File size: 25,115 Bytes
72e7967 |
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 14:54:04,117 ----------------------------------------------------------------------------------------------------
2023-10-06 14:54:04,118 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 14:54:04,118 ----------------------------------------------------------------------------------------------------
2023-10-06 14:54:04,118 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 14:54:04,119 ----------------------------------------------------------------------------------------------------
2023-10-06 14:54:04,119 Train: 1214 sentences
2023-10-06 14:54:04,119 (train_with_dev=False, train_with_test=False)
2023-10-06 14:54:04,119 ----------------------------------------------------------------------------------------------------
2023-10-06 14:54:04,119 Training Params:
2023-10-06 14:54:04,119 - learning_rate: "0.00016"
2023-10-06 14:54:04,119 - mini_batch_size: "8"
2023-10-06 14:54:04,119 - max_epochs: "10"
2023-10-06 14:54:04,119 - shuffle: "True"
2023-10-06 14:54:04,119 ----------------------------------------------------------------------------------------------------
2023-10-06 14:54:04,119 Plugins:
2023-10-06 14:54:04,119 - TensorboardLogger
2023-10-06 14:54:04,119 - LinearScheduler | warmup_fraction: '0.1'
2023-10-06 14:54:04,119 ----------------------------------------------------------------------------------------------------
2023-10-06 14:54:04,119 Final evaluation on model from best epoch (best-model.pt)
2023-10-06 14:54:04,119 - metric: "('micro avg', 'f1-score')"
2023-10-06 14:54:04,119 ----------------------------------------------------------------------------------------------------
2023-10-06 14:54:04,120 Computation:
2023-10-06 14:54:04,120 - compute on device: cuda:0
2023-10-06 14:54:04,120 - embedding storage: none
2023-10-06 14:54:04,120 ----------------------------------------------------------------------------------------------------
2023-10-06 14:54:04,120 Model training base path: "hmbench-ajmc/en-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-4"
2023-10-06 14:54:04,120 ----------------------------------------------------------------------------------------------------
2023-10-06 14:54:04,120 ----------------------------------------------------------------------------------------------------
2023-10-06 14:54:04,120 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-06 14:54:15,045 epoch 1 - iter 15/152 - loss 3.24944544 - time (sec): 10.92 - samples/sec: 280.48 - lr: 0.000015 - momentum: 0.000000
2023-10-06 14:54:26,358 epoch 1 - iter 30/152 - loss 3.24250264 - time (sec): 22.24 - samples/sec: 277.56 - lr: 0.000031 - momentum: 0.000000
2023-10-06 14:54:37,201 epoch 1 - iter 45/152 - loss 3.23097880 - time (sec): 33.08 - samples/sec: 276.79 - lr: 0.000046 - momentum: 0.000000
2023-10-06 14:54:48,538 epoch 1 - iter 60/152 - loss 3.20599492 - time (sec): 44.42 - samples/sec: 275.76 - lr: 0.000062 - momentum: 0.000000
2023-10-06 14:55:00,091 epoch 1 - iter 75/152 - loss 3.15179797 - time (sec): 55.97 - samples/sec: 277.22 - lr: 0.000078 - momentum: 0.000000
2023-10-06 14:55:11,261 epoch 1 - iter 90/152 - loss 3.08113062 - time (sec): 67.14 - samples/sec: 276.63 - lr: 0.000094 - momentum: 0.000000
2023-10-06 14:55:22,033 epoch 1 - iter 105/152 - loss 2.99589247 - time (sec): 77.91 - samples/sec: 275.94 - lr: 0.000109 - momentum: 0.000000
2023-10-06 14:55:32,979 epoch 1 - iter 120/152 - loss 2.90320056 - time (sec): 88.86 - samples/sec: 274.66 - lr: 0.000125 - momentum: 0.000000
2023-10-06 14:55:44,611 epoch 1 - iter 135/152 - loss 2.79285826 - time (sec): 100.49 - samples/sec: 275.87 - lr: 0.000141 - momentum: 0.000000
2023-10-06 14:55:55,343 epoch 1 - iter 150/152 - loss 2.69377810 - time (sec): 111.22 - samples/sec: 274.81 - lr: 0.000157 - momentum: 0.000000
2023-10-06 14:55:56,812 ----------------------------------------------------------------------------------------------------
2023-10-06 14:55:56,813 EPOCH 1 done: loss 2.6808 - lr: 0.000157
2023-10-06 14:56:04,646 DEV : loss 1.5423763990402222 - f1-score (micro avg) 0.0
2023-10-06 14:56:04,653 ----------------------------------------------------------------------------------------------------
2023-10-06 14:56:16,114 epoch 2 - iter 15/152 - loss 1.45028936 - time (sec): 11.46 - samples/sec: 281.01 - lr: 0.000158 - momentum: 0.000000
2023-10-06 14:56:27,084 epoch 2 - iter 30/152 - loss 1.31419700 - time (sec): 22.43 - samples/sec: 277.85 - lr: 0.000157 - momentum: 0.000000
2023-10-06 14:56:38,327 epoch 2 - iter 45/152 - loss 1.22141358 - time (sec): 33.67 - samples/sec: 275.63 - lr: 0.000155 - momentum: 0.000000
2023-10-06 14:56:49,884 epoch 2 - iter 60/152 - loss 1.13843406 - time (sec): 45.23 - samples/sec: 277.78 - lr: 0.000153 - momentum: 0.000000
2023-10-06 14:57:00,600 epoch 2 - iter 75/152 - loss 1.08039112 - time (sec): 55.94 - samples/sec: 277.54 - lr: 0.000151 - momentum: 0.000000
2023-10-06 14:57:11,495 epoch 2 - iter 90/152 - loss 1.01221563 - time (sec): 66.84 - samples/sec: 277.48 - lr: 0.000150 - momentum: 0.000000
2023-10-06 14:57:22,606 epoch 2 - iter 105/152 - loss 0.97571095 - time (sec): 77.95 - samples/sec: 277.16 - lr: 0.000148 - momentum: 0.000000
2023-10-06 14:57:33,499 epoch 2 - iter 120/152 - loss 0.92119821 - time (sec): 88.84 - samples/sec: 276.72 - lr: 0.000146 - momentum: 0.000000
2023-10-06 14:57:44,980 epoch 2 - iter 135/152 - loss 0.87915528 - time (sec): 100.33 - samples/sec: 278.04 - lr: 0.000144 - momentum: 0.000000
2023-10-06 14:57:55,416 epoch 2 - iter 150/152 - loss 0.84045686 - time (sec): 110.76 - samples/sec: 277.49 - lr: 0.000143 - momentum: 0.000000
2023-10-06 14:57:56,460 ----------------------------------------------------------------------------------------------------
2023-10-06 14:57:56,460 EPOCH 2 done: loss 0.8375 - lr: 0.000143
2023-10-06 14:58:04,278 DEV : loss 0.5047337412834167 - f1-score (micro avg) 0.0142
2023-10-06 14:58:04,285 saving best model
2023-10-06 14:58:05,137 ----------------------------------------------------------------------------------------------------
2023-10-06 14:58:16,206 epoch 3 - iter 15/152 - loss 0.36830879 - time (sec): 11.07 - samples/sec: 269.53 - lr: 0.000141 - momentum: 0.000000
2023-10-06 14:58:27,597 epoch 3 - iter 30/152 - loss 0.37695983 - time (sec): 22.46 - samples/sec: 274.92 - lr: 0.000139 - momentum: 0.000000
2023-10-06 14:58:37,905 epoch 3 - iter 45/152 - loss 0.37352735 - time (sec): 32.77 - samples/sec: 271.68 - lr: 0.000137 - momentum: 0.000000
2023-10-06 14:58:48,591 epoch 3 - iter 60/152 - loss 0.37890694 - time (sec): 43.45 - samples/sec: 271.52 - lr: 0.000135 - momentum: 0.000000
2023-10-06 14:59:00,051 epoch 3 - iter 75/152 - loss 0.36339867 - time (sec): 54.91 - samples/sec: 275.66 - lr: 0.000134 - momentum: 0.000000
2023-10-06 14:59:10,655 epoch 3 - iter 90/152 - loss 0.35784250 - time (sec): 65.52 - samples/sec: 275.17 - lr: 0.000132 - momentum: 0.000000
2023-10-06 14:59:21,918 epoch 3 - iter 105/152 - loss 0.35112702 - time (sec): 76.78 - samples/sec: 277.68 - lr: 0.000130 - momentum: 0.000000
2023-10-06 14:59:33,043 epoch 3 - iter 120/152 - loss 0.34716540 - time (sec): 87.90 - samples/sec: 278.01 - lr: 0.000128 - momentum: 0.000000
2023-10-06 14:59:44,360 epoch 3 - iter 135/152 - loss 0.33725530 - time (sec): 99.22 - samples/sec: 278.72 - lr: 0.000127 - momentum: 0.000000
2023-10-06 14:59:55,038 epoch 3 - iter 150/152 - loss 0.33077678 - time (sec): 109.90 - samples/sec: 277.95 - lr: 0.000125 - momentum: 0.000000
2023-10-06 14:59:56,568 ----------------------------------------------------------------------------------------------------
2023-10-06 14:59:56,568 EPOCH 3 done: loss 0.3300 - lr: 0.000125
2023-10-06 15:00:04,361 DEV : loss 0.29971709847450256 - f1-score (micro avg) 0.5147
2023-10-06 15:00:04,367 saving best model
2023-10-06 15:00:08,657 ----------------------------------------------------------------------------------------------------
2023-10-06 15:00:19,398 epoch 4 - iter 15/152 - loss 0.22111745 - time (sec): 10.74 - samples/sec: 274.60 - lr: 0.000123 - momentum: 0.000000
2023-10-06 15:00:30,602 epoch 4 - iter 30/152 - loss 0.22991226 - time (sec): 21.94 - samples/sec: 279.98 - lr: 0.000121 - momentum: 0.000000
2023-10-06 15:00:41,086 epoch 4 - iter 45/152 - loss 0.22353854 - time (sec): 32.43 - samples/sec: 275.69 - lr: 0.000119 - momentum: 0.000000
2023-10-06 15:00:51,889 epoch 4 - iter 60/152 - loss 0.21205288 - time (sec): 43.23 - samples/sec: 275.40 - lr: 0.000118 - momentum: 0.000000
2023-10-06 15:01:02,655 epoch 4 - iter 75/152 - loss 0.20545775 - time (sec): 54.00 - samples/sec: 275.74 - lr: 0.000116 - momentum: 0.000000
2023-10-06 15:01:13,843 epoch 4 - iter 90/152 - loss 0.20843455 - time (sec): 65.19 - samples/sec: 276.49 - lr: 0.000114 - momentum: 0.000000
2023-10-06 15:01:25,156 epoch 4 - iter 105/152 - loss 0.20784356 - time (sec): 76.50 - samples/sec: 278.16 - lr: 0.000112 - momentum: 0.000000
2023-10-06 15:01:36,444 epoch 4 - iter 120/152 - loss 0.20651959 - time (sec): 87.79 - samples/sec: 277.86 - lr: 0.000111 - momentum: 0.000000
2023-10-06 15:01:47,962 epoch 4 - iter 135/152 - loss 0.20646267 - time (sec): 99.30 - samples/sec: 278.36 - lr: 0.000109 - momentum: 0.000000
2023-10-06 15:01:58,990 epoch 4 - iter 150/152 - loss 0.20044522 - time (sec): 110.33 - samples/sec: 278.64 - lr: 0.000107 - momentum: 0.000000
2023-10-06 15:02:00,046 ----------------------------------------------------------------------------------------------------
2023-10-06 15:02:00,046 EPOCH 4 done: loss 0.2003 - lr: 0.000107
2023-10-06 15:02:07,930 DEV : loss 0.2131974995136261 - f1-score (micro avg) 0.696
2023-10-06 15:02:07,937 saving best model
2023-10-06 15:02:12,292 ----------------------------------------------------------------------------------------------------
2023-10-06 15:02:23,063 epoch 5 - iter 15/152 - loss 0.13651155 - time (sec): 10.77 - samples/sec: 275.49 - lr: 0.000105 - momentum: 0.000000
2023-10-06 15:02:34,190 epoch 5 - iter 30/152 - loss 0.14246880 - time (sec): 21.90 - samples/sec: 285.98 - lr: 0.000104 - momentum: 0.000000
2023-10-06 15:02:45,061 epoch 5 - iter 45/152 - loss 0.14377517 - time (sec): 32.77 - samples/sec: 286.84 - lr: 0.000102 - momentum: 0.000000
2023-10-06 15:02:55,518 epoch 5 - iter 60/152 - loss 0.14293566 - time (sec): 43.23 - samples/sec: 285.25 - lr: 0.000100 - momentum: 0.000000
2023-10-06 15:03:05,926 epoch 5 - iter 75/152 - loss 0.13509826 - time (sec): 53.63 - samples/sec: 286.43 - lr: 0.000098 - momentum: 0.000000
2023-10-06 15:03:17,090 epoch 5 - iter 90/152 - loss 0.13560247 - time (sec): 64.80 - samples/sec: 289.58 - lr: 0.000097 - momentum: 0.000000
2023-10-06 15:03:27,733 epoch 5 - iter 105/152 - loss 0.14033493 - time (sec): 75.44 - samples/sec: 291.66 - lr: 0.000095 - momentum: 0.000000
2023-10-06 15:03:37,869 epoch 5 - iter 120/152 - loss 0.13801788 - time (sec): 85.58 - samples/sec: 290.50 - lr: 0.000093 - momentum: 0.000000
2023-10-06 15:03:48,244 epoch 5 - iter 135/152 - loss 0.13495779 - time (sec): 95.95 - samples/sec: 290.31 - lr: 0.000091 - momentum: 0.000000
2023-10-06 15:03:58,015 epoch 5 - iter 150/152 - loss 0.13282031 - time (sec): 105.72 - samples/sec: 289.40 - lr: 0.000090 - momentum: 0.000000
2023-10-06 15:03:59,322 ----------------------------------------------------------------------------------------------------
2023-10-06 15:03:59,323 EPOCH 5 done: loss 0.1328 - lr: 0.000090
2023-10-06 15:04:06,330 DEV : loss 0.16594909131526947 - f1-score (micro avg) 0.7552
2023-10-06 15:04:06,339 saving best model
2023-10-06 15:04:10,659 ----------------------------------------------------------------------------------------------------
2023-10-06 15:04:20,574 epoch 6 - iter 15/152 - loss 0.07286253 - time (sec): 9.91 - samples/sec: 297.17 - lr: 0.000088 - momentum: 0.000000
2023-10-06 15:04:30,596 epoch 6 - iter 30/152 - loss 0.08190653 - time (sec): 19.94 - samples/sec: 292.85 - lr: 0.000086 - momentum: 0.000000
2023-10-06 15:04:40,957 epoch 6 - iter 45/152 - loss 0.08201220 - time (sec): 30.30 - samples/sec: 288.51 - lr: 0.000084 - momentum: 0.000000
2023-10-06 15:04:51,641 epoch 6 - iter 60/152 - loss 0.08484603 - time (sec): 40.98 - samples/sec: 290.41 - lr: 0.000082 - momentum: 0.000000
2023-10-06 15:05:01,979 epoch 6 - iter 75/152 - loss 0.08338837 - time (sec): 51.32 - samples/sec: 293.50 - lr: 0.000081 - momentum: 0.000000
2023-10-06 15:05:12,517 epoch 6 - iter 90/152 - loss 0.08585472 - time (sec): 61.86 - samples/sec: 294.02 - lr: 0.000079 - momentum: 0.000000
2023-10-06 15:05:23,099 epoch 6 - iter 105/152 - loss 0.09058195 - time (sec): 72.44 - samples/sec: 293.12 - lr: 0.000077 - momentum: 0.000000
2023-10-06 15:05:33,403 epoch 6 - iter 120/152 - loss 0.09149141 - time (sec): 82.74 - samples/sec: 292.40 - lr: 0.000075 - momentum: 0.000000
2023-10-06 15:05:43,822 epoch 6 - iter 135/152 - loss 0.09044122 - time (sec): 93.16 - samples/sec: 292.03 - lr: 0.000074 - momentum: 0.000000
2023-10-06 15:05:54,785 epoch 6 - iter 150/152 - loss 0.08888543 - time (sec): 104.12 - samples/sec: 292.59 - lr: 0.000072 - momentum: 0.000000
2023-10-06 15:05:56,348 ----------------------------------------------------------------------------------------------------
2023-10-06 15:05:56,349 EPOCH 6 done: loss 0.0902 - lr: 0.000072
2023-10-06 15:06:03,654 DEV : loss 0.14003103971481323 - f1-score (micro avg) 0.8216
2023-10-06 15:06:03,661 saving best model
2023-10-06 15:06:07,966 ----------------------------------------------------------------------------------------------------
2023-10-06 15:06:19,336 epoch 7 - iter 15/152 - loss 0.09304549 - time (sec): 11.37 - samples/sec: 304.52 - lr: 0.000070 - momentum: 0.000000
2023-10-06 15:06:29,974 epoch 7 - iter 30/152 - loss 0.07562276 - time (sec): 22.01 - samples/sec: 292.14 - lr: 0.000068 - momentum: 0.000000
2023-10-06 15:06:40,446 epoch 7 - iter 45/152 - loss 0.07334318 - time (sec): 32.48 - samples/sec: 291.76 - lr: 0.000066 - momentum: 0.000000
2023-10-06 15:06:50,617 epoch 7 - iter 60/152 - loss 0.07407120 - time (sec): 42.65 - samples/sec: 284.56 - lr: 0.000065 - momentum: 0.000000
2023-10-06 15:07:01,846 epoch 7 - iter 75/152 - loss 0.07204809 - time (sec): 53.88 - samples/sec: 284.47 - lr: 0.000063 - momentum: 0.000000
2023-10-06 15:07:12,504 epoch 7 - iter 90/152 - loss 0.07172554 - time (sec): 64.54 - samples/sec: 282.82 - lr: 0.000061 - momentum: 0.000000
2023-10-06 15:07:23,447 epoch 7 - iter 105/152 - loss 0.06797482 - time (sec): 75.48 - samples/sec: 282.22 - lr: 0.000059 - momentum: 0.000000
2023-10-06 15:07:34,152 epoch 7 - iter 120/152 - loss 0.06978457 - time (sec): 86.18 - samples/sec: 281.50 - lr: 0.000058 - momentum: 0.000000
2023-10-06 15:07:45,419 epoch 7 - iter 135/152 - loss 0.06738419 - time (sec): 97.45 - samples/sec: 282.09 - lr: 0.000056 - momentum: 0.000000
2023-10-06 15:07:56,784 epoch 7 - iter 150/152 - loss 0.06940149 - time (sec): 108.82 - samples/sec: 281.76 - lr: 0.000054 - momentum: 0.000000
2023-10-06 15:07:58,063 ----------------------------------------------------------------------------------------------------
2023-10-06 15:07:58,063 EPOCH 7 done: loss 0.0689 - lr: 0.000054
2023-10-06 15:08:06,065 DEV : loss 0.14181111752986908 - f1-score (micro avg) 0.8228
2023-10-06 15:08:06,073 saving best model
2023-10-06 15:08:10,369 ----------------------------------------------------------------------------------------------------
2023-10-06 15:08:21,506 epoch 8 - iter 15/152 - loss 0.05064222 - time (sec): 11.14 - samples/sec: 274.15 - lr: 0.000052 - momentum: 0.000000
2023-10-06 15:08:32,890 epoch 8 - iter 30/152 - loss 0.07198941 - time (sec): 22.52 - samples/sec: 276.91 - lr: 0.000050 - momentum: 0.000000
2023-10-06 15:08:43,836 epoch 8 - iter 45/152 - loss 0.06404128 - time (sec): 33.47 - samples/sec: 277.15 - lr: 0.000049 - momentum: 0.000000
2023-10-06 15:08:55,623 epoch 8 - iter 60/152 - loss 0.05734585 - time (sec): 45.25 - samples/sec: 279.41 - lr: 0.000047 - momentum: 0.000000
2023-10-06 15:09:06,691 epoch 8 - iter 75/152 - loss 0.06032725 - time (sec): 56.32 - samples/sec: 278.33 - lr: 0.000045 - momentum: 0.000000
2023-10-06 15:09:17,087 epoch 8 - iter 90/152 - loss 0.05864651 - time (sec): 66.72 - samples/sec: 276.18 - lr: 0.000043 - momentum: 0.000000
2023-10-06 15:09:27,963 epoch 8 - iter 105/152 - loss 0.05785205 - time (sec): 77.59 - samples/sec: 275.58 - lr: 0.000042 - momentum: 0.000000
2023-10-06 15:09:38,769 epoch 8 - iter 120/152 - loss 0.05653871 - time (sec): 88.40 - samples/sec: 275.04 - lr: 0.000040 - momentum: 0.000000
2023-10-06 15:09:50,187 epoch 8 - iter 135/152 - loss 0.05700357 - time (sec): 99.82 - samples/sec: 274.75 - lr: 0.000038 - momentum: 0.000000
2023-10-06 15:10:01,519 epoch 8 - iter 150/152 - loss 0.05517782 - time (sec): 111.15 - samples/sec: 275.22 - lr: 0.000036 - momentum: 0.000000
2023-10-06 15:10:02,946 ----------------------------------------------------------------------------------------------------
2023-10-06 15:10:02,947 EPOCH 8 done: loss 0.0546 - lr: 0.000036
2023-10-06 15:10:10,834 DEV : loss 0.13887561857700348 - f1-score (micro avg) 0.8339
2023-10-06 15:10:10,841 saving best model
2023-10-06 15:10:15,124 ----------------------------------------------------------------------------------------------------
2023-10-06 15:10:26,833 epoch 9 - iter 15/152 - loss 0.03018573 - time (sec): 11.71 - samples/sec: 283.47 - lr: 0.000034 - momentum: 0.000000
2023-10-06 15:10:38,375 epoch 9 - iter 30/152 - loss 0.03319836 - time (sec): 23.25 - samples/sec: 280.52 - lr: 0.000033 - momentum: 0.000000
2023-10-06 15:10:48,902 epoch 9 - iter 45/152 - loss 0.03938238 - time (sec): 33.78 - samples/sec: 275.75 - lr: 0.000031 - momentum: 0.000000
2023-10-06 15:11:00,133 epoch 9 - iter 60/152 - loss 0.03910265 - time (sec): 45.01 - samples/sec: 277.55 - lr: 0.000029 - momentum: 0.000000
2023-10-06 15:11:11,452 epoch 9 - iter 75/152 - loss 0.04443264 - time (sec): 56.33 - samples/sec: 278.50 - lr: 0.000027 - momentum: 0.000000
2023-10-06 15:11:22,466 epoch 9 - iter 90/152 - loss 0.04504801 - time (sec): 67.34 - samples/sec: 278.51 - lr: 0.000026 - momentum: 0.000000
2023-10-06 15:11:33,644 epoch 9 - iter 105/152 - loss 0.04819797 - time (sec): 78.52 - samples/sec: 279.31 - lr: 0.000024 - momentum: 0.000000
2023-10-06 15:11:44,142 epoch 9 - iter 120/152 - loss 0.04718031 - time (sec): 89.02 - samples/sec: 277.53 - lr: 0.000022 - momentum: 0.000000
2023-10-06 15:11:55,011 epoch 9 - iter 135/152 - loss 0.04711017 - time (sec): 99.89 - samples/sec: 277.60 - lr: 0.000020 - momentum: 0.000000
2023-10-06 15:12:05,499 epoch 9 - iter 150/152 - loss 0.04672030 - time (sec): 110.37 - samples/sec: 276.95 - lr: 0.000019 - momentum: 0.000000
2023-10-06 15:12:06,962 ----------------------------------------------------------------------------------------------------
2023-10-06 15:12:06,962 EPOCH 9 done: loss 0.0464 - lr: 0.000019
2023-10-06 15:12:14,653 DEV : loss 0.137730672955513 - f1-score (micro avg) 0.83
2023-10-06 15:12:14,659 ----------------------------------------------------------------------------------------------------
2023-10-06 15:12:25,326 epoch 10 - iter 15/152 - loss 0.04516188 - time (sec): 10.67 - samples/sec: 280.54 - lr: 0.000017 - momentum: 0.000000
2023-10-06 15:12:36,332 epoch 10 - iter 30/152 - loss 0.04842008 - time (sec): 21.67 - samples/sec: 278.94 - lr: 0.000015 - momentum: 0.000000
2023-10-06 15:12:46,987 epoch 10 - iter 45/152 - loss 0.04303074 - time (sec): 32.33 - samples/sec: 280.51 - lr: 0.000013 - momentum: 0.000000
2023-10-06 15:12:57,904 epoch 10 - iter 60/152 - loss 0.04090721 - time (sec): 43.24 - samples/sec: 277.69 - lr: 0.000012 - momentum: 0.000000
2023-10-06 15:13:08,514 epoch 10 - iter 75/152 - loss 0.03829420 - time (sec): 53.85 - samples/sec: 273.85 - lr: 0.000010 - momentum: 0.000000
2023-10-06 15:13:19,914 epoch 10 - iter 90/152 - loss 0.04368647 - time (sec): 65.25 - samples/sec: 274.54 - lr: 0.000008 - momentum: 0.000000
2023-10-06 15:13:30,800 epoch 10 - iter 105/152 - loss 0.04082935 - time (sec): 76.14 - samples/sec: 274.42 - lr: 0.000006 - momentum: 0.000000
2023-10-06 15:13:42,223 epoch 10 - iter 120/152 - loss 0.04090036 - time (sec): 87.56 - samples/sec: 275.54 - lr: 0.000005 - momentum: 0.000000
2023-10-06 15:13:53,578 epoch 10 - iter 135/152 - loss 0.04032698 - time (sec): 98.92 - samples/sec: 277.23 - lr: 0.000003 - momentum: 0.000000
2023-10-06 15:14:04,848 epoch 10 - iter 150/152 - loss 0.04052295 - time (sec): 110.19 - samples/sec: 278.04 - lr: 0.000001 - momentum: 0.000000
2023-10-06 15:14:06,117 ----------------------------------------------------------------------------------------------------
2023-10-06 15:14:06,118 EPOCH 10 done: loss 0.0418 - lr: 0.000001
2023-10-06 15:14:13,851 DEV : loss 0.13857229053974152 - f1-score (micro avg) 0.8296
2023-10-06 15:14:14,654 ----------------------------------------------------------------------------------------------------
2023-10-06 15:14:14,655 Loading model from best epoch ...
2023-10-06 15:14:18,184 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 15:14:25,471
Results:
- F-score (micro) 0.7984
- F-score (macro) 0.4845
- Accuracy 0.6751
By class:
precision recall f1-score support
scope 0.7640 0.8146 0.7885 151
pers 0.7826 0.9375 0.8531 96
work 0.7130 0.8632 0.7810 95
loc 0.0000 0.0000 0.0000 3
date 0.0000 0.0000 0.0000 3
micro avg 0.7545 0.8477 0.7984 348
macro avg 0.4519 0.5230 0.4845 348
weighted avg 0.7420 0.8477 0.7906 348
2023-10-06 15:14:25,472 ----------------------------------------------------------------------------------------------------
|