): 39.31 - samples/sec: 1957.39 - lr: 0.000016 - momentum: 0.000000
2023-10-15 22:51:25,601 epoch 8 - iter 780/2606 - loss 0.02243684 - time (sec): 58.14 - samples/sec: 1942.91 - lr: 0.000015 - momentum: 0.000000
2023-10-15 22:51:44,843 epoch 8 - iter 1040/2606 - loss 0.02218227 - time (sec): 77.38 - samples/sec: 1932.78 - lr: 0.000014 - momentum: 0.000000
2023-10-15 22:52:04,003 epoch 8 - iter 1300/2606 - loss 0.02168277 - time (sec): 96.54 - samples/sec: 1928.59 - lr: 0.000014 - momentum: 0.000000
2023-10-15 22:52:23,397 epoch 8 - iter 1560/2606 - loss 0.02361754 - time (sec): 115.93 - samples/sec: 1929.53 - lr: 0.000013 - momentum: 0.000000
2023-10-15 22:52:41,555 epoch 8 - iter 1820/2606 - loss 0.02366463 - time (sec): 134.09 - samples/sec: 1934.57 - lr: 0.000013 - momentum: 0.000000
2023-10-15 22:53:00,142 epoch 8 - iter 2080/2606 - loss 0.02356939 - time (sec): 152.68 - samples/sec: 1934.05 - lr: 0.000012 - momentum: 0.000000
2023-10-15 22:53:18,836 epoch 8 - iter 2340/2606 - loss 0.02340726 - time (sec): 171.37 - samples/sec: 1924.57 - lr: 0.000012 - momentum: 0.000000
2023-10-15 22:53:37,933 epoch 8 - iter 2600/2606 - loss 0.02280304 - time (sec): 190.47 - samples/sec: 1926.00 - lr: 0.000011 - momentum: 0.000000
2023-10-15 22:53:38,313 ----------------------------------------------------------------------------------------------------
2023-10-15 22:53:38,313 EPOCH 8 done: loss 0.0228 - lr: 0.000011
2023-10-15 22:53:47,439 DEV : loss 0.4546719491481781 - f1-score (micro avg) 0.3671
2023-10-15 22:53:47,466 ----------------------------------------------------------------------------------------------------
2023-10-15 22:54:06,558 epoch 9 - iter 260/2606 - loss 0.01469418 - time (sec): 19.09 - samples/sec: 1971.09 - lr: 0.000011 - momentum: 0.000000
2023-10-15 22:54:24,684 epoch 9 - iter 520/2606 - loss 0.01564107 - time (sec): 37.22 - samples/sec: 1948.48 - lr: 0.000010 - momentum: 0.000000
2023-10-15 22:54:43,192 epoch 9 - iter 780/2606 - loss 0.01644586 - time (sec): 55.72 - samples/sec: 1925.64 - lr: 0.000009 - momentum: 0.000000
2023-10-15 22:55:01,624 epoch 9 - iter 1040/2606 - loss 0.01628689 - time (sec): 74.16 - samples/sec: 1925.25 - lr: 0.000009 - momentum: 0.000000
2023-10-15 22:55:21,440 epoch 9 - iter 1300/2606 - loss 0.01534667 - time (sec): 93.97 - samples/sec: 1929.47 - lr: 0.000008 - momentum: 0.000000
2023-10-15 22:55:40,355 epoch 9 - iter 1560/2606 - loss 0.01439162 - time (sec): 112.89 - samples/sec: 1935.81 - lr: 0.000008 - momentum: 0.000000
2023-10-15 22:55:59,468 epoch 9 - iter 1820/2606 - loss 0.01446621 - time (sec): 132.00 - samples/sec: 1933.75 - lr: 0.000007 - momentum: 0.000000
2023-10-15 22:56:18,466 epoch 9 - iter 2080/2606 - loss 0.01388150 - time (sec): 151.00 - samples/sec: 1938.34 - lr: 0.000007 - momentum: 0.000000
2023-10-15 22:56:38,019 epoch 9 - iter 2340/2606 - loss 0.01387827 - time (sec): 170.55 - samples/sec: 1938.36 - lr: 0.000006 - momentum: 0.000000
2023-10-15 22:56:56,829 epoch 9 - iter 2600/2606 - loss 0.01405118 - time (sec): 189.36 - samples/sec: 1935.71 - lr: 0.000006 - momentum: 0.000000
2023-10-15 22:56:57,290 ----------------------------------------------------------------------------------------------------
2023-10-15 22:56:57,290 EPOCH 9 done: loss 0.0140 - lr: 0.000006
2023-10-15 22:57:06,397 DEV : loss 0.46497806906700134 - f1-score (micro avg) 0.3541
2023-10-15 22:57:06,426 ----------------------------------------------------------------------------------------------------
2023-10-15 22:57:24,867 epoch 10 - iter 260/2606 - loss 0.01004813 - time (sec): 18.44 - samples/sec: 1931.32 - lr: 0.000005 - momentum: 0.000000
2023-10-15 22:57:43,792 epoch 10 - iter 520/2606 - loss 0.01164960 - time (sec): 37.36 - samples/sec: 1912.24 - lr: 0.000004 - momentum: 0.000000
2023-10-15 22:58:02,149 epoch 10 - iter 780/2606 - loss 0.01022217 - time (sec): 55.72 - samples/sec: 1914.91 - lr: 0.000004 - momentum: 0.000000
2023-10-15 22:58:20,621 epoch 10 - iter 1040/2606 - loss 0.01089894 - time (sec): 74.19 - samples/sec: 1922.29 - lr: 0.000003 - momentum: 0.000000
2023-10-15 22:58:39,283 epoch 10 - iter 1300/2606 - loss 0.01043785 - time (sec): 92.86 - samples/sec: 1917.68 - lr: 0.000003 - momentum: 0.000000
2023-10-15 22:58:58,972 epoch 10 - iter 1560/2606 - loss 0.01058729 - time (sec): 112.55 - samples/sec: 1924.74 - lr: 0.000002 - momentum: 0.000000
2023-10-15 22:59:18,245 epoch 10 - iter 1820/2606 - loss 0.01038180 - time (sec): 131.82 - samples/sec: 1931.33 - lr: 0.000002 - momentum: 0.000000
2023-10-15 22:59:38,495 epoch 10 - iter 2080/2606 - loss 0.00998178 - time (sec): 152.07 - samples/sec: 1930.07 - lr: 0.000001 - momentum: 0.000000
2023-10-15 22:59:57,836 epoch 10 - iter 2340/2606 - loss 0.01023802 - time (sec): 171.41 - samples/sec: 1931.92 - lr: 0.000001 - momentum: 0.000000
2023-10-15 23:00:16,128 epoch 10 - iter 2600/2606 - loss 0.01035035 - time (sec): 189.70 - samples/sec: 1933.34 - lr: 0.000000 - momentum: 0.000000
2023-10-15 23:00:16,510 ----------------------------------------------------------------------------------------------------
2023-10-15 23:00:16,510 EPOCH 10 done: loss 0.0103 - lr: 0.000000
2023-10-15 23:00:25,551 DEV : loss 0.4552249312400818 - f1-score (micro avg) 0.3675
2023-10-15 23:00:25,955 ----------------------------------------------------------------------------------------------------
2023-10-15 23:00:25,956 Loading model from best epoch ...
2023-10-15 23:00:27,429 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-15 23:00:42,906
Results:
- F-score (micro) 0.4196
- F-score (macro) 0.2764
- Accuracy 0.2689
By class:
precision recall f1-score support
LOC 0.4960 0.4547 0.4744 1214
PER 0.3994 0.4567 0.4261 808
ORG 0.2405 0.1785 0.2049 353
HumanProd 0.0000 0.0000 0.0000 15
micro avg 0.4278 0.4117 0.4196 2390
macro avg 0.2839 0.2725 0.2764 2390
weighted avg 0.4224 0.4117 0.4153 2390
2023-10-15 23:00:42,906 ----------------------------------------------------------------------------------------------------
|