File size: 25,148 Bytes
b4e8f5e |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 |
2023-10-06 20:41:12,480 ----------------------------------------------------------------------------------------------------
2023-10-06 20:41:12,481 Model: "SequenceTagger(
(embeddings): ByT5Embeddings(
(model): T5EncoderModel(
(shared): Embedding(384, 1472)
(encoder): T5Stack(
(embed_tokens): Embedding(384, 1472)
(block): ModuleList(
(0): T5Block(
(layer): ModuleList(
(0): T5LayerSelfAttention(
(SelfAttention): T5Attention(
(q): Linear(in_features=1472, out_features=384, bias=False)
(k): Linear(in_features=1472, out_features=384, bias=False)
(v): Linear(in_features=1472, out_features=384, bias=False)
(o): Linear(in_features=384, out_features=1472, bias=False)
(relative_attention_bias): Embedding(32, 6)
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
(1): T5LayerFF(
(DenseReluDense): T5DenseGatedActDense(
(wi_0): Linear(in_features=1472, out_features=3584, bias=False)
(wi_1): Linear(in_features=1472, out_features=3584, bias=False)
(wo): Linear(in_features=3584, out_features=1472, bias=False)
(dropout): Dropout(p=0.1, inplace=False)
(act): NewGELUActivation()
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(1-11): 11 x T5Block(
(layer): ModuleList(
(0): T5LayerSelfAttention(
(SelfAttention): T5Attention(
(q): Linear(in_features=1472, out_features=384, bias=False)
(k): Linear(in_features=1472, out_features=384, bias=False)
(v): Linear(in_features=1472, out_features=384, bias=False)
(o): Linear(in_features=384, out_features=1472, bias=False)
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
(1): T5LayerFF(
(DenseReluDense): T5DenseGatedActDense(
(wi_0): Linear(in_features=1472, out_features=3584, bias=False)
(wi_1): Linear(in_features=1472, out_features=3584, bias=False)
(wo): Linear(in_features=3584, out_features=1472, bias=False)
(dropout): Dropout(p=0.1, inplace=False)
(act): NewGELUActivation()
)
(layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(final_layer_norm): T5LayerNorm()
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=1472, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-06 20:41:12,482 ----------------------------------------------------------------------------------------------------
2023-10-06 20:41:12,482 MultiCorpus: 1100 train + 206 dev + 240 test sentences
- NER_HIPE_2022 Corpus: 1100 train + 206 dev + 240 test sentences - /app/.flair/datasets/ner_hipe_2022/v2.1/ajmc/de/with_doc_seperator
2023-10-06 20:41:12,482 ----------------------------------------------------------------------------------------------------
2023-10-06 20:41:12,482 Train: 1100 sentences
2023-10-06 20:41:12,482 (train_with_dev=False, train_with_test=False)
2023-10-06 20:41:12,482 ----------------------------------------------------------------------------------------------------
2023-10-06 20:41:12,482 Training Params:
2023-10-06 20:41:12,482 - learning_rate: "0.00015"
2023-10-06 20:41:12,482 - mini_batch_size: "4"
2023-10-06 20:41:12,482 - max_epochs: "10"
2023-10-06 20:41:12,482 - shuffle: "True"
2023-10-06 20:41:12,482 ----------------------------------------------------------------------------------------------------
2023-10-06 20:41:12,482 Plugins:
2023-10-06 20:41:12,482 - TensorboardLogger
2023-10-06 20:41:12,483 - LinearScheduler | warmup_fraction: '0.1'
2023-10-06 20:41:12,483 ----------------------------------------------------------------------------------------------------
2023-10-06 20:41:12,483 Final evaluation on model from best epoch (best-model.pt)
2023-10-06 20:41:12,483 - metric: "('micro avg', 'f1-score')"
2023-10-06 20:41:12,483 ----------------------------------------------------------------------------------------------------
2023-10-06 20:41:12,483 Computation:
2023-10-06 20:41:12,483 - compute on device: cuda:0
2023-10-06 20:41:12,483 - embedding storage: none
2023-10-06 20:41:12,483 ----------------------------------------------------------------------------------------------------
2023-10-06 20:41:12,483 Model training base path: "hmbench-ajmc/de-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-1"
2023-10-06 20:41:12,483 ----------------------------------------------------------------------------------------------------
2023-10-06 20:41:12,483 ----------------------------------------------------------------------------------------------------
2023-10-06 20:41:12,483 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-06 20:41:23,097 epoch 1 - iter 27/275 - loss 3.22851574 - time (sec): 10.61 - samples/sec: 208.24 - lr: 0.000014 - momentum: 0.000000
2023-10-06 20:41:33,019 epoch 1 - iter 54/275 - loss 3.21829774 - time (sec): 20.53 - samples/sec: 213.98 - lr: 0.000029 - momentum: 0.000000
2023-10-06 20:41:42,867 epoch 1 - iter 81/275 - loss 3.19979413 - time (sec): 30.38 - samples/sec: 215.91 - lr: 0.000044 - momentum: 0.000000
2023-10-06 20:41:52,594 epoch 1 - iter 108/275 - loss 3.15653748 - time (sec): 40.11 - samples/sec: 215.83 - lr: 0.000058 - momentum: 0.000000
2023-10-06 20:42:03,379 epoch 1 - iter 135/275 - loss 3.06509180 - time (sec): 50.89 - samples/sec: 218.86 - lr: 0.000073 - momentum: 0.000000
2023-10-06 20:42:13,893 epoch 1 - iter 162/275 - loss 2.96221944 - time (sec): 61.41 - samples/sec: 219.38 - lr: 0.000088 - momentum: 0.000000
2023-10-06 20:42:24,478 epoch 1 - iter 189/275 - loss 2.85684614 - time (sec): 71.99 - samples/sec: 219.80 - lr: 0.000103 - momentum: 0.000000
2023-10-06 20:42:34,575 epoch 1 - iter 216/275 - loss 2.74750307 - time (sec): 82.09 - samples/sec: 219.59 - lr: 0.000117 - momentum: 0.000000
2023-10-06 20:42:44,911 epoch 1 - iter 243/275 - loss 2.63108071 - time (sec): 92.43 - samples/sec: 219.19 - lr: 0.000132 - momentum: 0.000000
2023-10-06 20:42:54,630 epoch 1 - iter 270/275 - loss 2.53023247 - time (sec): 102.15 - samples/sec: 218.06 - lr: 0.000147 - momentum: 0.000000
2023-10-06 20:42:56,909 ----------------------------------------------------------------------------------------------------
2023-10-06 20:42:56,910 EPOCH 1 done: loss 2.5051 - lr: 0.000147
2023-10-06 20:43:03,213 DEV : loss 1.1949074268341064 - f1-score (micro avg) 0.0
2023-10-06 20:43:03,219 ----------------------------------------------------------------------------------------------------
2023-10-06 20:43:13,749 epoch 2 - iter 27/275 - loss 1.09853432 - time (sec): 10.53 - samples/sec: 209.06 - lr: 0.000148 - momentum: 0.000000
2023-10-06 20:43:24,679 epoch 2 - iter 54/275 - loss 0.97325545 - time (sec): 21.46 - samples/sec: 208.26 - lr: 0.000147 - momentum: 0.000000
2023-10-06 20:43:34,960 epoch 2 - iter 81/275 - loss 0.96420986 - time (sec): 31.74 - samples/sec: 207.53 - lr: 0.000145 - momentum: 0.000000
2023-10-06 20:43:44,907 epoch 2 - iter 108/275 - loss 0.90817676 - time (sec): 41.69 - samples/sec: 204.38 - lr: 0.000144 - momentum: 0.000000
2023-10-06 20:43:55,205 epoch 2 - iter 135/275 - loss 0.89704867 - time (sec): 51.98 - samples/sec: 203.33 - lr: 0.000142 - momentum: 0.000000
2023-10-06 20:44:06,528 epoch 2 - iter 162/275 - loss 0.86146920 - time (sec): 63.31 - samples/sec: 204.97 - lr: 0.000140 - momentum: 0.000000
2023-10-06 20:44:17,829 epoch 2 - iter 189/275 - loss 0.82102558 - time (sec): 74.61 - samples/sec: 206.69 - lr: 0.000139 - momentum: 0.000000
2023-10-06 20:44:28,642 epoch 2 - iter 216/275 - loss 0.78226959 - time (sec): 85.42 - samples/sec: 207.29 - lr: 0.000137 - momentum: 0.000000
2023-10-06 20:44:39,384 epoch 2 - iter 243/275 - loss 0.74764483 - time (sec): 96.16 - samples/sec: 208.07 - lr: 0.000135 - momentum: 0.000000
2023-10-06 20:44:49,949 epoch 2 - iter 270/275 - loss 0.72320304 - time (sec): 106.73 - samples/sec: 208.87 - lr: 0.000134 - momentum: 0.000000
2023-10-06 20:44:51,899 ----------------------------------------------------------------------------------------------------
2023-10-06 20:44:51,899 EPOCH 2 done: loss 0.7199 - lr: 0.000134
2023-10-06 20:44:58,340 DEV : loss 0.43983766436576843 - f1-score (micro avg) 0.2969
2023-10-06 20:44:58,345 saving best model
2023-10-06 20:44:59,199 ----------------------------------------------------------------------------------------------------
2023-10-06 20:45:09,642 epoch 3 - iter 27/275 - loss 0.40940852 - time (sec): 10.44 - samples/sec: 212.23 - lr: 0.000132 - momentum: 0.000000
2023-10-06 20:45:20,822 epoch 3 - iter 54/275 - loss 0.38706250 - time (sec): 21.62 - samples/sec: 215.90 - lr: 0.000130 - momentum: 0.000000
2023-10-06 20:45:31,632 epoch 3 - iter 81/275 - loss 0.38027867 - time (sec): 32.43 - samples/sec: 216.40 - lr: 0.000129 - momentum: 0.000000
2023-10-06 20:45:42,720 epoch 3 - iter 108/275 - loss 0.37562173 - time (sec): 43.52 - samples/sec: 215.70 - lr: 0.000127 - momentum: 0.000000
2023-10-06 20:45:53,200 epoch 3 - iter 135/275 - loss 0.36728998 - time (sec): 54.00 - samples/sec: 214.06 - lr: 0.000125 - momentum: 0.000000
2023-10-06 20:46:04,063 epoch 3 - iter 162/275 - loss 0.36364722 - time (sec): 64.86 - samples/sec: 213.19 - lr: 0.000124 - momentum: 0.000000
2023-10-06 20:46:14,084 epoch 3 - iter 189/275 - loss 0.35221302 - time (sec): 74.88 - samples/sec: 211.14 - lr: 0.000122 - momentum: 0.000000
2023-10-06 20:46:24,319 epoch 3 - iter 216/275 - loss 0.34021923 - time (sec): 85.12 - samples/sec: 209.43 - lr: 0.000120 - momentum: 0.000000
2023-10-06 20:46:35,911 epoch 3 - iter 243/275 - loss 0.33081667 - time (sec): 96.71 - samples/sec: 210.07 - lr: 0.000119 - momentum: 0.000000
2023-10-06 20:46:45,972 epoch 3 - iter 270/275 - loss 0.31882959 - time (sec): 106.77 - samples/sec: 208.92 - lr: 0.000117 - momentum: 0.000000
2023-10-06 20:46:48,190 ----------------------------------------------------------------------------------------------------
2023-10-06 20:46:48,190 EPOCH 3 done: loss 0.3166 - lr: 0.000117
2023-10-06 20:46:54,743 DEV : loss 0.22956699132919312 - f1-score (micro avg) 0.7306
2023-10-06 20:46:54,748 saving best model
2023-10-06 20:46:59,225 ----------------------------------------------------------------------------------------------------
2023-10-06 20:47:11,123 epoch 4 - iter 27/275 - loss 0.18862218 - time (sec): 11.90 - samples/sec: 219.13 - lr: 0.000115 - momentum: 0.000000
2023-10-06 20:47:21,611 epoch 4 - iter 54/275 - loss 0.17081658 - time (sec): 22.38 - samples/sec: 212.87 - lr: 0.000114 - momentum: 0.000000
2023-10-06 20:47:31,893 epoch 4 - iter 81/275 - loss 0.17792602 - time (sec): 32.67 - samples/sec: 207.43 - lr: 0.000112 - momentum: 0.000000
2023-10-06 20:47:42,500 epoch 4 - iter 108/275 - loss 0.18147922 - time (sec): 43.27 - samples/sec: 208.49 - lr: 0.000110 - momentum: 0.000000
2023-10-06 20:47:53,725 epoch 4 - iter 135/275 - loss 0.18415122 - time (sec): 54.50 - samples/sec: 210.10 - lr: 0.000109 - momentum: 0.000000
2023-10-06 20:48:04,226 epoch 4 - iter 162/275 - loss 0.17880585 - time (sec): 65.00 - samples/sec: 208.51 - lr: 0.000107 - momentum: 0.000000
2023-10-06 20:48:15,156 epoch 4 - iter 189/275 - loss 0.17398312 - time (sec): 75.93 - samples/sec: 208.20 - lr: 0.000105 - momentum: 0.000000
2023-10-06 20:48:25,454 epoch 4 - iter 216/275 - loss 0.17691687 - time (sec): 86.23 - samples/sec: 208.11 - lr: 0.000104 - momentum: 0.000000
2023-10-06 20:48:35,926 epoch 4 - iter 243/275 - loss 0.17122472 - time (sec): 96.70 - samples/sec: 208.21 - lr: 0.000102 - momentum: 0.000000
2023-10-06 20:48:46,776 epoch 4 - iter 270/275 - loss 0.16701451 - time (sec): 107.55 - samples/sec: 207.35 - lr: 0.000101 - momentum: 0.000000
2023-10-06 20:48:48,883 ----------------------------------------------------------------------------------------------------
2023-10-06 20:48:48,883 EPOCH 4 done: loss 0.1661 - lr: 0.000101
2023-10-06 20:48:55,611 DEV : loss 0.14865034818649292 - f1-score (micro avg) 0.8125
2023-10-06 20:48:55,616 saving best model
2023-10-06 20:48:59,970 ----------------------------------------------------------------------------------------------------
2023-10-06 20:49:10,651 epoch 5 - iter 27/275 - loss 0.13835944 - time (sec): 10.68 - samples/sec: 204.71 - lr: 0.000099 - momentum: 0.000000
2023-10-06 20:49:21,407 epoch 5 - iter 54/275 - loss 0.12077603 - time (sec): 21.44 - samples/sec: 203.68 - lr: 0.000097 - momentum: 0.000000
2023-10-06 20:49:31,869 epoch 5 - iter 81/275 - loss 0.12505616 - time (sec): 31.90 - samples/sec: 205.13 - lr: 0.000095 - momentum: 0.000000
2023-10-06 20:49:43,704 epoch 5 - iter 108/275 - loss 0.11524283 - time (sec): 43.73 - samples/sec: 208.68 - lr: 0.000094 - momentum: 0.000000
2023-10-06 20:49:55,748 epoch 5 - iter 135/275 - loss 0.11047258 - time (sec): 55.78 - samples/sec: 209.28 - lr: 0.000092 - momentum: 0.000000
2023-10-06 20:50:06,657 epoch 5 - iter 162/275 - loss 0.10857292 - time (sec): 66.68 - samples/sec: 207.63 - lr: 0.000090 - momentum: 0.000000
2023-10-06 20:50:17,559 epoch 5 - iter 189/275 - loss 0.10346163 - time (sec): 77.59 - samples/sec: 207.55 - lr: 0.000089 - momentum: 0.000000
2023-10-06 20:50:28,156 epoch 5 - iter 216/275 - loss 0.10354840 - time (sec): 88.18 - samples/sec: 208.70 - lr: 0.000087 - momentum: 0.000000
2023-10-06 20:50:38,244 epoch 5 - iter 243/275 - loss 0.10378967 - time (sec): 98.27 - samples/sec: 207.14 - lr: 0.000086 - momentum: 0.000000
2023-10-06 20:50:48,462 epoch 5 - iter 270/275 - loss 0.09977533 - time (sec): 108.49 - samples/sec: 206.57 - lr: 0.000084 - momentum: 0.000000
2023-10-06 20:50:50,302 ----------------------------------------------------------------------------------------------------
2023-10-06 20:50:50,302 EPOCH 5 done: loss 0.1025 - lr: 0.000084
2023-10-06 20:50:57,063 DEV : loss 0.12752477824687958 - f1-score (micro avg) 0.8685
2023-10-06 20:50:57,069 saving best model
2023-10-06 20:51:01,781 ----------------------------------------------------------------------------------------------------
2023-10-06 20:51:12,742 epoch 6 - iter 27/275 - loss 0.07310406 - time (sec): 10.96 - samples/sec: 214.16 - lr: 0.000082 - momentum: 0.000000
2023-10-06 20:51:23,205 epoch 6 - iter 54/275 - loss 0.07407181 - time (sec): 21.42 - samples/sec: 209.41 - lr: 0.000080 - momentum: 0.000000
2023-10-06 20:51:33,976 epoch 6 - iter 81/275 - loss 0.06876723 - time (sec): 32.19 - samples/sec: 209.02 - lr: 0.000079 - momentum: 0.000000
2023-10-06 20:51:45,214 epoch 6 - iter 108/275 - loss 0.06624089 - time (sec): 43.43 - samples/sec: 207.59 - lr: 0.000077 - momentum: 0.000000
2023-10-06 20:51:56,261 epoch 6 - iter 135/275 - loss 0.06596938 - time (sec): 54.48 - samples/sec: 206.93 - lr: 0.000075 - momentum: 0.000000
2023-10-06 20:52:07,021 epoch 6 - iter 162/275 - loss 0.06566254 - time (sec): 65.24 - samples/sec: 206.49 - lr: 0.000074 - momentum: 0.000000
2023-10-06 20:52:18,206 epoch 6 - iter 189/275 - loss 0.07319061 - time (sec): 76.42 - samples/sec: 208.38 - lr: 0.000072 - momentum: 0.000000
2023-10-06 20:52:28,891 epoch 6 - iter 216/275 - loss 0.07175614 - time (sec): 87.11 - samples/sec: 207.72 - lr: 0.000071 - momentum: 0.000000
2023-10-06 20:52:39,707 epoch 6 - iter 243/275 - loss 0.07368640 - time (sec): 97.92 - samples/sec: 207.27 - lr: 0.000069 - momentum: 0.000000
2023-10-06 20:52:49,870 epoch 6 - iter 270/275 - loss 0.07386029 - time (sec): 108.09 - samples/sec: 206.37 - lr: 0.000067 - momentum: 0.000000
2023-10-06 20:52:52,149 ----------------------------------------------------------------------------------------------------
2023-10-06 20:52:52,150 EPOCH 6 done: loss 0.0737 - lr: 0.000067
2023-10-06 20:52:58,861 DEV : loss 0.1320181041955948 - f1-score (micro avg) 0.8616
2023-10-06 20:52:58,867 ----------------------------------------------------------------------------------------------------
2023-10-06 20:53:09,121 epoch 7 - iter 27/275 - loss 0.04497303 - time (sec): 10.25 - samples/sec: 199.75 - lr: 0.000065 - momentum: 0.000000
2023-10-06 20:53:19,415 epoch 7 - iter 54/275 - loss 0.04642627 - time (sec): 20.55 - samples/sec: 198.86 - lr: 0.000064 - momentum: 0.000000
2023-10-06 20:53:30,703 epoch 7 - iter 81/275 - loss 0.05601185 - time (sec): 31.83 - samples/sec: 203.08 - lr: 0.000062 - momentum: 0.000000
2023-10-06 20:53:41,853 epoch 7 - iter 108/275 - loss 0.04747209 - time (sec): 42.98 - samples/sec: 207.33 - lr: 0.000060 - momentum: 0.000000
2023-10-06 20:53:53,189 epoch 7 - iter 135/275 - loss 0.04916826 - time (sec): 54.32 - samples/sec: 207.47 - lr: 0.000059 - momentum: 0.000000
2023-10-06 20:54:03,634 epoch 7 - iter 162/275 - loss 0.05453263 - time (sec): 64.77 - samples/sec: 205.22 - lr: 0.000057 - momentum: 0.000000
2023-10-06 20:54:14,657 epoch 7 - iter 189/275 - loss 0.05089647 - time (sec): 75.79 - samples/sec: 205.49 - lr: 0.000056 - momentum: 0.000000
2023-10-06 20:54:25,774 epoch 7 - iter 216/275 - loss 0.05096154 - time (sec): 86.91 - samples/sec: 205.92 - lr: 0.000054 - momentum: 0.000000
2023-10-06 20:54:37,052 epoch 7 - iter 243/275 - loss 0.05331658 - time (sec): 98.18 - samples/sec: 207.30 - lr: 0.000052 - momentum: 0.000000
2023-10-06 20:54:47,632 epoch 7 - iter 270/275 - loss 0.05694342 - time (sec): 108.76 - samples/sec: 206.60 - lr: 0.000051 - momentum: 0.000000
2023-10-06 20:54:49,366 ----------------------------------------------------------------------------------------------------
2023-10-06 20:54:49,366 EPOCH 7 done: loss 0.0568 - lr: 0.000051
2023-10-06 20:54:56,092 DEV : loss 0.12154703587293625 - f1-score (micro avg) 0.8695
2023-10-06 20:54:56,097 saving best model
2023-10-06 20:55:00,483 ----------------------------------------------------------------------------------------------------
2023-10-06 20:55:10,631 epoch 8 - iter 27/275 - loss 0.04565948 - time (sec): 10.15 - samples/sec: 191.40 - lr: 0.000049 - momentum: 0.000000
2023-10-06 20:55:21,037 epoch 8 - iter 54/275 - loss 0.06174029 - time (sec): 20.55 - samples/sec: 201.59 - lr: 0.000047 - momentum: 0.000000
2023-10-06 20:55:31,987 epoch 8 - iter 81/275 - loss 0.05427538 - time (sec): 31.50 - samples/sec: 205.38 - lr: 0.000045 - momentum: 0.000000
2023-10-06 20:55:42,945 epoch 8 - iter 108/275 - loss 0.05040491 - time (sec): 42.46 - samples/sec: 206.34 - lr: 0.000044 - momentum: 0.000000
2023-10-06 20:55:53,321 epoch 8 - iter 135/275 - loss 0.04921226 - time (sec): 52.84 - samples/sec: 203.27 - lr: 0.000042 - momentum: 0.000000
2023-10-06 20:56:04,857 epoch 8 - iter 162/275 - loss 0.05030560 - time (sec): 64.37 - samples/sec: 205.65 - lr: 0.000041 - momentum: 0.000000
2023-10-06 20:56:14,975 epoch 8 - iter 189/275 - loss 0.04905001 - time (sec): 74.49 - samples/sec: 205.13 - lr: 0.000039 - momentum: 0.000000
2023-10-06 20:56:26,443 epoch 8 - iter 216/275 - loss 0.04739645 - time (sec): 85.96 - samples/sec: 206.03 - lr: 0.000037 - momentum: 0.000000
2023-10-06 20:56:38,162 epoch 8 - iter 243/275 - loss 0.04880165 - time (sec): 97.68 - samples/sec: 206.19 - lr: 0.000036 - momentum: 0.000000
2023-10-06 20:56:48,961 epoch 8 - iter 270/275 - loss 0.04732420 - time (sec): 108.48 - samples/sec: 205.91 - lr: 0.000034 - momentum: 0.000000
2023-10-06 20:56:51,030 ----------------------------------------------------------------------------------------------------
2023-10-06 20:56:51,030 EPOCH 8 done: loss 0.0465 - lr: 0.000034
2023-10-06 20:56:57,714 DEV : loss 0.12373079359531403 - f1-score (micro avg) 0.8668
2023-10-06 20:56:57,719 ----------------------------------------------------------------------------------------------------
2023-10-06 20:57:08,225 epoch 9 - iter 27/275 - loss 0.02881863 - time (sec): 10.50 - samples/sec: 204.50 - lr: 0.000032 - momentum: 0.000000
2023-10-06 20:57:18,849 epoch 9 - iter 54/275 - loss 0.04766868 - time (sec): 21.13 - samples/sec: 204.81 - lr: 0.000030 - momentum: 0.000000
2023-10-06 20:57:29,593 epoch 9 - iter 81/275 - loss 0.04945213 - time (sec): 31.87 - samples/sec: 204.48 - lr: 0.000029 - momentum: 0.000000
2023-10-06 20:57:40,917 epoch 9 - iter 108/275 - loss 0.04948668 - time (sec): 43.20 - samples/sec: 204.81 - lr: 0.000027 - momentum: 0.000000
2023-10-06 20:57:51,779 epoch 9 - iter 135/275 - loss 0.05271936 - time (sec): 54.06 - samples/sec: 205.15 - lr: 0.000026 - momentum: 0.000000
2023-10-06 20:58:02,196 epoch 9 - iter 162/275 - loss 0.05315498 - time (sec): 64.47 - samples/sec: 205.43 - lr: 0.000024 - momentum: 0.000000
2023-10-06 20:58:14,113 epoch 9 - iter 189/275 - loss 0.04591171 - time (sec): 76.39 - samples/sec: 206.84 - lr: 0.000022 - momentum: 0.000000
2023-10-06 20:58:24,875 epoch 9 - iter 216/275 - loss 0.04283486 - time (sec): 87.15 - samples/sec: 206.73 - lr: 0.000021 - momentum: 0.000000
2023-10-06 20:58:35,090 epoch 9 - iter 243/275 - loss 0.03967070 - time (sec): 97.37 - samples/sec: 205.61 - lr: 0.000019 - momentum: 0.000000
2023-10-06 20:58:45,978 epoch 9 - iter 270/275 - loss 0.04036106 - time (sec): 108.26 - samples/sec: 206.32 - lr: 0.000017 - momentum: 0.000000
2023-10-06 20:58:47,978 ----------------------------------------------------------------------------------------------------
2023-10-06 20:58:47,978 EPOCH 9 done: loss 0.0402 - lr: 0.000017
2023-10-06 20:58:54,661 DEV : loss 0.126312717795372 - f1-score (micro avg) 0.8752
2023-10-06 20:58:54,667 saving best model
2023-10-06 20:58:59,057 ----------------------------------------------------------------------------------------------------
2023-10-06 20:59:10,177 epoch 10 - iter 27/275 - loss 0.06065561 - time (sec): 11.12 - samples/sec: 204.36 - lr: 0.000015 - momentum: 0.000000
2023-10-06 20:59:20,842 epoch 10 - iter 54/275 - loss 0.05483908 - time (sec): 21.78 - samples/sec: 203.50 - lr: 0.000014 - momentum: 0.000000
2023-10-06 20:59:31,444 epoch 10 - iter 81/275 - loss 0.04534089 - time (sec): 32.38 - samples/sec: 205.00 - lr: 0.000012 - momentum: 0.000000
2023-10-06 20:59:42,516 epoch 10 - iter 108/275 - loss 0.04445337 - time (sec): 43.46 - samples/sec: 206.32 - lr: 0.000011 - momentum: 0.000000
2023-10-06 20:59:53,025 epoch 10 - iter 135/275 - loss 0.03812986 - time (sec): 53.97 - samples/sec: 204.48 - lr: 0.000009 - momentum: 0.000000
2023-10-06 21:00:03,647 epoch 10 - iter 162/275 - loss 0.04007420 - time (sec): 64.59 - samples/sec: 204.17 - lr: 0.000007 - momentum: 0.000000
2023-10-06 21:00:14,653 epoch 10 - iter 189/275 - loss 0.03835348 - time (sec): 75.59 - samples/sec: 204.39 - lr: 0.000006 - momentum: 0.000000
2023-10-06 21:00:25,689 epoch 10 - iter 216/275 - loss 0.03808426 - time (sec): 86.63 - samples/sec: 205.10 - lr: 0.000004 - momentum: 0.000000
2023-10-06 21:00:36,553 epoch 10 - iter 243/275 - loss 0.03670371 - time (sec): 97.49 - samples/sec: 205.86 - lr: 0.000002 - momentum: 0.000000
2023-10-06 21:00:47,225 epoch 10 - iter 270/275 - loss 0.03786440 - time (sec): 108.17 - samples/sec: 206.28 - lr: 0.000001 - momentum: 0.000000
2023-10-06 21:00:49,405 ----------------------------------------------------------------------------------------------------
2023-10-06 21:00:49,405 EPOCH 10 done: loss 0.0376 - lr: 0.000001
2023-10-06 21:00:56,101 DEV : loss 0.12622681260108948 - f1-score (micro avg) 0.8786
2023-10-06 21:00:56,107 saving best model
2023-10-06 21:01:01,443 ----------------------------------------------------------------------------------------------------
2023-10-06 21:01:01,445 Loading model from best epoch ...
2023-10-06 21:01:06,293 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
2023-10-06 21:01:13,499
Results:
- F-score (micro) 0.8903
- F-score (macro) 0.6643
- Accuracy 0.8197
By class:
precision recall f1-score support
scope 0.8977 0.8977 0.8977 176
pers 0.9154 0.9297 0.9225 128
work 0.8182 0.8514 0.8344 74
loc 1.0000 0.5000 0.6667 2
object 0.0000 0.0000 0.0000 2
micro avg 0.8880 0.8927 0.8903 382
macro avg 0.7263 0.6358 0.6643 382
weighted avg 0.8841 0.8927 0.8879 382
2023-10-06 21:01:13,499 ----------------------------------------------------------------------------------------------------
|