File size: 25,116 Bytes
7e9aef1 |
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 12:02:00,712 ----------------------------------------------------------------------------------------------------
2023-10-06 12:02:00,713 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 12:02:00,714 ----------------------------------------------------------------------------------------------------
2023-10-06 12:02:00,714 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 12:02:00,714 ----------------------------------------------------------------------------------------------------
2023-10-06 12:02:00,714 Train: 1214 sentences
2023-10-06 12:02:00,715 (train_with_dev=False, train_with_test=False)
2023-10-06 12:02:00,715 ----------------------------------------------------------------------------------------------------
2023-10-06 12:02:00,715 Training Params:
2023-10-06 12:02:00,715 - learning_rate: "0.00016"
2023-10-06 12:02:00,715 - mini_batch_size: "8"
2023-10-06 12:02:00,715 - max_epochs: "10"
2023-10-06 12:02:00,715 - shuffle: "True"
2023-10-06 12:02:00,715 ----------------------------------------------------------------------------------------------------
2023-10-06 12:02:00,715 Plugins:
2023-10-06 12:02:00,715 - TensorboardLogger
2023-10-06 12:02:00,715 - LinearScheduler | warmup_fraction: '0.1'
2023-10-06 12:02:00,715 ----------------------------------------------------------------------------------------------------
2023-10-06 12:02:00,715 Final evaluation on model from best epoch (best-model.pt)
2023-10-06 12:02:00,715 - metric: "('micro avg', 'f1-score')"
2023-10-06 12:02:00,715 ----------------------------------------------------------------------------------------------------
2023-10-06 12:02:00,715 Computation:
2023-10-06 12:02:00,715 - compute on device: cuda:0
2023-10-06 12:02:00,715 - embedding storage: none
2023-10-06 12:02:00,715 ----------------------------------------------------------------------------------------------------
2023-10-06 12:02:00,715 Model training base path: "hmbench-ajmc/en-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-2"
2023-10-06 12:02:00,715 ----------------------------------------------------------------------------------------------------
2023-10-06 12:02:00,716 ----------------------------------------------------------------------------------------------------
2023-10-06 12:02:00,716 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-06 12:02:11,895 epoch 1 - iter 15/152 - loss 3.23637945 - time (sec): 11.18 - samples/sec: 276.71 - lr: 0.000015 - momentum: 0.000000
2023-10-06 12:02:23,232 epoch 1 - iter 30/152 - loss 3.23056237 - time (sec): 22.51 - samples/sec: 273.82 - lr: 0.000031 - momentum: 0.000000
2023-10-06 12:02:34,656 epoch 1 - iter 45/152 - loss 3.21826245 - time (sec): 33.94 - samples/sec: 276.05 - lr: 0.000046 - momentum: 0.000000
2023-10-06 12:02:46,712 epoch 1 - iter 60/152 - loss 3.19874969 - time (sec): 45.99 - samples/sec: 276.99 - lr: 0.000062 - momentum: 0.000000
2023-10-06 12:02:58,141 epoch 1 - iter 75/152 - loss 3.16240231 - time (sec): 57.42 - samples/sec: 275.75 - lr: 0.000078 - momentum: 0.000000
2023-10-06 12:03:09,690 epoch 1 - iter 90/152 - loss 3.09184866 - time (sec): 68.97 - samples/sec: 275.72 - lr: 0.000094 - momentum: 0.000000
2023-10-06 12:03:20,372 epoch 1 - iter 105/152 - loss 3.01217102 - time (sec): 79.66 - samples/sec: 273.94 - lr: 0.000109 - momentum: 0.000000
2023-10-06 12:03:31,376 epoch 1 - iter 120/152 - loss 2.91828001 - time (sec): 90.66 - samples/sec: 274.07 - lr: 0.000125 - momentum: 0.000000
2023-10-06 12:03:41,741 epoch 1 - iter 135/152 - loss 2.82691369 - time (sec): 101.02 - samples/sec: 272.84 - lr: 0.000141 - momentum: 0.000000
2023-10-06 12:03:52,791 epoch 1 - iter 150/152 - loss 2.71500413 - time (sec): 112.07 - samples/sec: 273.84 - lr: 0.000157 - momentum: 0.000000
2023-10-06 12:03:53,974 ----------------------------------------------------------------------------------------------------
2023-10-06 12:03:53,974 EPOCH 1 done: loss 2.7055 - lr: 0.000157
2023-10-06 12:04:01,808 DEV : loss 1.5486512184143066 - f1-score (micro avg) 0.0
2023-10-06 12:04:01,815 ----------------------------------------------------------------------------------------------------
2023-10-06 12:04:12,751 epoch 2 - iter 15/152 - loss 1.50030838 - time (sec): 10.93 - samples/sec: 281.31 - lr: 0.000158 - momentum: 0.000000
2023-10-06 12:04:23,678 epoch 2 - iter 30/152 - loss 1.33185743 - time (sec): 21.86 - samples/sec: 276.79 - lr: 0.000157 - momentum: 0.000000
2023-10-06 12:04:35,236 epoch 2 - iter 45/152 - loss 1.23214306 - time (sec): 33.42 - samples/sec: 275.44 - lr: 0.000155 - momentum: 0.000000
2023-10-06 12:04:46,142 epoch 2 - iter 60/152 - loss 1.14422649 - time (sec): 44.33 - samples/sec: 275.80 - lr: 0.000153 - momentum: 0.000000
2023-10-06 12:04:56,961 epoch 2 - iter 75/152 - loss 1.07649727 - time (sec): 55.14 - samples/sec: 273.52 - lr: 0.000151 - momentum: 0.000000
2023-10-06 12:05:08,495 epoch 2 - iter 90/152 - loss 1.00943491 - time (sec): 66.68 - samples/sec: 275.17 - lr: 0.000150 - momentum: 0.000000
2023-10-06 12:05:19,664 epoch 2 - iter 105/152 - loss 0.95211533 - time (sec): 77.85 - samples/sec: 275.28 - lr: 0.000148 - momentum: 0.000000
2023-10-06 12:05:30,154 epoch 2 - iter 120/152 - loss 0.92045512 - time (sec): 88.34 - samples/sec: 274.65 - lr: 0.000146 - momentum: 0.000000
2023-10-06 12:05:41,453 epoch 2 - iter 135/152 - loss 0.88873144 - time (sec): 99.64 - samples/sec: 275.74 - lr: 0.000144 - momentum: 0.000000
2023-10-06 12:05:52,434 epoch 2 - iter 150/152 - loss 0.85889893 - time (sec): 110.62 - samples/sec: 276.36 - lr: 0.000143 - momentum: 0.000000
2023-10-06 12:05:53,810 ----------------------------------------------------------------------------------------------------
2023-10-06 12:05:53,810 EPOCH 2 done: loss 0.8552 - lr: 0.000143
2023-10-06 12:06:01,705 DEV : loss 0.5358220338821411 - f1-score (micro avg) 0.0
2023-10-06 12:06:01,713 ----------------------------------------------------------------------------------------------------
2023-10-06 12:06:12,801 epoch 3 - iter 15/152 - loss 0.49234408 - time (sec): 11.09 - samples/sec: 278.45 - lr: 0.000141 - momentum: 0.000000
2023-10-06 12:06:23,338 epoch 3 - iter 30/152 - loss 0.44760087 - time (sec): 21.62 - samples/sec: 276.18 - lr: 0.000139 - momentum: 0.000000
2023-10-06 12:06:33,930 epoch 3 - iter 45/152 - loss 0.43275129 - time (sec): 32.22 - samples/sec: 275.92 - lr: 0.000137 - momentum: 0.000000
2023-10-06 12:06:46,232 epoch 3 - iter 60/152 - loss 0.42500690 - time (sec): 44.52 - samples/sec: 280.76 - lr: 0.000135 - momentum: 0.000000
2023-10-06 12:06:57,377 epoch 3 - iter 75/152 - loss 0.42265947 - time (sec): 55.66 - samples/sec: 279.97 - lr: 0.000134 - momentum: 0.000000
2023-10-06 12:07:08,635 epoch 3 - iter 90/152 - loss 0.39858999 - time (sec): 66.92 - samples/sec: 279.31 - lr: 0.000132 - momentum: 0.000000
2023-10-06 12:07:19,407 epoch 3 - iter 105/152 - loss 0.38199531 - time (sec): 77.69 - samples/sec: 277.89 - lr: 0.000130 - momentum: 0.000000
2023-10-06 12:07:30,430 epoch 3 - iter 120/152 - loss 0.37465901 - time (sec): 88.72 - samples/sec: 278.99 - lr: 0.000128 - momentum: 0.000000
2023-10-06 12:07:41,424 epoch 3 - iter 135/152 - loss 0.36271381 - time (sec): 99.71 - samples/sec: 278.43 - lr: 0.000127 - momentum: 0.000000
2023-10-06 12:07:52,291 epoch 3 - iter 150/152 - loss 0.36018016 - time (sec): 110.58 - samples/sec: 277.04 - lr: 0.000125 - momentum: 0.000000
2023-10-06 12:07:53,598 ----------------------------------------------------------------------------------------------------
2023-10-06 12:07:53,598 EPOCH 3 done: loss 0.3599 - lr: 0.000125
2023-10-06 12:08:01,407 DEV : loss 0.3210684657096863 - f1-score (micro avg) 0.488
2023-10-06 12:08:01,414 saving best model
2023-10-06 12:08:02,261 ----------------------------------------------------------------------------------------------------
2023-10-06 12:08:12,909 epoch 4 - iter 15/152 - loss 0.25179135 - time (sec): 10.65 - samples/sec: 264.71 - lr: 0.000123 - momentum: 0.000000
2023-10-06 12:08:23,932 epoch 4 - iter 30/152 - loss 0.25281864 - time (sec): 21.67 - samples/sec: 264.02 - lr: 0.000121 - momentum: 0.000000
2023-10-06 12:08:35,466 epoch 4 - iter 45/152 - loss 0.24279470 - time (sec): 33.20 - samples/sec: 268.71 - lr: 0.000119 - momentum: 0.000000
2023-10-06 12:08:46,611 epoch 4 - iter 60/152 - loss 0.24022027 - time (sec): 44.35 - samples/sec: 269.46 - lr: 0.000118 - momentum: 0.000000
2023-10-06 12:08:58,501 epoch 4 - iter 75/152 - loss 0.23491480 - time (sec): 56.24 - samples/sec: 273.32 - lr: 0.000116 - momentum: 0.000000
2023-10-06 12:09:10,362 epoch 4 - iter 90/152 - loss 0.23215777 - time (sec): 68.10 - samples/sec: 275.54 - lr: 0.000114 - momentum: 0.000000
2023-10-06 12:09:21,698 epoch 4 - iter 105/152 - loss 0.22764006 - time (sec): 79.43 - samples/sec: 275.02 - lr: 0.000112 - momentum: 0.000000
2023-10-06 12:09:32,393 epoch 4 - iter 120/152 - loss 0.22243235 - time (sec): 90.13 - samples/sec: 274.39 - lr: 0.000111 - momentum: 0.000000
2023-10-06 12:09:43,952 epoch 4 - iter 135/152 - loss 0.21993044 - time (sec): 101.69 - samples/sec: 271.75 - lr: 0.000109 - momentum: 0.000000
2023-10-06 12:09:54,757 epoch 4 - iter 150/152 - loss 0.21504401 - time (sec): 112.49 - samples/sec: 271.47 - lr: 0.000107 - momentum: 0.000000
2023-10-06 12:09:56,300 ----------------------------------------------------------------------------------------------------
2023-10-06 12:09:56,301 EPOCH 4 done: loss 0.2163 - lr: 0.000107
2023-10-06 12:10:04,248 DEV : loss 0.22751779854297638 - f1-score (micro avg) 0.687
2023-10-06 12:10:04,257 saving best model
2023-10-06 12:10:08,601 ----------------------------------------------------------------------------------------------------
2023-10-06 12:10:19,647 epoch 5 - iter 15/152 - loss 0.15951381 - time (sec): 11.04 - samples/sec: 269.83 - lr: 0.000105 - momentum: 0.000000
2023-10-06 12:10:30,188 epoch 5 - iter 30/152 - loss 0.15429107 - time (sec): 21.58 - samples/sec: 270.00 - lr: 0.000104 - momentum: 0.000000
2023-10-06 12:10:42,171 epoch 5 - iter 45/152 - loss 0.14455687 - time (sec): 33.57 - samples/sec: 276.63 - lr: 0.000102 - momentum: 0.000000
2023-10-06 12:10:53,584 epoch 5 - iter 60/152 - loss 0.14741911 - time (sec): 44.98 - samples/sec: 276.52 - lr: 0.000100 - momentum: 0.000000
2023-10-06 12:11:05,039 epoch 5 - iter 75/152 - loss 0.14554318 - time (sec): 56.44 - samples/sec: 276.72 - lr: 0.000098 - momentum: 0.000000
2023-10-06 12:11:16,575 epoch 5 - iter 90/152 - loss 0.14521097 - time (sec): 67.97 - samples/sec: 275.89 - lr: 0.000097 - momentum: 0.000000
2023-10-06 12:11:26,737 epoch 5 - iter 105/152 - loss 0.14238142 - time (sec): 78.13 - samples/sec: 273.16 - lr: 0.000095 - momentum: 0.000000
2023-10-06 12:11:37,443 epoch 5 - iter 120/152 - loss 0.14306461 - time (sec): 88.84 - samples/sec: 272.12 - lr: 0.000093 - momentum: 0.000000
2023-10-06 12:11:48,868 epoch 5 - iter 135/152 - loss 0.14286929 - time (sec): 100.26 - samples/sec: 273.02 - lr: 0.000091 - momentum: 0.000000
2023-10-06 12:12:00,148 epoch 5 - iter 150/152 - loss 0.14522024 - time (sec): 111.55 - samples/sec: 274.43 - lr: 0.000090 - momentum: 0.000000
2023-10-06 12:12:01,553 ----------------------------------------------------------------------------------------------------
2023-10-06 12:12:01,553 EPOCH 5 done: loss 0.1444 - lr: 0.000090
2023-10-06 12:12:09,579 DEV : loss 0.17563626170158386 - f1-score (micro avg) 0.7021
2023-10-06 12:12:09,587 saving best model
2023-10-06 12:12:13,967 ----------------------------------------------------------------------------------------------------
2023-10-06 12:12:24,723 epoch 6 - iter 15/152 - loss 0.12403351 - time (sec): 10.75 - samples/sec: 274.13 - lr: 0.000088 - momentum: 0.000000
2023-10-06 12:12:36,005 epoch 6 - iter 30/152 - loss 0.11700282 - time (sec): 22.04 - samples/sec: 272.00 - lr: 0.000086 - momentum: 0.000000
2023-10-06 12:12:46,760 epoch 6 - iter 45/152 - loss 0.11345796 - time (sec): 32.79 - samples/sec: 272.55 - lr: 0.000084 - momentum: 0.000000
2023-10-06 12:12:57,704 epoch 6 - iter 60/152 - loss 0.11484332 - time (sec): 43.74 - samples/sec: 272.92 - lr: 0.000082 - momentum: 0.000000
2023-10-06 12:13:08,688 epoch 6 - iter 75/152 - loss 0.10922840 - time (sec): 54.72 - samples/sec: 274.64 - lr: 0.000081 - momentum: 0.000000
2023-10-06 12:13:20,364 epoch 6 - iter 90/152 - loss 0.10891341 - time (sec): 66.40 - samples/sec: 276.42 - lr: 0.000079 - momentum: 0.000000
2023-10-06 12:13:31,121 epoch 6 - iter 105/152 - loss 0.11126689 - time (sec): 77.15 - samples/sec: 275.91 - lr: 0.000077 - momentum: 0.000000
2023-10-06 12:13:42,989 epoch 6 - iter 120/152 - loss 0.10581120 - time (sec): 89.02 - samples/sec: 276.97 - lr: 0.000075 - momentum: 0.000000
2023-10-06 12:13:54,120 epoch 6 - iter 135/152 - loss 0.10175369 - time (sec): 100.15 - samples/sec: 276.36 - lr: 0.000074 - momentum: 0.000000
2023-10-06 12:14:04,976 epoch 6 - iter 150/152 - loss 0.10151653 - time (sec): 111.01 - samples/sec: 276.63 - lr: 0.000072 - momentum: 0.000000
2023-10-06 12:14:06,092 ----------------------------------------------------------------------------------------------------
2023-10-06 12:14:06,093 EPOCH 6 done: loss 0.1013 - lr: 0.000072
2023-10-06 12:14:13,389 DEV : loss 0.1543136090040207 - f1-score (micro avg) 0.7738
2023-10-06 12:14:13,396 saving best model
2023-10-06 12:14:17,728 ----------------------------------------------------------------------------------------------------
2023-10-06 12:14:28,364 epoch 7 - iter 15/152 - loss 0.08930482 - time (sec): 10.63 - samples/sec: 290.31 - lr: 0.000070 - momentum: 0.000000
2023-10-06 12:14:38,909 epoch 7 - iter 30/152 - loss 0.06578091 - time (sec): 21.18 - samples/sec: 293.41 - lr: 0.000068 - momentum: 0.000000
2023-10-06 12:14:49,555 epoch 7 - iter 45/152 - loss 0.07308786 - time (sec): 31.83 - samples/sec: 295.61 - lr: 0.000066 - momentum: 0.000000
2023-10-06 12:15:00,223 epoch 7 - iter 60/152 - loss 0.07966302 - time (sec): 42.49 - samples/sec: 295.46 - lr: 0.000065 - momentum: 0.000000
2023-10-06 12:15:10,037 epoch 7 - iter 75/152 - loss 0.08268068 - time (sec): 52.31 - samples/sec: 293.80 - lr: 0.000063 - momentum: 0.000000
2023-10-06 12:15:19,996 epoch 7 - iter 90/152 - loss 0.08050613 - time (sec): 62.27 - samples/sec: 294.03 - lr: 0.000061 - momentum: 0.000000
2023-10-06 12:15:30,376 epoch 7 - iter 105/152 - loss 0.07816602 - time (sec): 72.65 - samples/sec: 294.44 - lr: 0.000059 - momentum: 0.000000
2023-10-06 12:15:40,470 epoch 7 - iter 120/152 - loss 0.07702418 - time (sec): 82.74 - samples/sec: 293.94 - lr: 0.000058 - momentum: 0.000000
2023-10-06 12:15:51,413 epoch 7 - iter 135/152 - loss 0.07564384 - time (sec): 93.68 - samples/sec: 294.17 - lr: 0.000056 - momentum: 0.000000
2023-10-06 12:16:01,727 epoch 7 - iter 150/152 - loss 0.07638745 - time (sec): 104.00 - samples/sec: 294.28 - lr: 0.000054 - momentum: 0.000000
2023-10-06 12:16:03,002 ----------------------------------------------------------------------------------------------------
2023-10-06 12:16:03,003 EPOCH 7 done: loss 0.0757 - lr: 0.000054
2023-10-06 12:16:10,199 DEV : loss 0.1410299688577652 - f1-score (micro avg) 0.8187
2023-10-06 12:16:10,208 saving best model
2023-10-06 12:16:14,545 ----------------------------------------------------------------------------------------------------
2023-10-06 12:16:24,980 epoch 8 - iter 15/152 - loss 0.06033134 - time (sec): 10.43 - samples/sec: 277.10 - lr: 0.000052 - momentum: 0.000000
2023-10-06 12:16:35,514 epoch 8 - iter 30/152 - loss 0.06192308 - time (sec): 20.97 - samples/sec: 287.31 - lr: 0.000050 - momentum: 0.000000
2023-10-06 12:16:45,833 epoch 8 - iter 45/152 - loss 0.05196584 - time (sec): 31.29 - samples/sec: 291.32 - lr: 0.000049 - momentum: 0.000000
2023-10-06 12:16:56,942 epoch 8 - iter 60/152 - loss 0.05954812 - time (sec): 42.40 - samples/sec: 293.57 - lr: 0.000047 - momentum: 0.000000
2023-10-06 12:17:07,250 epoch 8 - iter 75/152 - loss 0.06352304 - time (sec): 52.70 - samples/sec: 293.91 - lr: 0.000045 - momentum: 0.000000
2023-10-06 12:17:17,508 epoch 8 - iter 90/152 - loss 0.06273933 - time (sec): 62.96 - samples/sec: 291.37 - lr: 0.000043 - momentum: 0.000000
2023-10-06 12:17:28,305 epoch 8 - iter 105/152 - loss 0.06082397 - time (sec): 73.76 - samples/sec: 292.22 - lr: 0.000042 - momentum: 0.000000
2023-10-06 12:17:38,935 epoch 8 - iter 120/152 - loss 0.06101682 - time (sec): 84.39 - samples/sec: 291.45 - lr: 0.000040 - momentum: 0.000000
2023-10-06 12:17:49,247 epoch 8 - iter 135/152 - loss 0.06145730 - time (sec): 94.70 - samples/sec: 289.71 - lr: 0.000038 - momentum: 0.000000
2023-10-06 12:18:00,377 epoch 8 - iter 150/152 - loss 0.06003389 - time (sec): 105.83 - samples/sec: 289.70 - lr: 0.000036 - momentum: 0.000000
2023-10-06 12:18:01,540 ----------------------------------------------------------------------------------------------------
2023-10-06 12:18:01,540 EPOCH 8 done: loss 0.0599 - lr: 0.000036
2023-10-06 12:18:09,011 DEV : loss 0.14382629096508026 - f1-score (micro avg) 0.8183
2023-10-06 12:18:09,019 ----------------------------------------------------------------------------------------------------
2023-10-06 12:18:20,037 epoch 9 - iter 15/152 - loss 0.03421405 - time (sec): 11.02 - samples/sec: 278.41 - lr: 0.000034 - momentum: 0.000000
2023-10-06 12:18:30,568 epoch 9 - iter 30/152 - loss 0.04338140 - time (sec): 21.55 - samples/sec: 278.59 - lr: 0.000033 - momentum: 0.000000
2023-10-06 12:18:41,390 epoch 9 - iter 45/152 - loss 0.04255641 - time (sec): 32.37 - samples/sec: 275.36 - lr: 0.000031 - momentum: 0.000000
2023-10-06 12:18:52,804 epoch 9 - iter 60/152 - loss 0.04732788 - time (sec): 43.78 - samples/sec: 278.44 - lr: 0.000029 - momentum: 0.000000
2023-10-06 12:19:03,735 epoch 9 - iter 75/152 - loss 0.05095845 - time (sec): 54.71 - samples/sec: 278.79 - lr: 0.000027 - momentum: 0.000000
2023-10-06 12:19:15,387 epoch 9 - iter 90/152 - loss 0.05702162 - time (sec): 66.37 - samples/sec: 281.75 - lr: 0.000026 - momentum: 0.000000
2023-10-06 12:19:26,028 epoch 9 - iter 105/152 - loss 0.05426125 - time (sec): 77.01 - samples/sec: 279.62 - lr: 0.000024 - momentum: 0.000000
2023-10-06 12:19:37,420 epoch 9 - iter 120/152 - loss 0.05428196 - time (sec): 88.40 - samples/sec: 279.08 - lr: 0.000022 - momentum: 0.000000
2023-10-06 12:19:48,285 epoch 9 - iter 135/152 - loss 0.05402795 - time (sec): 99.26 - samples/sec: 277.60 - lr: 0.000020 - momentum: 0.000000
2023-10-06 12:19:59,301 epoch 9 - iter 150/152 - loss 0.05197514 - time (sec): 110.28 - samples/sec: 277.58 - lr: 0.000019 - momentum: 0.000000
2023-10-06 12:20:00,718 ----------------------------------------------------------------------------------------------------
2023-10-06 12:20:00,718 EPOCH 9 done: loss 0.0521 - lr: 0.000019
2023-10-06 12:20:08,506 DEV : loss 0.14206436276435852 - f1-score (micro avg) 0.8191
2023-10-06 12:20:08,515 saving best model
2023-10-06 12:20:12,858 ----------------------------------------------------------------------------------------------------
2023-10-06 12:20:23,682 epoch 10 - iter 15/152 - loss 0.06250421 - time (sec): 10.82 - samples/sec: 280.53 - lr: 0.000017 - momentum: 0.000000
2023-10-06 12:20:34,055 epoch 10 - iter 30/152 - loss 0.04309332 - time (sec): 21.19 - samples/sec: 281.25 - lr: 0.000015 - momentum: 0.000000
2023-10-06 12:20:45,084 epoch 10 - iter 45/152 - loss 0.03714810 - time (sec): 32.22 - samples/sec: 286.70 - lr: 0.000013 - momentum: 0.000000
2023-10-06 12:20:55,339 epoch 10 - iter 60/152 - loss 0.04248445 - time (sec): 42.48 - samples/sec: 289.39 - lr: 0.000012 - momentum: 0.000000
2023-10-06 12:21:06,465 epoch 10 - iter 75/152 - loss 0.04273133 - time (sec): 53.61 - samples/sec: 292.38 - lr: 0.000010 - momentum: 0.000000
2023-10-06 12:21:17,073 epoch 10 - iter 90/152 - loss 0.04245136 - time (sec): 64.21 - samples/sec: 292.79 - lr: 0.000008 - momentum: 0.000000
2023-10-06 12:21:27,178 epoch 10 - iter 105/152 - loss 0.04490524 - time (sec): 74.32 - samples/sec: 293.57 - lr: 0.000006 - momentum: 0.000000
2023-10-06 12:21:37,146 epoch 10 - iter 120/152 - loss 0.04663582 - time (sec): 84.29 - samples/sec: 292.79 - lr: 0.000005 - momentum: 0.000000
2023-10-06 12:21:47,292 epoch 10 - iter 135/152 - loss 0.04702965 - time (sec): 94.43 - samples/sec: 291.90 - lr: 0.000003 - momentum: 0.000000
2023-10-06 12:21:57,888 epoch 10 - iter 150/152 - loss 0.04626855 - time (sec): 105.03 - samples/sec: 291.32 - lr: 0.000001 - momentum: 0.000000
2023-10-06 12:21:59,175 ----------------------------------------------------------------------------------------------------
2023-10-06 12:21:59,175 EPOCH 10 done: loss 0.0459 - lr: 0.000001
2023-10-06 12:22:06,340 DEV : loss 0.14375203847885132 - f1-score (micro avg) 0.8205
2023-10-06 12:22:06,348 saving best model
2023-10-06 12:22:11,518 ----------------------------------------------------------------------------------------------------
2023-10-06 12:22:11,519 Loading model from best epoch ...
2023-10-06 12:22:14,117 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 12:22:20,882
Results:
- F-score (micro) 0.7929
- F-score (macro) 0.4838
- Accuracy 0.6659
By class:
precision recall f1-score support
scope 0.7329 0.7815 0.7564 151
pers 0.8070 0.9583 0.8762 96
work 0.7297 0.8526 0.7864 95
loc 0.0000 0.0000 0.0000 3
date 0.0000 0.0000 0.0000 3
micro avg 0.7539 0.8362 0.7929 348
macro avg 0.4539 0.5185 0.4838 348
weighted avg 0.7399 0.8362 0.7846 348
2023-10-06 12:22:20,882 ----------------------------------------------------------------------------------------------------
|