File size: 25,156 Bytes
cbdac7b |
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 15:14:54,549 ----------------------------------------------------------------------------------------------------
2023-10-06 15:14:54,550 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 15:14:54,550 ----------------------------------------------------------------------------------------------------
2023-10-06 15:14:54,551 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 15:14:54,551 ----------------------------------------------------------------------------------------------------
2023-10-06 15:14:54,551 Train: 1214 sentences
2023-10-06 15:14:54,551 (train_with_dev=False, train_with_test=False)
2023-10-06 15:14:54,551 ----------------------------------------------------------------------------------------------------
2023-10-06 15:14:54,551 Training Params:
2023-10-06 15:14:54,551 - learning_rate: "0.00015"
2023-10-06 15:14:54,551 - mini_batch_size: "4"
2023-10-06 15:14:54,551 - max_epochs: "10"
2023-10-06 15:14:54,551 - shuffle: "True"
2023-10-06 15:14:54,551 ----------------------------------------------------------------------------------------------------
2023-10-06 15:14:54,551 Plugins:
2023-10-06 15:14:54,551 - TensorboardLogger
2023-10-06 15:14:54,551 - LinearScheduler | warmup_fraction: '0.1'
2023-10-06 15:14:54,551 ----------------------------------------------------------------------------------------------------
2023-10-06 15:14:54,551 Final evaluation on model from best epoch (best-model.pt)
2023-10-06 15:14:54,551 - metric: "('micro avg', 'f1-score')"
2023-10-06 15:14:54,551 ----------------------------------------------------------------------------------------------------
2023-10-06 15:14:54,552 Computation:
2023-10-06 15:14:54,552 - compute on device: cuda:0
2023-10-06 15:14:54,552 - embedding storage: none
2023-10-06 15:14:54,552 ----------------------------------------------------------------------------------------------------
2023-10-06 15:14:54,552 Model training base path: "hmbench-ajmc/en-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-5"
2023-10-06 15:14:54,552 ----------------------------------------------------------------------------------------------------
2023-10-06 15:14:54,552 ----------------------------------------------------------------------------------------------------
2023-10-06 15:14:54,552 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-06 15:15:06,477 epoch 1 - iter 30/304 - loss 3.23343998 - time (sec): 11.92 - samples/sec: 268.11 - lr: 0.000014 - momentum: 0.000000
2023-10-06 15:15:17,847 epoch 1 - iter 60/304 - loss 3.22288360 - time (sec): 23.29 - samples/sec: 264.57 - lr: 0.000029 - momentum: 0.000000
2023-10-06 15:15:29,460 epoch 1 - iter 90/304 - loss 3.19705274 - time (sec): 34.91 - samples/sec: 256.68 - lr: 0.000044 - momentum: 0.000000
2023-10-06 15:15:41,241 epoch 1 - iter 120/304 - loss 3.13184691 - time (sec): 46.69 - samples/sec: 255.61 - lr: 0.000059 - momentum: 0.000000
2023-10-06 15:15:52,963 epoch 1 - iter 150/304 - loss 3.03341121 - time (sec): 58.41 - samples/sec: 254.94 - lr: 0.000074 - momentum: 0.000000
2023-10-06 15:16:05,924 epoch 1 - iter 180/304 - loss 2.90838872 - time (sec): 71.37 - samples/sec: 256.59 - lr: 0.000088 - momentum: 0.000000
2023-10-06 15:16:16,857 epoch 1 - iter 210/304 - loss 2.80542255 - time (sec): 82.30 - samples/sec: 253.58 - lr: 0.000103 - momentum: 0.000000
2023-10-06 15:16:29,245 epoch 1 - iter 240/304 - loss 2.66521471 - time (sec): 94.69 - samples/sec: 253.98 - lr: 0.000118 - momentum: 0.000000
2023-10-06 15:16:41,369 epoch 1 - iter 270/304 - loss 2.51887914 - time (sec): 106.82 - samples/sec: 254.87 - lr: 0.000133 - momentum: 0.000000
2023-10-06 15:16:53,851 epoch 1 - iter 300/304 - loss 2.36748077 - time (sec): 119.30 - samples/sec: 256.24 - lr: 0.000148 - momentum: 0.000000
2023-10-06 15:16:55,437 ----------------------------------------------------------------------------------------------------
2023-10-06 15:16:55,437 EPOCH 1 done: loss 2.3481 - lr: 0.000148
2023-10-06 15:17:03,270 DEV : loss 0.9383919835090637 - f1-score (micro avg) 0.0
2023-10-06 15:17:03,278 ----------------------------------------------------------------------------------------------------
2023-10-06 15:17:15,100 epoch 2 - iter 30/304 - loss 0.83648571 - time (sec): 11.82 - samples/sec: 259.23 - lr: 0.000148 - momentum: 0.000000
2023-10-06 15:17:26,764 epoch 2 - iter 60/304 - loss 0.76649649 - time (sec): 23.48 - samples/sec: 259.62 - lr: 0.000147 - momentum: 0.000000
2023-10-06 15:17:39,176 epoch 2 - iter 90/304 - loss 0.74619072 - time (sec): 35.90 - samples/sec: 259.50 - lr: 0.000145 - momentum: 0.000000
2023-10-06 15:17:51,335 epoch 2 - iter 120/304 - loss 0.70922269 - time (sec): 48.06 - samples/sec: 260.51 - lr: 0.000143 - momentum: 0.000000
2023-10-06 15:18:02,189 epoch 2 - iter 150/304 - loss 0.66968302 - time (sec): 58.91 - samples/sec: 255.92 - lr: 0.000142 - momentum: 0.000000
2023-10-06 15:18:13,839 epoch 2 - iter 180/304 - loss 0.61907359 - time (sec): 70.56 - samples/sec: 256.62 - lr: 0.000140 - momentum: 0.000000
2023-10-06 15:18:26,033 epoch 2 - iter 210/304 - loss 0.59912542 - time (sec): 82.75 - samples/sec: 255.28 - lr: 0.000139 - momentum: 0.000000
2023-10-06 15:18:38,038 epoch 2 - iter 240/304 - loss 0.56612844 - time (sec): 94.76 - samples/sec: 256.12 - lr: 0.000137 - momentum: 0.000000
2023-10-06 15:18:50,269 epoch 2 - iter 270/304 - loss 0.53713759 - time (sec): 106.99 - samples/sec: 256.87 - lr: 0.000135 - momentum: 0.000000
2023-10-06 15:19:02,163 epoch 2 - iter 300/304 - loss 0.51203772 - time (sec): 118.88 - samples/sec: 257.54 - lr: 0.000134 - momentum: 0.000000
2023-10-06 15:19:03,653 ----------------------------------------------------------------------------------------------------
2023-10-06 15:19:03,653 EPOCH 2 done: loss 0.5116 - lr: 0.000134
2023-10-06 15:19:11,497 DEV : loss 0.3360927402973175 - f1-score (micro avg) 0.4839
2023-10-06 15:19:11,505 saving best model
2023-10-06 15:19:12,492 ----------------------------------------------------------------------------------------------------
2023-10-06 15:19:24,035 epoch 3 - iter 30/304 - loss 0.34846495 - time (sec): 11.54 - samples/sec: 250.32 - lr: 0.000132 - momentum: 0.000000
2023-10-06 15:19:35,707 epoch 3 - iter 60/304 - loss 0.29775882 - time (sec): 23.21 - samples/sec: 252.45 - lr: 0.000130 - momentum: 0.000000
2023-10-06 15:19:46,823 epoch 3 - iter 90/304 - loss 0.27016891 - time (sec): 34.33 - samples/sec: 250.66 - lr: 0.000128 - momentum: 0.000000
2023-10-06 15:19:59,541 epoch 3 - iter 120/304 - loss 0.26645099 - time (sec): 47.05 - samples/sec: 254.47 - lr: 0.000127 - momentum: 0.000000
2023-10-06 15:20:11,289 epoch 3 - iter 150/304 - loss 0.25301882 - time (sec): 58.80 - samples/sec: 253.29 - lr: 0.000125 - momentum: 0.000000
2023-10-06 15:20:23,403 epoch 3 - iter 180/304 - loss 0.24885049 - time (sec): 70.91 - samples/sec: 253.23 - lr: 0.000124 - momentum: 0.000000
2023-10-06 15:20:35,584 epoch 3 - iter 210/304 - loss 0.23877391 - time (sec): 83.09 - samples/sec: 253.89 - lr: 0.000122 - momentum: 0.000000
2023-10-06 15:20:47,834 epoch 3 - iter 240/304 - loss 0.23063199 - time (sec): 95.34 - samples/sec: 254.60 - lr: 0.000120 - momentum: 0.000000
2023-10-06 15:20:59,711 epoch 3 - iter 270/304 - loss 0.22124203 - time (sec): 107.22 - samples/sec: 254.68 - lr: 0.000119 - momentum: 0.000000
2023-10-06 15:21:11,967 epoch 3 - iter 300/304 - loss 0.21317194 - time (sec): 119.47 - samples/sec: 255.63 - lr: 0.000117 - momentum: 0.000000
2023-10-06 15:21:13,656 ----------------------------------------------------------------------------------------------------
2023-10-06 15:21:13,656 EPOCH 3 done: loss 0.2127 - lr: 0.000117
2023-10-06 15:21:21,677 DEV : loss 0.19596518576145172 - f1-score (micro avg) 0.6897
2023-10-06 15:21:21,685 saving best model
2023-10-06 15:21:25,879 ----------------------------------------------------------------------------------------------------
2023-10-06 15:21:38,154 epoch 4 - iter 30/304 - loss 0.16356280 - time (sec): 12.27 - samples/sec: 257.95 - lr: 0.000115 - momentum: 0.000000
2023-10-06 15:21:49,440 epoch 4 - iter 60/304 - loss 0.15394459 - time (sec): 23.56 - samples/sec: 251.28 - lr: 0.000113 - momentum: 0.000000
2023-10-06 15:22:01,355 epoch 4 - iter 90/304 - loss 0.14506530 - time (sec): 35.47 - samples/sec: 252.52 - lr: 0.000112 - momentum: 0.000000
2023-10-06 15:22:13,307 epoch 4 - iter 120/304 - loss 0.13973248 - time (sec): 47.43 - samples/sec: 253.42 - lr: 0.000110 - momentum: 0.000000
2023-10-06 15:22:25,445 epoch 4 - iter 150/304 - loss 0.13837821 - time (sec): 59.57 - samples/sec: 253.71 - lr: 0.000109 - momentum: 0.000000
2023-10-06 15:22:37,347 epoch 4 - iter 180/304 - loss 0.13092858 - time (sec): 71.47 - samples/sec: 254.19 - lr: 0.000107 - momentum: 0.000000
2023-10-06 15:22:48,932 epoch 4 - iter 210/304 - loss 0.12807551 - time (sec): 83.05 - samples/sec: 253.80 - lr: 0.000105 - momentum: 0.000000
2023-10-06 15:23:01,772 epoch 4 - iter 240/304 - loss 0.12378548 - time (sec): 95.89 - samples/sec: 256.80 - lr: 0.000104 - momentum: 0.000000
2023-10-06 15:23:14,397 epoch 4 - iter 270/304 - loss 0.12188333 - time (sec): 108.52 - samples/sec: 257.16 - lr: 0.000102 - momentum: 0.000000
2023-10-06 15:23:25,691 epoch 4 - iter 300/304 - loss 0.11711521 - time (sec): 119.81 - samples/sec: 255.83 - lr: 0.000100 - momentum: 0.000000
2023-10-06 15:23:27,076 ----------------------------------------------------------------------------------------------------
2023-10-06 15:23:27,077 EPOCH 4 done: loss 0.1165 - lr: 0.000100
2023-10-06 15:23:34,950 DEV : loss 0.14813771843910217 - f1-score (micro avg) 0.7958
2023-10-06 15:23:34,958 saving best model
2023-10-06 15:23:39,268 ----------------------------------------------------------------------------------------------------
2023-10-06 15:23:51,238 epoch 5 - iter 30/304 - loss 0.05189033 - time (sec): 11.97 - samples/sec: 256.91 - lr: 0.000098 - momentum: 0.000000
2023-10-06 15:24:02,832 epoch 5 - iter 60/304 - loss 0.07692114 - time (sec): 23.56 - samples/sec: 257.49 - lr: 0.000097 - momentum: 0.000000
2023-10-06 15:24:15,080 epoch 5 - iter 90/304 - loss 0.07768443 - time (sec): 35.81 - samples/sec: 258.38 - lr: 0.000095 - momentum: 0.000000
2023-10-06 15:24:27,053 epoch 5 - iter 120/304 - loss 0.07680150 - time (sec): 47.78 - samples/sec: 259.83 - lr: 0.000094 - momentum: 0.000000
2023-10-06 15:24:39,247 epoch 5 - iter 150/304 - loss 0.08172290 - time (sec): 59.98 - samples/sec: 260.31 - lr: 0.000092 - momentum: 0.000000
2023-10-06 15:24:50,853 epoch 5 - iter 180/304 - loss 0.07660559 - time (sec): 71.58 - samples/sec: 259.08 - lr: 0.000090 - momentum: 0.000000
2023-10-06 15:25:03,175 epoch 5 - iter 210/304 - loss 0.07248568 - time (sec): 83.91 - samples/sec: 259.14 - lr: 0.000089 - momentum: 0.000000
2023-10-06 15:25:15,136 epoch 5 - iter 240/304 - loss 0.07334292 - time (sec): 95.87 - samples/sec: 259.80 - lr: 0.000087 - momentum: 0.000000
2023-10-06 15:25:26,971 epoch 5 - iter 270/304 - loss 0.07069816 - time (sec): 107.70 - samples/sec: 258.59 - lr: 0.000085 - momentum: 0.000000
2023-10-06 15:25:38,568 epoch 5 - iter 300/304 - loss 0.07232378 - time (sec): 119.30 - samples/sec: 257.86 - lr: 0.000084 - momentum: 0.000000
2023-10-06 15:25:39,696 ----------------------------------------------------------------------------------------------------
2023-10-06 15:25:39,697 EPOCH 5 done: loss 0.0719 - lr: 0.000084
2023-10-06 15:25:47,674 DEV : loss 0.14675813913345337 - f1-score (micro avg) 0.7995
2023-10-06 15:25:47,683 saving best model
2023-10-06 15:25:52,000 ----------------------------------------------------------------------------------------------------
2023-10-06 15:26:03,984 epoch 6 - iter 30/304 - loss 0.08138966 - time (sec): 11.98 - samples/sec: 265.72 - lr: 0.000082 - momentum: 0.000000
2023-10-06 15:26:16,058 epoch 6 - iter 60/304 - loss 0.06957171 - time (sec): 24.06 - samples/sec: 267.62 - lr: 0.000080 - momentum: 0.000000
2023-10-06 15:26:27,991 epoch 6 - iter 90/304 - loss 0.06031100 - time (sec): 35.99 - samples/sec: 270.50 - lr: 0.000079 - momentum: 0.000000
2023-10-06 15:26:39,420 epoch 6 - iter 120/304 - loss 0.05820154 - time (sec): 47.42 - samples/sec: 270.86 - lr: 0.000077 - momentum: 0.000000
2023-10-06 15:26:50,563 epoch 6 - iter 150/304 - loss 0.05286555 - time (sec): 58.56 - samples/sec: 271.87 - lr: 0.000075 - momentum: 0.000000
2023-10-06 15:27:00,994 epoch 6 - iter 180/304 - loss 0.04973384 - time (sec): 68.99 - samples/sec: 269.38 - lr: 0.000074 - momentum: 0.000000
2023-10-06 15:27:12,623 epoch 6 - iter 210/304 - loss 0.05049495 - time (sec): 80.62 - samples/sec: 270.82 - lr: 0.000072 - momentum: 0.000000
2023-10-06 15:27:23,331 epoch 6 - iter 240/304 - loss 0.05100559 - time (sec): 91.33 - samples/sec: 270.37 - lr: 0.000070 - momentum: 0.000000
2023-10-06 15:27:34,415 epoch 6 - iter 270/304 - loss 0.04940749 - time (sec): 102.41 - samples/sec: 269.52 - lr: 0.000069 - momentum: 0.000000
2023-10-06 15:27:45,754 epoch 6 - iter 300/304 - loss 0.05137662 - time (sec): 113.75 - samples/sec: 269.68 - lr: 0.000067 - momentum: 0.000000
2023-10-06 15:27:46,967 ----------------------------------------------------------------------------------------------------
2023-10-06 15:27:46,968 EPOCH 6 done: loss 0.0525 - lr: 0.000067
2023-10-06 15:27:54,072 DEV : loss 0.1478758007287979 - f1-score (micro avg) 0.8151
2023-10-06 15:27:54,080 saving best model
2023-10-06 15:27:58,409 ----------------------------------------------------------------------------------------------------
2023-10-06 15:28:09,207 epoch 7 - iter 30/304 - loss 0.02944793 - time (sec): 10.80 - samples/sec: 265.64 - lr: 0.000065 - momentum: 0.000000
2023-10-06 15:28:19,973 epoch 7 - iter 60/304 - loss 0.03347190 - time (sec): 21.56 - samples/sec: 266.85 - lr: 0.000063 - momentum: 0.000000
2023-10-06 15:28:31,176 epoch 7 - iter 90/304 - loss 0.02834258 - time (sec): 32.77 - samples/sec: 267.36 - lr: 0.000062 - momentum: 0.000000
2023-10-06 15:28:42,314 epoch 7 - iter 120/304 - loss 0.03031835 - time (sec): 43.90 - samples/sec: 269.43 - lr: 0.000060 - momentum: 0.000000
2023-10-06 15:28:53,456 epoch 7 - iter 150/304 - loss 0.02974656 - time (sec): 55.04 - samples/sec: 269.16 - lr: 0.000059 - momentum: 0.000000
2023-10-06 15:29:05,155 epoch 7 - iter 180/304 - loss 0.03371158 - time (sec): 66.74 - samples/sec: 271.20 - lr: 0.000057 - momentum: 0.000000
2023-10-06 15:29:16,354 epoch 7 - iter 210/304 - loss 0.03208932 - time (sec): 77.94 - samples/sec: 272.03 - lr: 0.000055 - momentum: 0.000000
2023-10-06 15:29:27,622 epoch 7 - iter 240/304 - loss 0.03349765 - time (sec): 89.21 - samples/sec: 270.88 - lr: 0.000054 - momentum: 0.000000
2023-10-06 15:29:39,306 epoch 7 - iter 270/304 - loss 0.03330430 - time (sec): 100.90 - samples/sec: 272.61 - lr: 0.000052 - momentum: 0.000000
2023-10-06 15:29:50,864 epoch 7 - iter 300/304 - loss 0.04026473 - time (sec): 112.45 - samples/sec: 272.22 - lr: 0.000050 - momentum: 0.000000
2023-10-06 15:29:52,226 ----------------------------------------------------------------------------------------------------
2023-10-06 15:29:52,226 EPOCH 7 done: loss 0.0398 - lr: 0.000050
2023-10-06 15:29:59,692 DEV : loss 0.15869873762130737 - f1-score (micro avg) 0.8254
2023-10-06 15:29:59,700 saving best model
2023-10-06 15:30:04,025 ----------------------------------------------------------------------------------------------------
2023-10-06 15:30:15,233 epoch 8 - iter 30/304 - loss 0.04093775 - time (sec): 11.21 - samples/sec: 270.27 - lr: 0.000048 - momentum: 0.000000
2023-10-06 15:30:26,471 epoch 8 - iter 60/304 - loss 0.04151933 - time (sec): 22.44 - samples/sec: 266.16 - lr: 0.000047 - momentum: 0.000000
2023-10-06 15:30:37,321 epoch 8 - iter 90/304 - loss 0.03716224 - time (sec): 33.30 - samples/sec: 261.21 - lr: 0.000045 - momentum: 0.000000
2023-10-06 15:30:48,960 epoch 8 - iter 120/304 - loss 0.03713580 - time (sec): 44.93 - samples/sec: 262.01 - lr: 0.000044 - momentum: 0.000000
2023-10-06 15:31:01,047 epoch 8 - iter 150/304 - loss 0.03630827 - time (sec): 57.02 - samples/sec: 262.85 - lr: 0.000042 - momentum: 0.000000
2023-10-06 15:31:12,792 epoch 8 - iter 180/304 - loss 0.03460352 - time (sec): 68.77 - samples/sec: 261.06 - lr: 0.000040 - momentum: 0.000000
2023-10-06 15:31:25,105 epoch 8 - iter 210/304 - loss 0.03214098 - time (sec): 81.08 - samples/sec: 261.36 - lr: 0.000039 - momentum: 0.000000
2023-10-06 15:31:37,569 epoch 8 - iter 240/304 - loss 0.03134757 - time (sec): 93.54 - samples/sec: 262.97 - lr: 0.000037 - momentum: 0.000000
2023-10-06 15:31:49,741 epoch 8 - iter 270/304 - loss 0.03276907 - time (sec): 105.72 - samples/sec: 262.73 - lr: 0.000035 - momentum: 0.000000
2023-10-06 15:32:00,861 epoch 8 - iter 300/304 - loss 0.03405317 - time (sec): 116.83 - samples/sec: 261.69 - lr: 0.000034 - momentum: 0.000000
2023-10-06 15:32:02,424 ----------------------------------------------------------------------------------------------------
2023-10-06 15:32:02,424 EPOCH 8 done: loss 0.0339 - lr: 0.000034
2023-10-06 15:32:09,779 DEV : loss 0.1572294384241104 - f1-score (micro avg) 0.837
2023-10-06 15:32:09,791 saving best model
2023-10-06 15:32:14,497 ----------------------------------------------------------------------------------------------------
2023-10-06 15:32:26,418 epoch 9 - iter 30/304 - loss 0.03926648 - time (sec): 11.92 - samples/sec: 283.50 - lr: 0.000032 - momentum: 0.000000
2023-10-06 15:32:37,728 epoch 9 - iter 60/304 - loss 0.03292966 - time (sec): 23.23 - samples/sec: 282.53 - lr: 0.000030 - momentum: 0.000000
2023-10-06 15:32:48,745 epoch 9 - iter 90/304 - loss 0.03106427 - time (sec): 34.25 - samples/sec: 278.04 - lr: 0.000029 - momentum: 0.000000
2023-10-06 15:32:59,485 epoch 9 - iter 120/304 - loss 0.03029571 - time (sec): 44.99 - samples/sec: 275.64 - lr: 0.000027 - momentum: 0.000000
2023-10-06 15:33:10,627 epoch 9 - iter 150/304 - loss 0.02641133 - time (sec): 56.13 - samples/sec: 273.14 - lr: 0.000025 - momentum: 0.000000
2023-10-06 15:33:22,017 epoch 9 - iter 180/304 - loss 0.02607911 - time (sec): 67.52 - samples/sec: 273.51 - lr: 0.000024 - momentum: 0.000000
2023-10-06 15:33:33,639 epoch 9 - iter 210/304 - loss 0.02502963 - time (sec): 79.14 - samples/sec: 273.97 - lr: 0.000022 - momentum: 0.000000
2023-10-06 15:33:44,669 epoch 9 - iter 240/304 - loss 0.02840878 - time (sec): 90.17 - samples/sec: 272.65 - lr: 0.000020 - momentum: 0.000000
2023-10-06 15:33:55,629 epoch 9 - iter 270/304 - loss 0.02606547 - time (sec): 101.13 - samples/sec: 272.82 - lr: 0.000019 - momentum: 0.000000
2023-10-06 15:34:06,757 epoch 9 - iter 300/304 - loss 0.02663716 - time (sec): 112.26 - samples/sec: 272.69 - lr: 0.000017 - momentum: 0.000000
2023-10-06 15:34:08,047 ----------------------------------------------------------------------------------------------------
2023-10-06 15:34:08,047 EPOCH 9 done: loss 0.0270 - lr: 0.000017
2023-10-06 15:34:15,070 DEV : loss 0.16326496005058289 - f1-score (micro avg) 0.8359
2023-10-06 15:34:15,077 ----------------------------------------------------------------------------------------------------
2023-10-06 15:34:25,865 epoch 10 - iter 30/304 - loss 0.03134249 - time (sec): 10.79 - samples/sec: 266.42 - lr: 0.000015 - momentum: 0.000000
2023-10-06 15:34:38,025 epoch 10 - iter 60/304 - loss 0.02757941 - time (sec): 22.95 - samples/sec: 277.85 - lr: 0.000014 - momentum: 0.000000
2023-10-06 15:34:49,214 epoch 10 - iter 90/304 - loss 0.02547701 - time (sec): 34.14 - samples/sec: 274.40 - lr: 0.000012 - momentum: 0.000000
2023-10-06 15:35:00,419 epoch 10 - iter 120/304 - loss 0.02355033 - time (sec): 45.34 - samples/sec: 272.55 - lr: 0.000010 - momentum: 0.000000
2023-10-06 15:35:11,177 epoch 10 - iter 150/304 - loss 0.02498782 - time (sec): 56.10 - samples/sec: 269.65 - lr: 0.000009 - momentum: 0.000000
2023-10-06 15:35:22,614 epoch 10 - iter 180/304 - loss 0.02443102 - time (sec): 67.54 - samples/sec: 270.51 - lr: 0.000007 - momentum: 0.000000
2023-10-06 15:35:33,818 epoch 10 - iter 210/304 - loss 0.02205130 - time (sec): 78.74 - samples/sec: 267.78 - lr: 0.000005 - momentum: 0.000000
2023-10-06 15:35:45,708 epoch 10 - iter 240/304 - loss 0.02222944 - time (sec): 90.63 - samples/sec: 268.27 - lr: 0.000004 - momentum: 0.000000
2023-10-06 15:35:57,393 epoch 10 - iter 270/304 - loss 0.02155449 - time (sec): 102.31 - samples/sec: 268.74 - lr: 0.000002 - momentum: 0.000000
2023-10-06 15:36:09,165 epoch 10 - iter 300/304 - loss 0.02313283 - time (sec): 114.09 - samples/sec: 269.33 - lr: 0.000000 - momentum: 0.000000
2023-10-06 15:36:10,383 ----------------------------------------------------------------------------------------------------
2023-10-06 15:36:10,383 EPOCH 10 done: loss 0.0237 - lr: 0.000000
2023-10-06 15:36:18,002 DEV : loss 0.16279256343841553 - f1-score (micro avg) 0.8302
2023-10-06 15:36:18,831 ----------------------------------------------------------------------------------------------------
2023-10-06 15:36:18,832 Loading model from best epoch ...
2023-10-06 15:36:21,516 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 15:36:28,563
Results:
- F-score (micro) 0.8104
- F-score (macro) 0.6512
- Accuracy 0.686
By class:
precision recall f1-score support
scope 0.7707 0.8013 0.7857 151
work 0.7321 0.8632 0.7923 95
pers 0.8257 0.9375 0.8780 96
loc 1.0000 0.6667 0.8000 3
date 0.0000 0.0000 0.0000 3
micro avg 0.7763 0.8477 0.8104 348
macro avg 0.6657 0.6537 0.6512 348
weighted avg 0.7707 0.8477 0.8063 348
2023-10-06 15:36:28,563 ----------------------------------------------------------------------------------------------------
|