File size: 25,078 Bytes
79abd98 |
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 |
2023-10-06 13:08:08,875 ----------------------------------------------------------------------------------------------------
2023-10-06 13:08:08,877 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 13:08:08,877 ----------------------------------------------------------------------------------------------------
2023-10-06 13:08:08,877 MultiCorpus: 1214 train + 266 dev + 251 test sentences
- NER_HIPE_2022 Corpus: 1214 train + 266 dev + 251 test sentences - /app/.flair/datasets/ner_hipe_2022/v2.1/ajmc/en/with_doc_seperator
2023-10-06 13:08:08,877 ----------------------------------------------------------------------------------------------------
2023-10-06 13:08:08,877 Train: 1214 sentences
2023-10-06 13:08:08,877 (train_with_dev=False, train_with_test=False)
2023-10-06 13:08:08,877 ----------------------------------------------------------------------------------------------------
2023-10-06 13:08:08,877 Training Params:
2023-10-06 13:08:08,877 - learning_rate: "0.00015"
2023-10-06 13:08:08,877 - mini_batch_size: "8"
2023-10-06 13:08:08,877 - max_epochs: "10"
2023-10-06 13:08:08,877 - shuffle: "True"
2023-10-06 13:08:08,877 ----------------------------------------------------------------------------------------------------
2023-10-06 13:08:08,878 Plugins:
2023-10-06 13:08:08,878 - TensorboardLogger
2023-10-06 13:08:08,878 - LinearScheduler | warmup_fraction: '0.1'
2023-10-06 13:08:08,878 ----------------------------------------------------------------------------------------------------
2023-10-06 13:08:08,878 Final evaluation on model from best epoch (best-model.pt)
2023-10-06 13:08:08,878 - metric: "('micro avg', 'f1-score')"
2023-10-06 13:08:08,878 ----------------------------------------------------------------------------------------------------
2023-10-06 13:08:08,878 Computation:
2023-10-06 13:08:08,878 - compute on device: cuda:0
2023-10-06 13:08:08,878 - embedding storage: none
2023-10-06 13:08:08,878 ----------------------------------------------------------------------------------------------------
2023-10-06 13:08:08,878 Model training base path: "hmbench-ajmc/en-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-3"
2023-10-06 13:08:08,878 ----------------------------------------------------------------------------------------------------
2023-10-06 13:08:08,878 ----------------------------------------------------------------------------------------------------
2023-10-06 13:08:08,878 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-06 13:08:19,434 epoch 1 - iter 15/152 - loss 3.21438031 - time (sec): 10.55 - samples/sec: 279.61 - lr: 0.000014 - momentum: 0.000000
2023-10-06 13:08:30,150 epoch 1 - iter 30/152 - loss 3.20540351 - time (sec): 21.27 - samples/sec: 280.91 - lr: 0.000029 - momentum: 0.000000
2023-10-06 13:08:40,815 epoch 1 - iter 45/152 - loss 3.19559217 - time (sec): 31.94 - samples/sec: 283.98 - lr: 0.000043 - momentum: 0.000000
2023-10-06 13:08:51,120 epoch 1 - iter 60/152 - loss 3.17725347 - time (sec): 42.24 - samples/sec: 282.24 - lr: 0.000058 - momentum: 0.000000
2023-10-06 13:09:01,354 epoch 1 - iter 75/152 - loss 3.14091169 - time (sec): 52.47 - samples/sec: 285.15 - lr: 0.000073 - momentum: 0.000000
2023-10-06 13:09:11,547 epoch 1 - iter 90/152 - loss 3.07834090 - time (sec): 62.67 - samples/sec: 286.56 - lr: 0.000088 - momentum: 0.000000
2023-10-06 13:09:21,906 epoch 1 - iter 105/152 - loss 2.99906148 - time (sec): 73.03 - samples/sec: 287.96 - lr: 0.000103 - momentum: 0.000000
2023-10-06 13:09:32,764 epoch 1 - iter 120/152 - loss 2.90338465 - time (sec): 83.88 - samples/sec: 290.77 - lr: 0.000117 - momentum: 0.000000
2023-10-06 13:09:43,670 epoch 1 - iter 135/152 - loss 2.80969823 - time (sec): 94.79 - samples/sec: 289.70 - lr: 0.000132 - momentum: 0.000000
2023-10-06 13:09:53,984 epoch 1 - iter 150/152 - loss 2.70892700 - time (sec): 105.10 - samples/sec: 291.40 - lr: 0.000147 - momentum: 0.000000
2023-10-06 13:09:55,194 ----------------------------------------------------------------------------------------------------
2023-10-06 13:09:55,194 EPOCH 1 done: loss 2.6974 - lr: 0.000147
2023-10-06 13:10:02,399 DEV : loss 1.6305921077728271 - f1-score (micro avg) 0.0
2023-10-06 13:10:02,412 ----------------------------------------------------------------------------------------------------
2023-10-06 13:10:12,934 epoch 2 - iter 15/152 - loss 1.57320305 - time (sec): 10.52 - samples/sec: 300.00 - lr: 0.000148 - momentum: 0.000000
2023-10-06 13:10:23,466 epoch 2 - iter 30/152 - loss 1.43312210 - time (sec): 21.05 - samples/sec: 300.59 - lr: 0.000147 - momentum: 0.000000
2023-10-06 13:10:33,164 epoch 2 - iter 45/152 - loss 1.30679236 - time (sec): 30.75 - samples/sec: 293.75 - lr: 0.000145 - momentum: 0.000000
2023-10-06 13:10:43,685 epoch 2 - iter 60/152 - loss 1.22453391 - time (sec): 41.27 - samples/sec: 292.66 - lr: 0.000144 - momentum: 0.000000
2023-10-06 13:10:54,229 epoch 2 - iter 75/152 - loss 1.15899750 - time (sec): 51.81 - samples/sec: 291.77 - lr: 0.000142 - momentum: 0.000000
2023-10-06 13:11:04,548 epoch 2 - iter 90/152 - loss 1.10009267 - time (sec): 62.13 - samples/sec: 292.05 - lr: 0.000140 - momentum: 0.000000
2023-10-06 13:11:14,863 epoch 2 - iter 105/152 - loss 1.03295088 - time (sec): 72.45 - samples/sec: 292.01 - lr: 0.000139 - momentum: 0.000000
2023-10-06 13:11:25,393 epoch 2 - iter 120/152 - loss 0.98652304 - time (sec): 82.98 - samples/sec: 291.54 - lr: 0.000137 - momentum: 0.000000
2023-10-06 13:11:36,515 epoch 2 - iter 135/152 - loss 0.93594003 - time (sec): 94.10 - samples/sec: 291.46 - lr: 0.000135 - momentum: 0.000000
2023-10-06 13:11:47,543 epoch 2 - iter 150/152 - loss 0.89467702 - time (sec): 105.13 - samples/sec: 292.02 - lr: 0.000134 - momentum: 0.000000
2023-10-06 13:11:48,621 ----------------------------------------------------------------------------------------------------
2023-10-06 13:11:48,621 EPOCH 2 done: loss 0.8913 - lr: 0.000134
2023-10-06 13:11:55,927 DEV : loss 0.5647168755531311 - f1-score (micro avg) 0.0
2023-10-06 13:11:55,934 ----------------------------------------------------------------------------------------------------
2023-10-06 13:12:06,278 epoch 3 - iter 15/152 - loss 0.55208629 - time (sec): 10.34 - samples/sec: 282.72 - lr: 0.000132 - momentum: 0.000000
2023-10-06 13:12:16,529 epoch 3 - iter 30/152 - loss 0.49683238 - time (sec): 20.59 - samples/sec: 280.97 - lr: 0.000130 - momentum: 0.000000
2023-10-06 13:12:27,371 epoch 3 - iter 45/152 - loss 0.46337786 - time (sec): 31.44 - samples/sec: 282.49 - lr: 0.000129 - momentum: 0.000000
2023-10-06 13:12:38,640 epoch 3 - iter 60/152 - loss 0.44837662 - time (sec): 42.70 - samples/sec: 285.22 - lr: 0.000127 - momentum: 0.000000
2023-10-06 13:12:49,081 epoch 3 - iter 75/152 - loss 0.41717800 - time (sec): 53.15 - samples/sec: 283.47 - lr: 0.000125 - momentum: 0.000000
2023-10-06 13:13:00,259 epoch 3 - iter 90/152 - loss 0.41097854 - time (sec): 64.32 - samples/sec: 282.62 - lr: 0.000124 - momentum: 0.000000
2023-10-06 13:13:11,316 epoch 3 - iter 105/152 - loss 0.40805842 - time (sec): 75.38 - samples/sec: 283.03 - lr: 0.000122 - momentum: 0.000000
2023-10-06 13:13:22,175 epoch 3 - iter 120/152 - loss 0.39498496 - time (sec): 86.24 - samples/sec: 282.03 - lr: 0.000120 - momentum: 0.000000
2023-10-06 13:13:32,934 epoch 3 - iter 135/152 - loss 0.38573774 - time (sec): 97.00 - samples/sec: 280.27 - lr: 0.000119 - momentum: 0.000000
2023-10-06 13:13:44,662 epoch 3 - iter 150/152 - loss 0.37384951 - time (sec): 108.73 - samples/sec: 281.55 - lr: 0.000117 - momentum: 0.000000
2023-10-06 13:13:46,031 ----------------------------------------------------------------------------------------------------
2023-10-06 13:13:46,032 EPOCH 3 done: loss 0.3753 - lr: 0.000117
2023-10-06 13:13:54,094 DEV : loss 0.32833757996559143 - f1-score (micro avg) 0.4878
2023-10-06 13:13:54,102 saving best model
2023-10-06 13:13:54,964 ----------------------------------------------------------------------------------------------------
2023-10-06 13:14:06,353 epoch 4 - iter 15/152 - loss 0.28006593 - time (sec): 11.39 - samples/sec: 274.59 - lr: 0.000115 - momentum: 0.000000
2023-10-06 13:14:17,205 epoch 4 - iter 30/152 - loss 0.28639218 - time (sec): 22.24 - samples/sec: 274.51 - lr: 0.000114 - momentum: 0.000000
2023-10-06 13:14:28,850 epoch 4 - iter 45/152 - loss 0.27015973 - time (sec): 33.88 - samples/sec: 278.86 - lr: 0.000112 - momentum: 0.000000
2023-10-06 13:14:40,557 epoch 4 - iter 60/152 - loss 0.26084485 - time (sec): 45.59 - samples/sec: 278.63 - lr: 0.000110 - momentum: 0.000000
2023-10-06 13:14:51,743 epoch 4 - iter 75/152 - loss 0.25334043 - time (sec): 56.78 - samples/sec: 279.25 - lr: 0.000109 - momentum: 0.000000
2023-10-06 13:15:03,188 epoch 4 - iter 90/152 - loss 0.24462008 - time (sec): 68.22 - samples/sec: 278.78 - lr: 0.000107 - momentum: 0.000000
2023-10-06 13:15:13,402 epoch 4 - iter 105/152 - loss 0.24376650 - time (sec): 78.44 - samples/sec: 277.79 - lr: 0.000105 - momentum: 0.000000
2023-10-06 13:15:24,629 epoch 4 - iter 120/152 - loss 0.24370889 - time (sec): 89.66 - samples/sec: 276.29 - lr: 0.000104 - momentum: 0.000000
2023-10-06 13:15:35,373 epoch 4 - iter 135/152 - loss 0.23999875 - time (sec): 100.41 - samples/sec: 275.42 - lr: 0.000102 - momentum: 0.000000
2023-10-06 13:15:46,458 epoch 4 - iter 150/152 - loss 0.23025240 - time (sec): 111.49 - samples/sec: 274.23 - lr: 0.000101 - momentum: 0.000000
2023-10-06 13:15:47,939 ----------------------------------------------------------------------------------------------------
2023-10-06 13:15:47,940 EPOCH 4 done: loss 0.2287 - lr: 0.000101
2023-10-06 13:15:56,008 DEV : loss 0.2277979999780655 - f1-score (micro avg) 0.7074
2023-10-06 13:15:56,016 saving best model
2023-10-06 13:16:00,368 ----------------------------------------------------------------------------------------------------
2023-10-06 13:16:11,445 epoch 5 - iter 15/152 - loss 0.12246111 - time (sec): 11.08 - samples/sec: 276.56 - lr: 0.000099 - momentum: 0.000000
2023-10-06 13:16:22,448 epoch 5 - iter 30/152 - loss 0.15322928 - time (sec): 22.08 - samples/sec: 273.43 - lr: 0.000097 - momentum: 0.000000
2023-10-06 13:16:34,190 epoch 5 - iter 45/152 - loss 0.15417556 - time (sec): 33.82 - samples/sec: 273.18 - lr: 0.000095 - momentum: 0.000000
2023-10-06 13:16:45,527 epoch 5 - iter 60/152 - loss 0.16403261 - time (sec): 45.16 - samples/sec: 274.29 - lr: 0.000094 - momentum: 0.000000
2023-10-06 13:16:56,764 epoch 5 - iter 75/152 - loss 0.16365678 - time (sec): 56.39 - samples/sec: 274.74 - lr: 0.000092 - momentum: 0.000000
2023-10-06 13:17:07,818 epoch 5 - iter 90/152 - loss 0.16139091 - time (sec): 67.45 - samples/sec: 273.62 - lr: 0.000091 - momentum: 0.000000
2023-10-06 13:17:18,564 epoch 5 - iter 105/152 - loss 0.15382459 - time (sec): 78.19 - samples/sec: 271.81 - lr: 0.000089 - momentum: 0.000000
2023-10-06 13:17:30,211 epoch 5 - iter 120/152 - loss 0.15240417 - time (sec): 89.84 - samples/sec: 273.78 - lr: 0.000087 - momentum: 0.000000
2023-10-06 13:17:41,001 epoch 5 - iter 135/152 - loss 0.15079292 - time (sec): 100.63 - samples/sec: 273.95 - lr: 0.000086 - momentum: 0.000000
2023-10-06 13:17:51,812 epoch 5 - iter 150/152 - loss 0.15298419 - time (sec): 111.44 - samples/sec: 274.10 - lr: 0.000084 - momentum: 0.000000
2023-10-06 13:17:53,395 ----------------------------------------------------------------------------------------------------
2023-10-06 13:17:53,395 EPOCH 5 done: loss 0.1519 - lr: 0.000084
2023-10-06 13:18:01,435 DEV : loss 0.17695870995521545 - f1-score (micro avg) 0.7258
2023-10-06 13:18:01,444 saving best model
2023-10-06 13:18:05,779 ----------------------------------------------------------------------------------------------------
2023-10-06 13:18:16,697 epoch 6 - iter 15/152 - loss 0.09270192 - time (sec): 10.92 - samples/sec: 271.43 - lr: 0.000082 - momentum: 0.000000
2023-10-06 13:18:27,716 epoch 6 - iter 30/152 - loss 0.11720667 - time (sec): 21.94 - samples/sec: 267.47 - lr: 0.000080 - momentum: 0.000000
2023-10-06 13:18:39,303 epoch 6 - iter 45/152 - loss 0.11175077 - time (sec): 33.52 - samples/sec: 270.35 - lr: 0.000079 - momentum: 0.000000
2023-10-06 13:18:50,781 epoch 6 - iter 60/152 - loss 0.11436783 - time (sec): 45.00 - samples/sec: 272.19 - lr: 0.000077 - momentum: 0.000000
2023-10-06 13:19:01,772 epoch 6 - iter 75/152 - loss 0.11082759 - time (sec): 55.99 - samples/sec: 271.26 - lr: 0.000076 - momentum: 0.000000
2023-10-06 13:19:13,070 epoch 6 - iter 90/152 - loss 0.10598764 - time (sec): 67.29 - samples/sec: 271.48 - lr: 0.000074 - momentum: 0.000000
2023-10-06 13:19:24,104 epoch 6 - iter 105/152 - loss 0.10720574 - time (sec): 78.32 - samples/sec: 270.72 - lr: 0.000072 - momentum: 0.000000
2023-10-06 13:19:35,155 epoch 6 - iter 120/152 - loss 0.11087016 - time (sec): 89.37 - samples/sec: 272.16 - lr: 0.000071 - momentum: 0.000000
2023-10-06 13:19:46,330 epoch 6 - iter 135/152 - loss 0.10716799 - time (sec): 100.55 - samples/sec: 273.19 - lr: 0.000069 - momentum: 0.000000
2023-10-06 13:19:57,472 epoch 6 - iter 150/152 - loss 0.10886342 - time (sec): 111.69 - samples/sec: 273.88 - lr: 0.000067 - momentum: 0.000000
2023-10-06 13:19:58,850 ----------------------------------------------------------------------------------------------------
2023-10-06 13:19:58,850 EPOCH 6 done: loss 0.1091 - lr: 0.000067
2023-10-06 13:20:06,860 DEV : loss 0.1589815467596054 - f1-score (micro avg) 0.7943
2023-10-06 13:20:06,868 saving best model
2023-10-06 13:20:11,225 ----------------------------------------------------------------------------------------------------
2023-10-06 13:20:22,447 epoch 7 - iter 15/152 - loss 0.12776069 - time (sec): 11.22 - samples/sec: 265.52 - lr: 0.000066 - momentum: 0.000000
2023-10-06 13:20:34,020 epoch 7 - iter 30/152 - loss 0.10877098 - time (sec): 22.79 - samples/sec: 273.51 - lr: 0.000064 - momentum: 0.000000
2023-10-06 13:20:44,843 epoch 7 - iter 45/152 - loss 0.09444140 - time (sec): 33.62 - samples/sec: 273.80 - lr: 0.000062 - momentum: 0.000000
2023-10-06 13:20:56,426 epoch 7 - iter 60/152 - loss 0.09188632 - time (sec): 45.20 - samples/sec: 275.76 - lr: 0.000061 - momentum: 0.000000
2023-10-06 13:21:07,589 epoch 7 - iter 75/152 - loss 0.09200131 - time (sec): 56.36 - samples/sec: 275.34 - lr: 0.000059 - momentum: 0.000000
2023-10-06 13:21:18,216 epoch 7 - iter 90/152 - loss 0.08825078 - time (sec): 66.99 - samples/sec: 272.46 - lr: 0.000057 - momentum: 0.000000
2023-10-06 13:21:28,988 epoch 7 - iter 105/152 - loss 0.08636856 - time (sec): 77.76 - samples/sec: 272.23 - lr: 0.000056 - momentum: 0.000000
2023-10-06 13:21:40,417 epoch 7 - iter 120/152 - loss 0.08420198 - time (sec): 89.19 - samples/sec: 273.10 - lr: 0.000054 - momentum: 0.000000
2023-10-06 13:21:51,929 epoch 7 - iter 135/152 - loss 0.08667895 - time (sec): 100.70 - samples/sec: 274.30 - lr: 0.000052 - momentum: 0.000000
2023-10-06 13:22:02,903 epoch 7 - iter 150/152 - loss 0.08327537 - time (sec): 111.68 - samples/sec: 274.30 - lr: 0.000051 - momentum: 0.000000
2023-10-06 13:22:04,222 ----------------------------------------------------------------------------------------------------
2023-10-06 13:22:04,223 EPOCH 7 done: loss 0.0830 - lr: 0.000051
2023-10-06 13:22:12,223 DEV : loss 0.14461469650268555 - f1-score (micro avg) 0.8189
2023-10-06 13:22:12,231 saving best model
2023-10-06 13:22:16,577 ----------------------------------------------------------------------------------------------------
2023-10-06 13:22:27,756 epoch 8 - iter 15/152 - loss 0.06828884 - time (sec): 11.18 - samples/sec: 281.09 - lr: 0.000049 - momentum: 0.000000
2023-10-06 13:22:39,406 epoch 8 - iter 30/152 - loss 0.07361348 - time (sec): 22.83 - samples/sec: 283.65 - lr: 0.000047 - momentum: 0.000000
2023-10-06 13:22:51,204 epoch 8 - iter 45/152 - loss 0.08124757 - time (sec): 34.63 - samples/sec: 283.46 - lr: 0.000046 - momentum: 0.000000
2023-10-06 13:23:02,401 epoch 8 - iter 60/152 - loss 0.07871064 - time (sec): 45.82 - samples/sec: 281.54 - lr: 0.000044 - momentum: 0.000000
2023-10-06 13:23:13,854 epoch 8 - iter 75/152 - loss 0.07422431 - time (sec): 57.28 - samples/sec: 280.31 - lr: 0.000042 - momentum: 0.000000
2023-10-06 13:23:24,913 epoch 8 - iter 90/152 - loss 0.07333499 - time (sec): 68.33 - samples/sec: 278.69 - lr: 0.000041 - momentum: 0.000000
2023-10-06 13:23:35,270 epoch 8 - iter 105/152 - loss 0.07143430 - time (sec): 78.69 - samples/sec: 276.28 - lr: 0.000039 - momentum: 0.000000
2023-10-06 13:23:46,331 epoch 8 - iter 120/152 - loss 0.06850039 - time (sec): 89.75 - samples/sec: 276.17 - lr: 0.000037 - momentum: 0.000000
2023-10-06 13:23:57,183 epoch 8 - iter 135/152 - loss 0.06792511 - time (sec): 100.60 - samples/sec: 275.50 - lr: 0.000036 - momentum: 0.000000
2023-10-06 13:24:07,826 epoch 8 - iter 150/152 - loss 0.06593987 - time (sec): 111.25 - samples/sec: 274.54 - lr: 0.000034 - momentum: 0.000000
2023-10-06 13:24:09,273 ----------------------------------------------------------------------------------------------------
2023-10-06 13:24:09,273 EPOCH 8 done: loss 0.0681 - lr: 0.000034
2023-10-06 13:24:17,123 DEV : loss 0.14291059970855713 - f1-score (micro avg) 0.807
2023-10-06 13:24:17,132 ----------------------------------------------------------------------------------------------------
2023-10-06 13:24:28,361 epoch 9 - iter 15/152 - loss 0.05127601 - time (sec): 11.23 - samples/sec: 282.41 - lr: 0.000032 - momentum: 0.000000
2023-10-06 13:24:38,888 epoch 9 - iter 30/152 - loss 0.05020165 - time (sec): 21.75 - samples/sec: 274.51 - lr: 0.000031 - momentum: 0.000000
2023-10-06 13:24:49,557 epoch 9 - iter 45/152 - loss 0.04825827 - time (sec): 32.42 - samples/sec: 272.33 - lr: 0.000029 - momentum: 0.000000
2023-10-06 13:25:01,312 epoch 9 - iter 60/152 - loss 0.05048857 - time (sec): 44.18 - samples/sec: 276.71 - lr: 0.000027 - momentum: 0.000000
2023-10-06 13:25:12,244 epoch 9 - iter 75/152 - loss 0.05688874 - time (sec): 55.11 - samples/sec: 275.16 - lr: 0.000026 - momentum: 0.000000
2023-10-06 13:25:23,129 epoch 9 - iter 90/152 - loss 0.05601915 - time (sec): 66.00 - samples/sec: 274.97 - lr: 0.000024 - momentum: 0.000000
2023-10-06 13:25:34,216 epoch 9 - iter 105/152 - loss 0.05737830 - time (sec): 77.08 - samples/sec: 276.08 - lr: 0.000022 - momentum: 0.000000
2023-10-06 13:25:45,044 epoch 9 - iter 120/152 - loss 0.05834607 - time (sec): 87.91 - samples/sec: 276.19 - lr: 0.000021 - momentum: 0.000000
2023-10-06 13:25:56,736 epoch 9 - iter 135/152 - loss 0.05753999 - time (sec): 99.60 - samples/sec: 276.94 - lr: 0.000019 - momentum: 0.000000
2023-10-06 13:26:07,630 epoch 9 - iter 150/152 - loss 0.05804039 - time (sec): 110.50 - samples/sec: 276.72 - lr: 0.000018 - momentum: 0.000000
2023-10-06 13:26:09,098 ----------------------------------------------------------------------------------------------------
2023-10-06 13:26:09,098 EPOCH 9 done: loss 0.0574 - lr: 0.000018
2023-10-06 13:26:16,886 DEV : loss 0.14050251245498657 - f1-score (micro avg) 0.8132
2023-10-06 13:26:16,894 ----------------------------------------------------------------------------------------------------
2023-10-06 13:26:28,153 epoch 10 - iter 15/152 - loss 0.04766949 - time (sec): 11.26 - samples/sec: 265.79 - lr: 0.000016 - momentum: 0.000000
2023-10-06 13:26:39,619 epoch 10 - iter 30/152 - loss 0.06175718 - time (sec): 22.72 - samples/sec: 269.68 - lr: 0.000014 - momentum: 0.000000
2023-10-06 13:26:50,158 epoch 10 - iter 45/152 - loss 0.05547825 - time (sec): 33.26 - samples/sec: 268.72 - lr: 0.000012 - momentum: 0.000000
2023-10-06 13:27:01,057 epoch 10 - iter 60/152 - loss 0.05417530 - time (sec): 44.16 - samples/sec: 268.78 - lr: 0.000011 - momentum: 0.000000
2023-10-06 13:27:12,421 epoch 10 - iter 75/152 - loss 0.05152617 - time (sec): 55.53 - samples/sec: 270.74 - lr: 0.000009 - momentum: 0.000000
2023-10-06 13:27:23,494 epoch 10 - iter 90/152 - loss 0.05355677 - time (sec): 66.60 - samples/sec: 272.15 - lr: 0.000008 - momentum: 0.000000
2023-10-06 13:27:34,374 epoch 10 - iter 105/152 - loss 0.05240993 - time (sec): 77.48 - samples/sec: 271.92 - lr: 0.000006 - momentum: 0.000000
2023-10-06 13:27:45,675 epoch 10 - iter 120/152 - loss 0.05321465 - time (sec): 88.78 - samples/sec: 273.51 - lr: 0.000004 - momentum: 0.000000
2023-10-06 13:27:57,530 epoch 10 - iter 135/152 - loss 0.05326204 - time (sec): 100.63 - samples/sec: 274.63 - lr: 0.000003 - momentum: 0.000000
2023-10-06 13:28:08,225 epoch 10 - iter 150/152 - loss 0.05277666 - time (sec): 111.33 - samples/sec: 274.78 - lr: 0.000001 - momentum: 0.000000
2023-10-06 13:28:09,594 ----------------------------------------------------------------------------------------------------
2023-10-06 13:28:09,594 EPOCH 10 done: loss 0.0527 - lr: 0.000001
2023-10-06 13:28:17,372 DEV : loss 0.13832153379917145 - f1-score (micro avg) 0.8217
2023-10-06 13:28:17,379 saving best model
2023-10-06 13:28:22,740 ----------------------------------------------------------------------------------------------------
2023-10-06 13:28:22,742 Loading model from best epoch ...
2023-10-06 13:28:25,333 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
2023-10-06 13:28:32,461
Results:
- F-score (micro) 0.7891
- F-score (macro) 0.4828
- Accuracy 0.6561
By class:
precision recall f1-score support
scope 0.7205 0.7682 0.7436 151
work 0.7155 0.8737 0.7867 95
pers 0.8273 0.9479 0.8835 96
loc 0.0000 0.0000 0.0000 3
date 0.0000 0.0000 0.0000 3
micro avg 0.7494 0.8333 0.7891 348
macro avg 0.4527 0.5180 0.4828 348
weighted avg 0.7362 0.8333 0.7811 348
2023-10-06 13:28:32,462 ----------------------------------------------------------------------------------------------------
|