File size: 25,083 Bytes
b204486 |
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 |
2023-10-07 01:15:22,530 ----------------------------------------------------------------------------------------------------
2023-10-07 01:15:22,532 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()
)"
2023-10-07 01:15:22,532 ----------------------------------------------------------------------------------------------------
2023-10-07 01:15:22,532 MultiCorpus: 1100 train + 206 dev + 240 test sentences
- NER_HIPE_2022 Corpus: 1100 train + 206 dev + 240 test sentences - /app/.flair/datasets/ner_hipe_2022/v2.1/ajmc/de/with_doc_seperator
2023-10-07 01:15:22,532 ----------------------------------------------------------------------------------------------------
2023-10-07 01:15:22,532 Train: 1100 sentences
2023-10-07 01:15:22,532 (train_with_dev=False, train_with_test=False)
2023-10-07 01:15:22,532 ----------------------------------------------------------------------------------------------------
2023-10-07 01:15:22,532 Training Params:
2023-10-07 01:15:22,532 - learning_rate: "0.00015"
2023-10-07 01:15:22,532 - mini_batch_size: "8"
2023-10-07 01:15:22,532 - max_epochs: "10"
2023-10-07 01:15:22,532 - shuffle: "True"
2023-10-07 01:15:22,532 ----------------------------------------------------------------------------------------------------
2023-10-07 01:15:22,532 Plugins:
2023-10-07 01:15:22,532 - TensorboardLogger
2023-10-07 01:15:22,532 - LinearScheduler | warmup_fraction: '0.1'
2023-10-07 01:15:22,532 ----------------------------------------------------------------------------------------------------
2023-10-07 01:15:22,533 Final evaluation on model from best epoch (best-model.pt)
2023-10-07 01:15:22,533 - metric: "('micro avg', 'f1-score')"
2023-10-07 01:15:22,533 ----------------------------------------------------------------------------------------------------
2023-10-07 01:15:22,533 Computation:
2023-10-07 01:15:22,533 - compute on device: cuda:0
2023-10-07 01:15:22,533 - embedding storage: none
2023-10-07 01:15:22,533 ----------------------------------------------------------------------------------------------------
2023-10-07 01:15:22,533 Model training base path: "hmbench-ajmc/de-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-4"
2023-10-07 01:15:22,533 ----------------------------------------------------------------------------------------------------
2023-10-07 01:15:22,533 ----------------------------------------------------------------------------------------------------
2023-10-07 01:15:22,533 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-07 01:15:31,538 epoch 1 - iter 13/138 - loss 3.24828859 - time (sec): 9.00 - samples/sec: 224.23 - lr: 0.000013 - momentum: 0.000000
2023-10-07 01:15:40,922 epoch 1 - iter 26/138 - loss 3.24306645 - time (sec): 18.39 - samples/sec: 229.50 - lr: 0.000027 - momentum: 0.000000
2023-10-07 01:15:50,749 epoch 1 - iter 39/138 - loss 3.23291278 - time (sec): 28.22 - samples/sec: 232.92 - lr: 0.000041 - momentum: 0.000000
2023-10-07 01:16:00,629 epoch 1 - iter 52/138 - loss 3.21835524 - time (sec): 38.09 - samples/sec: 232.53 - lr: 0.000055 - momentum: 0.000000
2023-10-07 01:16:09,702 epoch 1 - iter 65/138 - loss 3.19377640 - time (sec): 47.17 - samples/sec: 230.75 - lr: 0.000070 - momentum: 0.000000
2023-10-07 01:16:19,190 epoch 1 - iter 78/138 - loss 3.14821775 - time (sec): 56.66 - samples/sec: 228.98 - lr: 0.000084 - momentum: 0.000000
2023-10-07 01:16:28,019 epoch 1 - iter 91/138 - loss 3.08866051 - time (sec): 65.48 - samples/sec: 228.63 - lr: 0.000098 - momentum: 0.000000
2023-10-07 01:16:38,138 epoch 1 - iter 104/138 - loss 3.00670410 - time (sec): 75.60 - samples/sec: 229.38 - lr: 0.000112 - momentum: 0.000000
2023-10-07 01:16:47,382 epoch 1 - iter 117/138 - loss 2.93035609 - time (sec): 84.85 - samples/sec: 229.73 - lr: 0.000126 - momentum: 0.000000
2023-10-07 01:16:56,570 epoch 1 - iter 130/138 - loss 2.85220031 - time (sec): 94.04 - samples/sec: 228.72 - lr: 0.000140 - momentum: 0.000000
2023-10-07 01:17:02,058 ----------------------------------------------------------------------------------------------------
2023-10-07 01:17:02,059 EPOCH 1 done: loss 2.8027 - lr: 0.000140
2023-10-07 01:17:08,371 DEV : loss 1.8371483087539673 - f1-score (micro avg) 0.0
2023-10-07 01:17:08,376 ----------------------------------------------------------------------------------------------------
2023-10-07 01:17:17,992 epoch 2 - iter 13/138 - loss 1.78002652 - time (sec): 9.61 - samples/sec: 232.76 - lr: 0.000149 - momentum: 0.000000
2023-10-07 01:17:27,209 epoch 2 - iter 26/138 - loss 1.70494439 - time (sec): 18.83 - samples/sec: 224.88 - lr: 0.000147 - momentum: 0.000000
2023-10-07 01:17:35,792 epoch 2 - iter 39/138 - loss 1.60987014 - time (sec): 27.42 - samples/sec: 221.41 - lr: 0.000145 - momentum: 0.000000
2023-10-07 01:17:45,114 epoch 2 - iter 52/138 - loss 1.51958579 - time (sec): 36.74 - samples/sec: 224.76 - lr: 0.000144 - momentum: 0.000000
2023-10-07 01:17:54,597 epoch 2 - iter 65/138 - loss 1.42527343 - time (sec): 46.22 - samples/sec: 225.77 - lr: 0.000142 - momentum: 0.000000
2023-10-07 01:18:04,025 epoch 2 - iter 78/138 - loss 1.35043793 - time (sec): 55.65 - samples/sec: 225.44 - lr: 0.000141 - momentum: 0.000000
2023-10-07 01:18:13,985 epoch 2 - iter 91/138 - loss 1.29544695 - time (sec): 65.61 - samples/sec: 226.39 - lr: 0.000139 - momentum: 0.000000
2023-10-07 01:18:23,880 epoch 2 - iter 104/138 - loss 1.23545553 - time (sec): 75.50 - samples/sec: 226.96 - lr: 0.000138 - momentum: 0.000000
2023-10-07 01:18:32,973 epoch 2 - iter 117/138 - loss 1.21725016 - time (sec): 84.60 - samples/sec: 228.38 - lr: 0.000136 - momentum: 0.000000
2023-10-07 01:18:42,496 epoch 2 - iter 130/138 - loss 1.17433177 - time (sec): 94.12 - samples/sec: 228.75 - lr: 0.000134 - momentum: 0.000000
2023-10-07 01:18:47,903 ----------------------------------------------------------------------------------------------------
2023-10-07 01:18:47,903 EPOCH 2 done: loss 1.1550 - lr: 0.000134
2023-10-07 01:18:54,337 DEV : loss 0.845672607421875 - f1-score (micro avg) 0.0
2023-10-07 01:18:54,342 ----------------------------------------------------------------------------------------------------
2023-10-07 01:19:03,480 epoch 3 - iter 13/138 - loss 0.76719476 - time (sec): 9.14 - samples/sec: 225.45 - lr: 0.000132 - momentum: 0.000000
2023-10-07 01:19:12,704 epoch 3 - iter 26/138 - loss 0.75545689 - time (sec): 18.36 - samples/sec: 229.73 - lr: 0.000130 - momentum: 0.000000
2023-10-07 01:19:22,852 epoch 3 - iter 39/138 - loss 0.69073662 - time (sec): 28.51 - samples/sec: 230.21 - lr: 0.000129 - momentum: 0.000000
2023-10-07 01:19:32,459 epoch 3 - iter 52/138 - loss 0.64569560 - time (sec): 38.12 - samples/sec: 229.11 - lr: 0.000127 - momentum: 0.000000
2023-10-07 01:19:41,993 epoch 3 - iter 65/138 - loss 0.63526192 - time (sec): 47.65 - samples/sec: 229.30 - lr: 0.000126 - momentum: 0.000000
2023-10-07 01:19:51,185 epoch 3 - iter 78/138 - loss 0.61936500 - time (sec): 56.84 - samples/sec: 228.12 - lr: 0.000124 - momentum: 0.000000
2023-10-07 01:20:00,681 epoch 3 - iter 91/138 - loss 0.59743899 - time (sec): 66.34 - samples/sec: 229.93 - lr: 0.000123 - momentum: 0.000000
2023-10-07 01:20:10,114 epoch 3 - iter 104/138 - loss 0.58003767 - time (sec): 75.77 - samples/sec: 229.74 - lr: 0.000121 - momentum: 0.000000
2023-10-07 01:20:19,561 epoch 3 - iter 117/138 - loss 0.56742023 - time (sec): 85.22 - samples/sec: 229.06 - lr: 0.000119 - momentum: 0.000000
2023-10-07 01:20:28,595 epoch 3 - iter 130/138 - loss 0.55633486 - time (sec): 94.25 - samples/sec: 228.21 - lr: 0.000118 - momentum: 0.000000
2023-10-07 01:20:34,112 ----------------------------------------------------------------------------------------------------
2023-10-07 01:20:34,113 EPOCH 3 done: loss 0.5499 - lr: 0.000118
2023-10-07 01:20:40,531 DEV : loss 0.4134528338909149 - f1-score (micro avg) 0.5185
2023-10-07 01:20:40,537 saving best model
2023-10-07 01:20:41,383 ----------------------------------------------------------------------------------------------------
2023-10-07 01:20:49,674 epoch 4 - iter 13/138 - loss 0.41651853 - time (sec): 8.29 - samples/sec: 206.66 - lr: 0.000115 - momentum: 0.000000
2023-10-07 01:20:59,027 epoch 4 - iter 26/138 - loss 0.37358104 - time (sec): 17.64 - samples/sec: 219.36 - lr: 0.000114 - momentum: 0.000000
2023-10-07 01:21:08,798 epoch 4 - iter 39/138 - loss 0.37215514 - time (sec): 27.41 - samples/sec: 225.62 - lr: 0.000112 - momentum: 0.000000
2023-10-07 01:21:18,418 epoch 4 - iter 52/138 - loss 0.37238785 - time (sec): 37.03 - samples/sec: 229.25 - lr: 0.000111 - momentum: 0.000000
2023-10-07 01:21:27,308 epoch 4 - iter 65/138 - loss 0.37104112 - time (sec): 45.92 - samples/sec: 228.16 - lr: 0.000109 - momentum: 0.000000
2023-10-07 01:21:36,664 epoch 4 - iter 78/138 - loss 0.35234194 - time (sec): 55.28 - samples/sec: 228.37 - lr: 0.000107 - momentum: 0.000000
2023-10-07 01:21:45,175 epoch 4 - iter 91/138 - loss 0.34394455 - time (sec): 63.79 - samples/sec: 226.31 - lr: 0.000106 - momentum: 0.000000
2023-10-07 01:21:55,033 epoch 4 - iter 104/138 - loss 0.33542247 - time (sec): 73.65 - samples/sec: 227.54 - lr: 0.000104 - momentum: 0.000000
2023-10-07 01:22:04,919 epoch 4 - iter 117/138 - loss 0.33674126 - time (sec): 83.53 - samples/sec: 228.41 - lr: 0.000103 - momentum: 0.000000
2023-10-07 01:22:15,458 epoch 4 - iter 130/138 - loss 0.32805211 - time (sec): 94.07 - samples/sec: 229.30 - lr: 0.000101 - momentum: 0.000000
2023-10-07 01:22:21,001 ----------------------------------------------------------------------------------------------------
2023-10-07 01:22:21,001 EPOCH 4 done: loss 0.3246 - lr: 0.000101
2023-10-07 01:22:27,423 DEV : loss 0.2694088816642761 - f1-score (micro avg) 0.7589
2023-10-07 01:22:27,428 saving best model
2023-10-07 01:22:28,303 ----------------------------------------------------------------------------------------------------
2023-10-07 01:22:37,145 epoch 5 - iter 13/138 - loss 0.30090111 - time (sec): 8.84 - samples/sec: 237.87 - lr: 0.000099 - momentum: 0.000000
2023-10-07 01:22:46,256 epoch 5 - iter 26/138 - loss 0.26577049 - time (sec): 17.95 - samples/sec: 226.73 - lr: 0.000097 - momentum: 0.000000
2023-10-07 01:22:56,278 epoch 5 - iter 39/138 - loss 0.22896971 - time (sec): 27.97 - samples/sec: 229.18 - lr: 0.000096 - momentum: 0.000000
2023-10-07 01:23:05,442 epoch 5 - iter 52/138 - loss 0.21383942 - time (sec): 37.14 - samples/sec: 228.64 - lr: 0.000094 - momentum: 0.000000
2023-10-07 01:23:14,990 epoch 5 - iter 65/138 - loss 0.21585880 - time (sec): 46.69 - samples/sec: 229.79 - lr: 0.000092 - momentum: 0.000000
2023-10-07 01:23:24,460 epoch 5 - iter 78/138 - loss 0.21291348 - time (sec): 56.16 - samples/sec: 229.15 - lr: 0.000091 - momentum: 0.000000
2023-10-07 01:23:33,904 epoch 5 - iter 91/138 - loss 0.21386054 - time (sec): 65.60 - samples/sec: 229.06 - lr: 0.000089 - momentum: 0.000000
2023-10-07 01:23:43,519 epoch 5 - iter 104/138 - loss 0.21509638 - time (sec): 75.21 - samples/sec: 229.56 - lr: 0.000088 - momentum: 0.000000
2023-10-07 01:23:53,245 epoch 5 - iter 117/138 - loss 0.21472409 - time (sec): 84.94 - samples/sec: 230.45 - lr: 0.000086 - momentum: 0.000000
2023-10-07 01:24:02,719 epoch 5 - iter 130/138 - loss 0.21814584 - time (sec): 94.41 - samples/sec: 230.02 - lr: 0.000085 - momentum: 0.000000
2023-10-07 01:24:07,647 ----------------------------------------------------------------------------------------------------
2023-10-07 01:24:07,647 EPOCH 5 done: loss 0.2169 - lr: 0.000085
2023-10-07 01:24:14,043 DEV : loss 0.19746224582195282 - f1-score (micro avg) 0.7865
2023-10-07 01:24:14,048 saving best model
2023-10-07 01:24:14,906 ----------------------------------------------------------------------------------------------------
2023-10-07 01:24:24,467 epoch 6 - iter 13/138 - loss 0.15336532 - time (sec): 9.56 - samples/sec: 241.22 - lr: 0.000082 - momentum: 0.000000
2023-10-07 01:24:34,331 epoch 6 - iter 26/138 - loss 0.16846546 - time (sec): 19.42 - samples/sec: 236.46 - lr: 0.000080 - momentum: 0.000000
2023-10-07 01:24:43,301 epoch 6 - iter 39/138 - loss 0.17300366 - time (sec): 28.39 - samples/sec: 234.63 - lr: 0.000079 - momentum: 0.000000
2023-10-07 01:24:52,559 epoch 6 - iter 52/138 - loss 0.16181288 - time (sec): 37.65 - samples/sec: 230.64 - lr: 0.000077 - momentum: 0.000000
2023-10-07 01:25:01,549 epoch 6 - iter 65/138 - loss 0.16255739 - time (sec): 46.64 - samples/sec: 229.11 - lr: 0.000076 - momentum: 0.000000
2023-10-07 01:25:11,148 epoch 6 - iter 78/138 - loss 0.16099642 - time (sec): 56.24 - samples/sec: 229.16 - lr: 0.000074 - momentum: 0.000000
2023-10-07 01:25:20,519 epoch 6 - iter 91/138 - loss 0.16207104 - time (sec): 65.61 - samples/sec: 228.56 - lr: 0.000073 - momentum: 0.000000
2023-10-07 01:25:30,256 epoch 6 - iter 104/138 - loss 0.15770321 - time (sec): 75.35 - samples/sec: 229.04 - lr: 0.000071 - momentum: 0.000000
2023-10-07 01:25:39,585 epoch 6 - iter 117/138 - loss 0.15414663 - time (sec): 84.68 - samples/sec: 228.34 - lr: 0.000070 - momentum: 0.000000
2023-10-07 01:25:48,985 epoch 6 - iter 130/138 - loss 0.15062209 - time (sec): 94.08 - samples/sec: 227.10 - lr: 0.000068 - momentum: 0.000000
2023-10-07 01:25:55,024 ----------------------------------------------------------------------------------------------------
2023-10-07 01:25:55,025 EPOCH 6 done: loss 0.1525 - lr: 0.000068
2023-10-07 01:26:01,693 DEV : loss 0.15915270149707794 - f1-score (micro avg) 0.8373
2023-10-07 01:26:01,698 saving best model
2023-10-07 01:26:02,575 ----------------------------------------------------------------------------------------------------
2023-10-07 01:26:12,347 epoch 7 - iter 13/138 - loss 0.13219699 - time (sec): 9.77 - samples/sec: 227.53 - lr: 0.000065 - momentum: 0.000000
2023-10-07 01:26:21,486 epoch 7 - iter 26/138 - loss 0.13025357 - time (sec): 18.91 - samples/sec: 227.82 - lr: 0.000064 - momentum: 0.000000
2023-10-07 01:26:30,405 epoch 7 - iter 39/138 - loss 0.13191905 - time (sec): 27.83 - samples/sec: 222.90 - lr: 0.000062 - momentum: 0.000000
2023-10-07 01:26:40,469 epoch 7 - iter 52/138 - loss 0.12075848 - time (sec): 37.89 - samples/sec: 221.21 - lr: 0.000061 - momentum: 0.000000
2023-10-07 01:26:49,595 epoch 7 - iter 65/138 - loss 0.11734169 - time (sec): 47.02 - samples/sec: 221.11 - lr: 0.000059 - momentum: 0.000000
2023-10-07 01:26:58,753 epoch 7 - iter 78/138 - loss 0.11326191 - time (sec): 56.18 - samples/sec: 220.45 - lr: 0.000058 - momentum: 0.000000
2023-10-07 01:27:09,090 epoch 7 - iter 91/138 - loss 0.11408908 - time (sec): 66.51 - samples/sec: 222.26 - lr: 0.000056 - momentum: 0.000000
2023-10-07 01:27:18,707 epoch 7 - iter 104/138 - loss 0.11252134 - time (sec): 76.13 - samples/sec: 223.84 - lr: 0.000054 - momentum: 0.000000
2023-10-07 01:27:28,634 epoch 7 - iter 117/138 - loss 0.11428033 - time (sec): 86.06 - samples/sec: 224.33 - lr: 0.000053 - momentum: 0.000000
2023-10-07 01:27:38,838 epoch 7 - iter 130/138 - loss 0.11554929 - time (sec): 96.26 - samples/sec: 224.36 - lr: 0.000051 - momentum: 0.000000
2023-10-07 01:27:44,390 ----------------------------------------------------------------------------------------------------
2023-10-07 01:27:44,390 EPOCH 7 done: loss 0.1145 - lr: 0.000051
2023-10-07 01:27:51,074 DEV : loss 0.14204637706279755 - f1-score (micro avg) 0.8422
2023-10-07 01:27:51,081 saving best model
2023-10-07 01:27:51,965 ----------------------------------------------------------------------------------------------------
2023-10-07 01:28:01,706 epoch 8 - iter 13/138 - loss 0.07932656 - time (sec): 9.74 - samples/sec: 227.12 - lr: 0.000049 - momentum: 0.000000
2023-10-07 01:28:11,678 epoch 8 - iter 26/138 - loss 0.07910965 - time (sec): 19.71 - samples/sec: 227.39 - lr: 0.000047 - momentum: 0.000000
2023-10-07 01:28:20,730 epoch 8 - iter 39/138 - loss 0.08042982 - time (sec): 28.76 - samples/sec: 224.98 - lr: 0.000046 - momentum: 0.000000
2023-10-07 01:28:30,382 epoch 8 - iter 52/138 - loss 0.08587795 - time (sec): 38.41 - samples/sec: 225.30 - lr: 0.000044 - momentum: 0.000000
2023-10-07 01:28:39,368 epoch 8 - iter 65/138 - loss 0.08723165 - time (sec): 47.40 - samples/sec: 224.41 - lr: 0.000043 - momentum: 0.000000
2023-10-07 01:28:48,684 epoch 8 - iter 78/138 - loss 0.08915768 - time (sec): 56.72 - samples/sec: 223.74 - lr: 0.000041 - momentum: 0.000000
2023-10-07 01:28:58,262 epoch 8 - iter 91/138 - loss 0.09084241 - time (sec): 66.30 - samples/sec: 223.35 - lr: 0.000039 - momentum: 0.000000
2023-10-07 01:29:08,052 epoch 8 - iter 104/138 - loss 0.08892701 - time (sec): 76.08 - samples/sec: 224.54 - lr: 0.000038 - momentum: 0.000000
2023-10-07 01:29:18,376 epoch 8 - iter 117/138 - loss 0.08692053 - time (sec): 86.41 - samples/sec: 225.35 - lr: 0.000036 - momentum: 0.000000
2023-10-07 01:29:27,508 epoch 8 - iter 130/138 - loss 0.08573298 - time (sec): 95.54 - samples/sec: 225.11 - lr: 0.000035 - momentum: 0.000000
2023-10-07 01:29:32,893 ----------------------------------------------------------------------------------------------------
2023-10-07 01:29:32,893 EPOCH 8 done: loss 0.0922 - lr: 0.000035
2023-10-07 01:29:39,367 DEV : loss 0.13560204207897186 - f1-score (micro avg) 0.8487
2023-10-07 01:29:39,372 saving best model
2023-10-07 01:29:40,250 ----------------------------------------------------------------------------------------------------
2023-10-07 01:29:49,586 epoch 9 - iter 13/138 - loss 0.07874872 - time (sec): 9.33 - samples/sec: 229.78 - lr: 0.000032 - momentum: 0.000000
2023-10-07 01:29:59,591 epoch 9 - iter 26/138 - loss 0.07917520 - time (sec): 19.34 - samples/sec: 231.28 - lr: 0.000031 - momentum: 0.000000
2023-10-07 01:30:09,142 epoch 9 - iter 39/138 - loss 0.07841021 - time (sec): 28.89 - samples/sec: 232.68 - lr: 0.000029 - momentum: 0.000000
2023-10-07 01:30:19,580 epoch 9 - iter 52/138 - loss 0.07759241 - time (sec): 39.33 - samples/sec: 234.51 - lr: 0.000027 - momentum: 0.000000
2023-10-07 01:30:28,813 epoch 9 - iter 65/138 - loss 0.07677469 - time (sec): 48.56 - samples/sec: 231.89 - lr: 0.000026 - momentum: 0.000000
2023-10-07 01:30:39,300 epoch 9 - iter 78/138 - loss 0.07753629 - time (sec): 59.05 - samples/sec: 230.42 - lr: 0.000024 - momentum: 0.000000
2023-10-07 01:30:48,249 epoch 9 - iter 91/138 - loss 0.07580801 - time (sec): 68.00 - samples/sec: 228.01 - lr: 0.000023 - momentum: 0.000000
2023-10-07 01:30:57,903 epoch 9 - iter 104/138 - loss 0.07869909 - time (sec): 77.65 - samples/sec: 227.93 - lr: 0.000021 - momentum: 0.000000
2023-10-07 01:31:06,851 epoch 9 - iter 117/138 - loss 0.07986836 - time (sec): 86.60 - samples/sec: 226.44 - lr: 0.000020 - momentum: 0.000000
2023-10-07 01:31:15,797 epoch 9 - iter 130/138 - loss 0.07964601 - time (sec): 95.55 - samples/sec: 226.71 - lr: 0.000018 - momentum: 0.000000
2023-10-07 01:31:21,066 ----------------------------------------------------------------------------------------------------
2023-10-07 01:31:21,066 EPOCH 9 done: loss 0.0797 - lr: 0.000018
2023-10-07 01:31:27,542 DEV : loss 0.13089357316493988 - f1-score (micro avg) 0.8609
2023-10-07 01:31:27,547 saving best model
2023-10-07 01:31:28,443 ----------------------------------------------------------------------------------------------------
2023-10-07 01:31:37,497 epoch 10 - iter 13/138 - loss 0.05251554 - time (sec): 9.05 - samples/sec: 224.90 - lr: 0.000016 - momentum: 0.000000
2023-10-07 01:31:46,636 epoch 10 - iter 26/138 - loss 0.05742682 - time (sec): 18.19 - samples/sec: 226.20 - lr: 0.000014 - momentum: 0.000000
2023-10-07 01:31:55,887 epoch 10 - iter 39/138 - loss 0.06060797 - time (sec): 27.44 - samples/sec: 226.48 - lr: 0.000012 - momentum: 0.000000
2023-10-07 01:32:05,632 epoch 10 - iter 52/138 - loss 0.06450398 - time (sec): 37.19 - samples/sec: 226.39 - lr: 0.000011 - momentum: 0.000000
2023-10-07 01:32:15,827 epoch 10 - iter 65/138 - loss 0.06574069 - time (sec): 47.38 - samples/sec: 226.98 - lr: 0.000009 - momentum: 0.000000
2023-10-07 01:32:25,236 epoch 10 - iter 78/138 - loss 0.06923330 - time (sec): 56.79 - samples/sec: 226.30 - lr: 0.000008 - momentum: 0.000000
2023-10-07 01:32:35,271 epoch 10 - iter 91/138 - loss 0.06896849 - time (sec): 66.83 - samples/sec: 227.11 - lr: 0.000006 - momentum: 0.000000
2023-10-07 01:32:44,202 epoch 10 - iter 104/138 - loss 0.06823662 - time (sec): 75.76 - samples/sec: 225.11 - lr: 0.000005 - momentum: 0.000000
2023-10-07 01:32:54,442 epoch 10 - iter 117/138 - loss 0.06764845 - time (sec): 86.00 - samples/sec: 225.88 - lr: 0.000003 - momentum: 0.000000
2023-10-07 01:33:03,573 epoch 10 - iter 130/138 - loss 0.07308496 - time (sec): 95.13 - samples/sec: 225.56 - lr: 0.000001 - momentum: 0.000000
2023-10-07 01:33:09,302 ----------------------------------------------------------------------------------------------------
2023-10-07 01:33:09,302 EPOCH 10 done: loss 0.0725 - lr: 0.000001
2023-10-07 01:33:15,797 DEV : loss 0.1304098665714264 - f1-score (micro avg) 0.8589
2023-10-07 01:33:16,632 ----------------------------------------------------------------------------------------------------
2023-10-07 01:33:16,634 Loading model from best epoch ...
2023-10-07 01:33:19,817 SequenceTagger predicts: Dictionary with 25 tags: O, S-scope, B-scope, E-scope, I-scope, S-pers, B-pers, E-pers, I-pers, S-work, B-work, E-work, I-work, S-loc, B-loc, E-loc, I-loc, S-object, B-object, E-object, I-object, S-date, B-date, E-date, I-date
2023-10-07 01:33:26,784
Results:
- F-score (micro) 0.8903
- F-score (macro) 0.5298
- Accuracy 0.8257
By class:
precision recall f1-score support
scope 0.9148 0.9148 0.9148 176
pers 0.9070 0.9141 0.9105 128
work 0.7975 0.8514 0.8235 74
object 0.0000 0.0000 0.0000 2
loc 0.0000 0.0000 0.0000 2
micro avg 0.8880 0.8927 0.8903 382
macro avg 0.5238 0.5360 0.5298 382
weighted avg 0.8799 0.8927 0.8861 382
2023-10-07 01:33:26,784 ----------------------------------------------------------------------------------------------------
|