File size: 24,061 Bytes
f3fc8e2 |
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 |
2023-10-13 09:04:20,676 ----------------------------------------------------------------------------------------------------
2023-10-13 09:04:20,677 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 09:04:20,677 ----------------------------------------------------------------------------------------------------
2023-10-13 09:04:20,677 MultiCorpus: 1214 train + 266 dev + 251 test sentences
- NER_HIPE_2022 Corpus: 1214 train + 266 dev + 251 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/en/with_doc_seperator
2023-10-13 09:04:20,677 ----------------------------------------------------------------------------------------------------
2023-10-13 09:04:20,677 Train: 1214 sentences
2023-10-13 09:04:20,677 (train_with_dev=False, train_with_test=False)
2023-10-13 09:04:20,677 ----------------------------------------------------------------------------------------------------
2023-10-13 09:04:20,677 Training Params:
2023-10-13 09:04:20,678 - learning_rate: "5e-05"
2023-10-13 09:04:20,678 - mini_batch_size: "4"
2023-10-13 09:04:20,678 - max_epochs: "10"
2023-10-13 09:04:20,678 - shuffle: "True"
2023-10-13 09:04:20,678 ----------------------------------------------------------------------------------------------------
2023-10-13 09:04:20,678 Plugins:
2023-10-13 09:04:20,678 - LinearScheduler | warmup_fraction: '0.1'
2023-10-13 09:04:20,678 ----------------------------------------------------------------------------------------------------
2023-10-13 09:04:20,678 Final evaluation on model from best epoch (best-model.pt)
2023-10-13 09:04:20,678 - metric: "('micro avg', 'f1-score')"
2023-10-13 09:04:20,678 ----------------------------------------------------------------------------------------------------
2023-10-13 09:04:20,678 Computation:
2023-10-13 09:04:20,678 - compute on device: cuda:0
2023-10-13 09:04:20,678 - embedding storage: none
2023-10-13 09:04:20,678 ----------------------------------------------------------------------------------------------------
2023-10-13 09:04:20,678 Model training base path: "hmbench-ajmc/en-dbmdz/bert-base-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
2023-10-13 09:04:20,678 ----------------------------------------------------------------------------------------------------
2023-10-13 09:04:20,678 ----------------------------------------------------------------------------------------------------
2023-10-13 09:04:22,176 epoch 1 - iter 30/304 - loss 3.31425647 - time (sec): 1.50 - samples/sec: 2161.50 - lr: 0.000005 - momentum: 0.000000
2023-10-13 09:04:23,708 epoch 1 - iter 60/304 - loss 2.53179623 - time (sec): 3.03 - samples/sec: 2106.27 - lr: 0.000010 - momentum: 0.000000
2023-10-13 09:04:25,220 epoch 1 - iter 90/304 - loss 1.89558294 - time (sec): 4.54 - samples/sec: 2141.58 - lr: 0.000015 - momentum: 0.000000
2023-10-13 09:04:26,744 epoch 1 - iter 120/304 - loss 1.58723990 - time (sec): 6.06 - samples/sec: 2119.98 - lr: 0.000020 - momentum: 0.000000
2023-10-13 09:04:28,296 epoch 1 - iter 150/304 - loss 1.36936470 - time (sec): 7.62 - samples/sec: 2087.56 - lr: 0.000025 - momentum: 0.000000
2023-10-13 09:04:29,830 epoch 1 - iter 180/304 - loss 1.21901354 - time (sec): 9.15 - samples/sec: 2038.01 - lr: 0.000029 - momentum: 0.000000
2023-10-13 09:04:31,359 epoch 1 - iter 210/304 - loss 1.07903038 - time (sec): 10.68 - samples/sec: 2051.23 - lr: 0.000034 - momentum: 0.000000
2023-10-13 09:04:32,814 epoch 1 - iter 240/304 - loss 0.98215090 - time (sec): 12.13 - samples/sec: 2041.84 - lr: 0.000039 - momentum: 0.000000
2023-10-13 09:04:34,195 epoch 1 - iter 270/304 - loss 0.90752162 - time (sec): 13.52 - samples/sec: 2040.06 - lr: 0.000044 - momentum: 0.000000
2023-10-13 09:04:35,739 epoch 1 - iter 300/304 - loss 0.84165624 - time (sec): 15.06 - samples/sec: 2033.56 - lr: 0.000049 - momentum: 0.000000
2023-10-13 09:04:35,946 ----------------------------------------------------------------------------------------------------
2023-10-13 09:04:35,946 EPOCH 1 done: loss 0.8341 - lr: 0.000049
2023-10-13 09:04:36,776 DEV : loss 0.202301487326622 - f1-score (micro avg) 0.5797
2023-10-13 09:04:36,782 saving best model
2023-10-13 09:04:37,219 ----------------------------------------------------------------------------------------------------
2023-10-13 09:04:38,808 epoch 2 - iter 30/304 - loss 0.19650663 - time (sec): 1.59 - samples/sec: 1945.46 - lr: 0.000049 - momentum: 0.000000
2023-10-13 09:04:40,388 epoch 2 - iter 60/304 - loss 0.20528328 - time (sec): 3.17 - samples/sec: 1890.30 - lr: 0.000049 - momentum: 0.000000
2023-10-13 09:04:41,946 epoch 2 - iter 90/304 - loss 0.17434151 - time (sec): 4.73 - samples/sec: 1951.81 - lr: 0.000048 - momentum: 0.000000
2023-10-13 09:04:43,541 epoch 2 - iter 120/304 - loss 0.16749938 - time (sec): 6.32 - samples/sec: 1910.32 - lr: 0.000048 - momentum: 0.000000
2023-10-13 09:04:45,159 epoch 2 - iter 150/304 - loss 0.15293309 - time (sec): 7.94 - samples/sec: 1938.35 - lr: 0.000047 - momentum: 0.000000
2023-10-13 09:04:46,685 epoch 2 - iter 180/304 - loss 0.15217789 - time (sec): 9.46 - samples/sec: 1955.49 - lr: 0.000047 - momentum: 0.000000
2023-10-13 09:04:48,205 epoch 2 - iter 210/304 - loss 0.15455635 - time (sec): 10.98 - samples/sec: 1985.53 - lr: 0.000046 - momentum: 0.000000
2023-10-13 09:04:49,776 epoch 2 - iter 240/304 - loss 0.15812685 - time (sec): 12.56 - samples/sec: 1968.72 - lr: 0.000046 - momentum: 0.000000
2023-10-13 09:04:51,276 epoch 2 - iter 270/304 - loss 0.15344707 - time (sec): 14.06 - samples/sec: 1958.21 - lr: 0.000045 - momentum: 0.000000
2023-10-13 09:04:52,734 epoch 2 - iter 300/304 - loss 0.14666209 - time (sec): 15.51 - samples/sec: 1981.15 - lr: 0.000045 - momentum: 0.000000
2023-10-13 09:04:52,904 ----------------------------------------------------------------------------------------------------
2023-10-13 09:04:52,904 EPOCH 2 done: loss 0.1462 - lr: 0.000045
2023-10-13 09:04:54,001 DEV : loss 0.161252960562706 - f1-score (micro avg) 0.7976
2023-10-13 09:04:54,007 saving best model
2023-10-13 09:04:54,486 ----------------------------------------------------------------------------------------------------
2023-10-13 09:04:55,859 epoch 3 - iter 30/304 - loss 0.06096573 - time (sec): 1.37 - samples/sec: 2198.32 - lr: 0.000044 - momentum: 0.000000
2023-10-13 09:04:57,252 epoch 3 - iter 60/304 - loss 0.06899069 - time (sec): 2.76 - samples/sec: 2198.66 - lr: 0.000043 - momentum: 0.000000
2023-10-13 09:04:58,630 epoch 3 - iter 90/304 - loss 0.08640571 - time (sec): 4.14 - samples/sec: 2179.58 - lr: 0.000043 - momentum: 0.000000
2023-10-13 09:04:59,930 epoch 3 - iter 120/304 - loss 0.08755042 - time (sec): 5.44 - samples/sec: 2222.55 - lr: 0.000042 - momentum: 0.000000
2023-10-13 09:05:01,227 epoch 3 - iter 150/304 - loss 0.10222509 - time (sec): 6.74 - samples/sec: 2224.09 - lr: 0.000042 - momentum: 0.000000
2023-10-13 09:05:02,524 epoch 3 - iter 180/304 - loss 0.09486876 - time (sec): 8.03 - samples/sec: 2247.32 - lr: 0.000041 - momentum: 0.000000
2023-10-13 09:05:03,856 epoch 3 - iter 210/304 - loss 0.09811310 - time (sec): 9.37 - samples/sec: 2299.44 - lr: 0.000041 - momentum: 0.000000
2023-10-13 09:05:05,167 epoch 3 - iter 240/304 - loss 0.09881164 - time (sec): 10.68 - samples/sec: 2308.84 - lr: 0.000040 - momentum: 0.000000
2023-10-13 09:05:06,463 epoch 3 - iter 270/304 - loss 0.09401562 - time (sec): 11.97 - samples/sec: 2297.62 - lr: 0.000040 - momentum: 0.000000
2023-10-13 09:05:07,787 epoch 3 - iter 300/304 - loss 0.09134469 - time (sec): 13.30 - samples/sec: 2307.52 - lr: 0.000039 - momentum: 0.000000
2023-10-13 09:05:07,957 ----------------------------------------------------------------------------------------------------
2023-10-13 09:05:07,957 EPOCH 3 done: loss 0.0929 - lr: 0.000039
2023-10-13 09:05:08,860 DEV : loss 0.2110685259103775 - f1-score (micro avg) 0.7938
2023-10-13 09:05:08,866 ----------------------------------------------------------------------------------------------------
2023-10-13 09:05:10,177 epoch 4 - iter 30/304 - loss 0.02405006 - time (sec): 1.31 - samples/sec: 2298.53 - lr: 0.000038 - momentum: 0.000000
2023-10-13 09:05:11,527 epoch 4 - iter 60/304 - loss 0.07462171 - time (sec): 2.66 - samples/sec: 2303.08 - lr: 0.000038 - momentum: 0.000000
2023-10-13 09:05:12,844 epoch 4 - iter 90/304 - loss 0.06704429 - time (sec): 3.98 - samples/sec: 2314.71 - lr: 0.000037 - momentum: 0.000000
2023-10-13 09:05:14,143 epoch 4 - iter 120/304 - loss 0.06888938 - time (sec): 5.28 - samples/sec: 2308.40 - lr: 0.000037 - momentum: 0.000000
2023-10-13 09:05:15,460 epoch 4 - iter 150/304 - loss 0.07104858 - time (sec): 6.59 - samples/sec: 2288.83 - lr: 0.000036 - momentum: 0.000000
2023-10-13 09:05:16,819 epoch 4 - iter 180/304 - loss 0.06984543 - time (sec): 7.95 - samples/sec: 2291.86 - lr: 0.000036 - momentum: 0.000000
2023-10-13 09:05:18,148 epoch 4 - iter 210/304 - loss 0.06550416 - time (sec): 9.28 - samples/sec: 2289.44 - lr: 0.000035 - momentum: 0.000000
2023-10-13 09:05:19,465 epoch 4 - iter 240/304 - loss 0.06294752 - time (sec): 10.60 - samples/sec: 2272.78 - lr: 0.000035 - momentum: 0.000000
2023-10-13 09:05:20,792 epoch 4 - iter 270/304 - loss 0.06091572 - time (sec): 11.93 - samples/sec: 2304.00 - lr: 0.000034 - momentum: 0.000000
2023-10-13 09:05:22,112 epoch 4 - iter 300/304 - loss 0.06786459 - time (sec): 13.25 - samples/sec: 2311.04 - lr: 0.000033 - momentum: 0.000000
2023-10-13 09:05:22,284 ----------------------------------------------------------------------------------------------------
2023-10-13 09:05:22,284 EPOCH 4 done: loss 0.0689 - lr: 0.000033
2023-10-13 09:05:23,200 DEV : loss 0.19346584379673004 - f1-score (micro avg) 0.8168
2023-10-13 09:05:23,206 saving best model
2023-10-13 09:05:23,706 ----------------------------------------------------------------------------------------------------
2023-10-13 09:05:25,132 epoch 5 - iter 30/304 - loss 0.05275074 - time (sec): 1.42 - samples/sec: 2346.92 - lr: 0.000033 - momentum: 0.000000
2023-10-13 09:05:26,496 epoch 5 - iter 60/304 - loss 0.04726932 - time (sec): 2.78 - samples/sec: 2216.65 - lr: 0.000032 - momentum: 0.000000
2023-10-13 09:05:27,823 epoch 5 - iter 90/304 - loss 0.04916465 - time (sec): 4.11 - samples/sec: 2303.28 - lr: 0.000032 - momentum: 0.000000
2023-10-13 09:05:29,201 epoch 5 - iter 120/304 - loss 0.04542877 - time (sec): 5.49 - samples/sec: 2289.47 - lr: 0.000031 - momentum: 0.000000
2023-10-13 09:05:30,568 epoch 5 - iter 150/304 - loss 0.04512704 - time (sec): 6.86 - samples/sec: 2265.53 - lr: 0.000031 - momentum: 0.000000
2023-10-13 09:05:31,949 epoch 5 - iter 180/304 - loss 0.04701365 - time (sec): 8.24 - samples/sec: 2265.97 - lr: 0.000030 - momentum: 0.000000
2023-10-13 09:05:33,285 epoch 5 - iter 210/304 - loss 0.04782925 - time (sec): 9.57 - samples/sec: 2257.83 - lr: 0.000030 - momentum: 0.000000
2023-10-13 09:05:34,626 epoch 5 - iter 240/304 - loss 0.04537712 - time (sec): 10.91 - samples/sec: 2288.50 - lr: 0.000029 - momentum: 0.000000
2023-10-13 09:05:35,973 epoch 5 - iter 270/304 - loss 0.04648129 - time (sec): 12.26 - samples/sec: 2271.88 - lr: 0.000028 - momentum: 0.000000
2023-10-13 09:05:37,299 epoch 5 - iter 300/304 - loss 0.05241658 - time (sec): 13.59 - samples/sec: 2259.52 - lr: 0.000028 - momentum: 0.000000
2023-10-13 09:05:37,468 ----------------------------------------------------------------------------------------------------
2023-10-13 09:05:37,469 EPOCH 5 done: loss 0.0519 - lr: 0.000028
2023-10-13 09:05:38,383 DEV : loss 0.18774794042110443 - f1-score (micro avg) 0.831
2023-10-13 09:05:38,389 saving best model
2023-10-13 09:05:38,883 ----------------------------------------------------------------------------------------------------
2023-10-13 09:05:40,202 epoch 6 - iter 30/304 - loss 0.02969662 - time (sec): 1.32 - samples/sec: 2536.21 - lr: 0.000027 - momentum: 0.000000
2023-10-13 09:05:41,491 epoch 6 - iter 60/304 - loss 0.02657460 - time (sec): 2.60 - samples/sec: 2336.38 - lr: 0.000027 - momentum: 0.000000
2023-10-13 09:05:42,788 epoch 6 - iter 90/304 - loss 0.02967375 - time (sec): 3.90 - samples/sec: 2354.17 - lr: 0.000026 - momentum: 0.000000
2023-10-13 09:05:44,139 epoch 6 - iter 120/304 - loss 0.03422691 - time (sec): 5.25 - samples/sec: 2414.54 - lr: 0.000026 - momentum: 0.000000
2023-10-13 09:05:45,525 epoch 6 - iter 150/304 - loss 0.03521797 - time (sec): 6.64 - samples/sec: 2344.38 - lr: 0.000025 - momentum: 0.000000
2023-10-13 09:05:46,921 epoch 6 - iter 180/304 - loss 0.03531895 - time (sec): 8.03 - samples/sec: 2308.09 - lr: 0.000025 - momentum: 0.000000
2023-10-13 09:05:48,220 epoch 6 - iter 210/304 - loss 0.03429962 - time (sec): 9.33 - samples/sec: 2272.86 - lr: 0.000024 - momentum: 0.000000
2023-10-13 09:05:49,545 epoch 6 - iter 240/304 - loss 0.03350496 - time (sec): 10.66 - samples/sec: 2286.37 - lr: 0.000023 - momentum: 0.000000
2023-10-13 09:05:50,900 epoch 6 - iter 270/304 - loss 0.03633404 - time (sec): 12.01 - samples/sec: 2295.78 - lr: 0.000023 - momentum: 0.000000
2023-10-13 09:05:52,254 epoch 6 - iter 300/304 - loss 0.03609254 - time (sec): 13.37 - samples/sec: 2288.90 - lr: 0.000022 - momentum: 0.000000
2023-10-13 09:05:52,424 ----------------------------------------------------------------------------------------------------
2023-10-13 09:05:52,424 EPOCH 6 done: loss 0.0360 - lr: 0.000022
2023-10-13 09:05:53,341 DEV : loss 0.20084916055202484 - f1-score (micro avg) 0.8353
2023-10-13 09:05:53,348 saving best model
2023-10-13 09:05:53,828 ----------------------------------------------------------------------------------------------------
2023-10-13 09:05:55,131 epoch 7 - iter 30/304 - loss 0.02132131 - time (sec): 1.30 - samples/sec: 2162.59 - lr: 0.000022 - momentum: 0.000000
2023-10-13 09:05:56,449 epoch 7 - iter 60/304 - loss 0.01941644 - time (sec): 2.62 - samples/sec: 2243.89 - lr: 0.000021 - momentum: 0.000000
2023-10-13 09:05:57,761 epoch 7 - iter 90/304 - loss 0.02483462 - time (sec): 3.93 - samples/sec: 2262.16 - lr: 0.000021 - momentum: 0.000000
2023-10-13 09:05:59,104 epoch 7 - iter 120/304 - loss 0.02458270 - time (sec): 5.27 - samples/sec: 2347.95 - lr: 0.000020 - momentum: 0.000000
2023-10-13 09:06:00,428 epoch 7 - iter 150/304 - loss 0.02301087 - time (sec): 6.60 - samples/sec: 2301.18 - lr: 0.000020 - momentum: 0.000000
2023-10-13 09:06:01,764 epoch 7 - iter 180/304 - loss 0.02219513 - time (sec): 7.93 - samples/sec: 2316.48 - lr: 0.000019 - momentum: 0.000000
2023-10-13 09:06:03,008 epoch 7 - iter 210/304 - loss 0.02073113 - time (sec): 9.18 - samples/sec: 2302.92 - lr: 0.000018 - momentum: 0.000000
2023-10-13 09:06:04,318 epoch 7 - iter 240/304 - loss 0.02757178 - time (sec): 10.49 - samples/sec: 2300.42 - lr: 0.000018 - momentum: 0.000000
2023-10-13 09:06:05,660 epoch 7 - iter 270/304 - loss 0.02714443 - time (sec): 11.83 - samples/sec: 2307.53 - lr: 0.000017 - momentum: 0.000000
2023-10-13 09:06:06,961 epoch 7 - iter 300/304 - loss 0.02645846 - time (sec): 13.13 - samples/sec: 2328.18 - lr: 0.000017 - momentum: 0.000000
2023-10-13 09:06:07,150 ----------------------------------------------------------------------------------------------------
2023-10-13 09:06:07,150 EPOCH 7 done: loss 0.0261 - lr: 0.000017
2023-10-13 09:06:08,090 DEV : loss 0.20895600318908691 - f1-score (micro avg) 0.8424
2023-10-13 09:06:08,096 saving best model
2023-10-13 09:06:08,606 ----------------------------------------------------------------------------------------------------
2023-10-13 09:06:09,966 epoch 8 - iter 30/304 - loss 0.01056526 - time (sec): 1.36 - samples/sec: 2142.94 - lr: 0.000016 - momentum: 0.000000
2023-10-13 09:06:11,404 epoch 8 - iter 60/304 - loss 0.00730266 - time (sec): 2.80 - samples/sec: 2092.65 - lr: 0.000016 - momentum: 0.000000
2023-10-13 09:06:12,738 epoch 8 - iter 90/304 - loss 0.00608566 - time (sec): 4.13 - samples/sec: 2171.41 - lr: 0.000015 - momentum: 0.000000
2023-10-13 09:06:14,066 epoch 8 - iter 120/304 - loss 0.01279583 - time (sec): 5.46 - samples/sec: 2218.58 - lr: 0.000015 - momentum: 0.000000
2023-10-13 09:06:15,385 epoch 8 - iter 150/304 - loss 0.01346908 - time (sec): 6.78 - samples/sec: 2257.75 - lr: 0.000014 - momentum: 0.000000
2023-10-13 09:06:16,696 epoch 8 - iter 180/304 - loss 0.01285972 - time (sec): 8.09 - samples/sec: 2234.68 - lr: 0.000013 - momentum: 0.000000
2023-10-13 09:06:18,013 epoch 8 - iter 210/304 - loss 0.01281828 - time (sec): 9.40 - samples/sec: 2254.16 - lr: 0.000013 - momentum: 0.000000
2023-10-13 09:06:19,337 epoch 8 - iter 240/304 - loss 0.01223233 - time (sec): 10.73 - samples/sec: 2270.97 - lr: 0.000012 - momentum: 0.000000
2023-10-13 09:06:20,662 epoch 8 - iter 270/304 - loss 0.01625321 - time (sec): 12.05 - samples/sec: 2296.37 - lr: 0.000012 - momentum: 0.000000
2023-10-13 09:06:21,972 epoch 8 - iter 300/304 - loss 0.01736944 - time (sec): 13.36 - samples/sec: 2294.13 - lr: 0.000011 - momentum: 0.000000
2023-10-13 09:06:22,147 ----------------------------------------------------------------------------------------------------
2023-10-13 09:06:22,148 EPOCH 8 done: loss 0.0172 - lr: 0.000011
2023-10-13 09:06:23,062 DEV : loss 0.22081097960472107 - f1-score (micro avg) 0.8281
2023-10-13 09:06:23,068 ----------------------------------------------------------------------------------------------------
2023-10-13 09:06:24,379 epoch 9 - iter 30/304 - loss 0.01196464 - time (sec): 1.31 - samples/sec: 2456.12 - lr: 0.000011 - momentum: 0.000000
2023-10-13 09:06:25,683 epoch 9 - iter 60/304 - loss 0.00801280 - time (sec): 2.61 - samples/sec: 2383.27 - lr: 0.000010 - momentum: 0.000000
2023-10-13 09:06:27,006 epoch 9 - iter 90/304 - loss 0.00586371 - time (sec): 3.94 - samples/sec: 2321.29 - lr: 0.000010 - momentum: 0.000000
2023-10-13 09:06:28,374 epoch 9 - iter 120/304 - loss 0.01255957 - time (sec): 5.31 - samples/sec: 2339.86 - lr: 0.000009 - momentum: 0.000000
2023-10-13 09:06:29,718 epoch 9 - iter 150/304 - loss 0.01240447 - time (sec): 6.65 - samples/sec: 2347.40 - lr: 0.000008 - momentum: 0.000000
2023-10-13 09:06:31,015 epoch 9 - iter 180/304 - loss 0.01195576 - time (sec): 7.95 - samples/sec: 2337.16 - lr: 0.000008 - momentum: 0.000000
2023-10-13 09:06:32,320 epoch 9 - iter 210/304 - loss 0.01295077 - time (sec): 9.25 - samples/sec: 2323.20 - lr: 0.000007 - momentum: 0.000000
2023-10-13 09:06:33,681 epoch 9 - iter 240/304 - loss 0.01168237 - time (sec): 10.61 - samples/sec: 2336.96 - lr: 0.000007 - momentum: 0.000000
2023-10-13 09:06:35,017 epoch 9 - iter 270/304 - loss 0.01202743 - time (sec): 11.95 - samples/sec: 2320.88 - lr: 0.000006 - momentum: 0.000000
2023-10-13 09:06:36,350 epoch 9 - iter 300/304 - loss 0.01178066 - time (sec): 13.28 - samples/sec: 2314.81 - lr: 0.000006 - momentum: 0.000000
2023-10-13 09:06:36,522 ----------------------------------------------------------------------------------------------------
2023-10-13 09:06:36,523 EPOCH 9 done: loss 0.0117 - lr: 0.000006
2023-10-13 09:06:37,425 DEV : loss 0.2193765640258789 - f1-score (micro avg) 0.8458
2023-10-13 09:06:37,431 saving best model
2023-10-13 09:06:37,926 ----------------------------------------------------------------------------------------------------
2023-10-13 09:06:39,259 epoch 10 - iter 30/304 - loss 0.00979948 - time (sec): 1.33 - samples/sec: 2350.60 - lr: 0.000005 - momentum: 0.000000
2023-10-13 09:06:40,577 epoch 10 - iter 60/304 - loss 0.00536257 - time (sec): 2.65 - samples/sec: 2333.35 - lr: 0.000005 - momentum: 0.000000
2023-10-13 09:06:41,871 epoch 10 - iter 90/304 - loss 0.00609703 - time (sec): 3.94 - samples/sec: 2317.60 - lr: 0.000004 - momentum: 0.000000
2023-10-13 09:06:43,182 epoch 10 - iter 120/304 - loss 0.00504606 - time (sec): 5.25 - samples/sec: 2292.91 - lr: 0.000003 - momentum: 0.000000
2023-10-13 09:06:44,555 epoch 10 - iter 150/304 - loss 0.00679310 - time (sec): 6.63 - samples/sec: 2261.60 - lr: 0.000003 - momentum: 0.000000
2023-10-13 09:06:45,866 epoch 10 - iter 180/304 - loss 0.00710765 - time (sec): 7.94 - samples/sec: 2290.19 - lr: 0.000002 - momentum: 0.000000
2023-10-13 09:06:47,144 epoch 10 - iter 210/304 - loss 0.00724926 - time (sec): 9.22 - samples/sec: 2318.40 - lr: 0.000002 - momentum: 0.000000
2023-10-13 09:06:48,455 epoch 10 - iter 240/304 - loss 0.00770948 - time (sec): 10.53 - samples/sec: 2323.53 - lr: 0.000001 - momentum: 0.000000
2023-10-13 09:06:49,867 epoch 10 - iter 270/304 - loss 0.00810010 - time (sec): 11.94 - samples/sec: 2302.68 - lr: 0.000001 - momentum: 0.000000
2023-10-13 09:06:51,172 epoch 10 - iter 300/304 - loss 0.00800972 - time (sec): 13.24 - samples/sec: 2308.48 - lr: 0.000000 - momentum: 0.000000
2023-10-13 09:06:51,342 ----------------------------------------------------------------------------------------------------
2023-10-13 09:06:51,343 EPOCH 10 done: loss 0.0080 - lr: 0.000000
2023-10-13 09:06:52,272 DEV : loss 0.21843381226062775 - f1-score (micro avg) 0.8498
2023-10-13 09:06:52,278 saving best model
2023-10-13 09:06:53,181 ----------------------------------------------------------------------------------------------------
2023-10-13 09:06:53,183 Loading model from best epoch ...
2023-10-13 09:06:55,102 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-date, B-date, E-date, I-date, S-object, B-object, E-object, I-object
2023-10-13 09:06:55,791
Results:
- F-score (micro) 0.7826
- F-score (macro) 0.5753
- Accuracy 0.6513
By class:
precision recall f1-score support
scope 0.7455 0.8146 0.7785 151
pers 0.7063 0.9271 0.8018 96
work 0.7241 0.8842 0.7962 95
date 0.0000 0.0000 0.0000 3
loc 1.0000 0.3333 0.5000 3
micro avg 0.7226 0.8534 0.7826 348
macro avg 0.6352 0.5918 0.5753 348
weighted avg 0.7246 0.8534 0.7806 348
2023-10-13 09:06:55,791 ----------------------------------------------------------------------------------------------------
|