File size: 25,249 Bytes
5d65ed0 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 |
2023-10-10 22:00:43,304 ----------------------------------------------------------------------------------------------------
2023-10-10 22:00:43,306 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): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(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): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(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): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(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): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(final_layer_norm): FusedRMSNorm(torch.Size([1472]), eps=1e-06, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=1472, out_features=17, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-10 22:00:43,306 ----------------------------------------------------------------------------------------------------
2023-10-10 22:00:43,307 MultiCorpus: 1166 train + 165 dev + 415 test sentences
- NER_HIPE_2022 Corpus: 1166 train + 165 dev + 415 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/newseye/fi/with_doc_seperator
2023-10-10 22:00:43,307 ----------------------------------------------------------------------------------------------------
2023-10-10 22:00:43,307 Train: 1166 sentences
2023-10-10 22:00:43,307 (train_with_dev=False, train_with_test=False)
2023-10-10 22:00:43,307 ----------------------------------------------------------------------------------------------------
2023-10-10 22:00:43,307 Training Params:
2023-10-10 22:00:43,307 - learning_rate: "0.00016"
2023-10-10 22:00:43,307 - mini_batch_size: "8"
2023-10-10 22:00:43,307 - max_epochs: "10"
2023-10-10 22:00:43,307 - shuffle: "True"
2023-10-10 22:00:43,307 ----------------------------------------------------------------------------------------------------
2023-10-10 22:00:43,307 Plugins:
2023-10-10 22:00:43,307 - TensorboardLogger
2023-10-10 22:00:43,307 - LinearScheduler | warmup_fraction: '0.1'
2023-10-10 22:00:43,308 ----------------------------------------------------------------------------------------------------
2023-10-10 22:00:43,308 Final evaluation on model from best epoch (best-model.pt)
2023-10-10 22:00:43,308 - metric: "('micro avg', 'f1-score')"
2023-10-10 22:00:43,308 ----------------------------------------------------------------------------------------------------
2023-10-10 22:00:43,308 Computation:
2023-10-10 22:00:43,308 - compute on device: cuda:0
2023-10-10 22:00:43,308 - embedding storage: none
2023-10-10 22:00:43,308 ----------------------------------------------------------------------------------------------------
2023-10-10 22:00:43,308 Model training base path: "hmbench-newseye/fi-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00016-poolingfirst-layers-1-crfFalse-1"
2023-10-10 22:00:43,308 ----------------------------------------------------------------------------------------------------
2023-10-10 22:00:43,308 ----------------------------------------------------------------------------------------------------
2023-10-10 22:00:43,308 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-10 22:00:52,394 epoch 1 - iter 14/146 - loss 2.83088289 - time (sec): 9.08 - samples/sec: 530.64 - lr: 0.000014 - momentum: 0.000000
2023-10-10 22:01:00,821 epoch 1 - iter 28/146 - loss 2.82628929 - time (sec): 17.51 - samples/sec: 492.26 - lr: 0.000030 - momentum: 0.000000
2023-10-10 22:01:10,396 epoch 1 - iter 42/146 - loss 2.81614191 - time (sec): 27.09 - samples/sec: 505.57 - lr: 0.000045 - momentum: 0.000000
2023-10-10 22:01:19,597 epoch 1 - iter 56/146 - loss 2.80155779 - time (sec): 36.29 - samples/sec: 497.39 - lr: 0.000060 - momentum: 0.000000
2023-10-10 22:01:28,047 epoch 1 - iter 70/146 - loss 2.77881902 - time (sec): 44.74 - samples/sec: 485.62 - lr: 0.000076 - momentum: 0.000000
2023-10-10 22:01:36,553 epoch 1 - iter 84/146 - loss 2.73383396 - time (sec): 53.24 - samples/sec: 482.56 - lr: 0.000091 - momentum: 0.000000
2023-10-10 22:01:44,474 epoch 1 - iter 98/146 - loss 2.67503100 - time (sec): 61.16 - samples/sec: 478.46 - lr: 0.000106 - momentum: 0.000000
2023-10-10 22:01:54,877 epoch 1 - iter 112/146 - loss 2.57315899 - time (sec): 71.57 - samples/sec: 485.69 - lr: 0.000122 - momentum: 0.000000
2023-10-10 22:02:03,653 epoch 1 - iter 126/146 - loss 2.49591763 - time (sec): 80.34 - samples/sec: 482.87 - lr: 0.000137 - momentum: 0.000000
2023-10-10 22:02:12,893 epoch 1 - iter 140/146 - loss 2.41324023 - time (sec): 89.58 - samples/sec: 479.18 - lr: 0.000152 - momentum: 0.000000
2023-10-10 22:02:16,311 ----------------------------------------------------------------------------------------------------
2023-10-10 22:02:16,311 EPOCH 1 done: loss 2.3834 - lr: 0.000152
2023-10-10 22:02:21,884 DEV : loss 1.2905865907669067 - f1-score (micro avg) 0.0
2023-10-10 22:02:21,893 ----------------------------------------------------------------------------------------------------
2023-10-10 22:02:31,067 epoch 2 - iter 14/146 - loss 1.32629947 - time (sec): 9.17 - samples/sec: 499.38 - lr: 0.000158 - momentum: 0.000000
2023-10-10 22:02:40,124 epoch 2 - iter 28/146 - loss 1.29112932 - time (sec): 18.23 - samples/sec: 488.68 - lr: 0.000157 - momentum: 0.000000
2023-10-10 22:02:49,458 epoch 2 - iter 42/146 - loss 1.15322127 - time (sec): 27.56 - samples/sec: 476.15 - lr: 0.000155 - momentum: 0.000000
2023-10-10 22:02:59,632 epoch 2 - iter 56/146 - loss 1.05695895 - time (sec): 37.74 - samples/sec: 467.96 - lr: 0.000153 - momentum: 0.000000
2023-10-10 22:03:10,417 epoch 2 - iter 70/146 - loss 0.98742812 - time (sec): 48.52 - samples/sec: 472.43 - lr: 0.000152 - momentum: 0.000000
2023-10-10 22:03:19,883 epoch 2 - iter 84/146 - loss 0.93247747 - time (sec): 57.99 - samples/sec: 463.34 - lr: 0.000150 - momentum: 0.000000
2023-10-10 22:03:29,139 epoch 2 - iter 98/146 - loss 0.88912171 - time (sec): 67.24 - samples/sec: 457.24 - lr: 0.000148 - momentum: 0.000000
2023-10-10 22:03:38,219 epoch 2 - iter 112/146 - loss 0.85474722 - time (sec): 76.32 - samples/sec: 452.98 - lr: 0.000147 - momentum: 0.000000
2023-10-10 22:03:46,844 epoch 2 - iter 126/146 - loss 0.82173264 - time (sec): 84.95 - samples/sec: 456.28 - lr: 0.000145 - momentum: 0.000000
2023-10-10 22:03:55,388 epoch 2 - iter 140/146 - loss 0.79360195 - time (sec): 93.49 - samples/sec: 459.36 - lr: 0.000143 - momentum: 0.000000
2023-10-10 22:03:58,923 ----------------------------------------------------------------------------------------------------
2023-10-10 22:03:58,923 EPOCH 2 done: loss 0.7905 - lr: 0.000143
2023-10-10 22:04:04,751 DEV : loss 0.3894149363040924 - f1-score (micro avg) 0.0
2023-10-10 22:04:04,760 ----------------------------------------------------------------------------------------------------
2023-10-10 22:04:13,057 epoch 3 - iter 14/146 - loss 0.51507142 - time (sec): 8.29 - samples/sec: 425.69 - lr: 0.000141 - momentum: 0.000000
2023-10-10 22:04:22,691 epoch 3 - iter 28/146 - loss 0.41692054 - time (sec): 17.93 - samples/sec: 477.51 - lr: 0.000139 - momentum: 0.000000
2023-10-10 22:04:31,502 epoch 3 - iter 42/146 - loss 0.43730739 - time (sec): 26.74 - samples/sec: 477.83 - lr: 0.000137 - momentum: 0.000000
2023-10-10 22:04:40,274 epoch 3 - iter 56/146 - loss 0.42787094 - time (sec): 35.51 - samples/sec: 470.86 - lr: 0.000136 - momentum: 0.000000
2023-10-10 22:04:49,689 epoch 3 - iter 70/146 - loss 0.41822906 - time (sec): 44.93 - samples/sec: 467.47 - lr: 0.000134 - momentum: 0.000000
2023-10-10 22:04:58,317 epoch 3 - iter 84/146 - loss 0.41836824 - time (sec): 53.55 - samples/sec: 459.77 - lr: 0.000132 - momentum: 0.000000
2023-10-10 22:05:07,820 epoch 3 - iter 98/146 - loss 0.44258611 - time (sec): 63.06 - samples/sec: 473.68 - lr: 0.000131 - momentum: 0.000000
2023-10-10 22:05:17,349 epoch 3 - iter 112/146 - loss 0.42988452 - time (sec): 72.59 - samples/sec: 478.20 - lr: 0.000129 - momentum: 0.000000
2023-10-10 22:05:26,127 epoch 3 - iter 126/146 - loss 0.41794490 - time (sec): 81.36 - samples/sec: 480.97 - lr: 0.000127 - momentum: 0.000000
2023-10-10 22:05:34,136 epoch 3 - iter 140/146 - loss 0.41620303 - time (sec): 89.37 - samples/sec: 476.97 - lr: 0.000125 - momentum: 0.000000
2023-10-10 22:05:37,768 ----------------------------------------------------------------------------------------------------
2023-10-10 22:05:37,768 EPOCH 3 done: loss 0.4112 - lr: 0.000125
2023-10-10 22:05:43,626 DEV : loss 0.28306612372398376 - f1-score (micro avg) 0.1609
2023-10-10 22:05:43,635 saving best model
2023-10-10 22:05:44,623 ----------------------------------------------------------------------------------------------------
2023-10-10 22:05:54,251 epoch 4 - iter 14/146 - loss 0.29881808 - time (sec): 9.62 - samples/sec: 492.36 - lr: 0.000123 - momentum: 0.000000
2023-10-10 22:06:02,849 epoch 4 - iter 28/146 - loss 0.27924207 - time (sec): 18.22 - samples/sec: 471.98 - lr: 0.000121 - momentum: 0.000000
2023-10-10 22:06:12,815 epoch 4 - iter 42/146 - loss 0.34183632 - time (sec): 28.19 - samples/sec: 478.45 - lr: 0.000120 - momentum: 0.000000
2023-10-10 22:06:21,685 epoch 4 - iter 56/146 - loss 0.34079281 - time (sec): 37.06 - samples/sec: 471.65 - lr: 0.000118 - momentum: 0.000000
2023-10-10 22:06:30,830 epoch 4 - iter 70/146 - loss 0.33661860 - time (sec): 46.20 - samples/sec: 478.46 - lr: 0.000116 - momentum: 0.000000
2023-10-10 22:06:39,823 epoch 4 - iter 84/146 - loss 0.33761968 - time (sec): 55.20 - samples/sec: 477.34 - lr: 0.000115 - momentum: 0.000000
2023-10-10 22:06:50,180 epoch 4 - iter 98/146 - loss 0.32990070 - time (sec): 65.55 - samples/sec: 472.88 - lr: 0.000113 - momentum: 0.000000
2023-10-10 22:06:59,582 epoch 4 - iter 112/146 - loss 0.32645728 - time (sec): 74.96 - samples/sec: 464.23 - lr: 0.000111 - momentum: 0.000000
2023-10-10 22:07:08,979 epoch 4 - iter 126/146 - loss 0.32140043 - time (sec): 84.35 - samples/sec: 460.47 - lr: 0.000109 - momentum: 0.000000
2023-10-10 22:07:17,392 epoch 4 - iter 140/146 - loss 0.32290848 - time (sec): 92.77 - samples/sec: 457.95 - lr: 0.000108 - momentum: 0.000000
2023-10-10 22:07:21,205 ----------------------------------------------------------------------------------------------------
2023-10-10 22:07:21,205 EPOCH 4 done: loss 0.3174 - lr: 0.000108
2023-10-10 22:07:26,848 DEV : loss 0.2454710602760315 - f1-score (micro avg) 0.3563
2023-10-10 22:07:26,857 saving best model
2023-10-10 22:07:35,506 ----------------------------------------------------------------------------------------------------
2023-10-10 22:07:44,550 epoch 5 - iter 14/146 - loss 0.27566779 - time (sec): 9.04 - samples/sec: 454.76 - lr: 0.000105 - momentum: 0.000000
2023-10-10 22:07:54,215 epoch 5 - iter 28/146 - loss 0.33438474 - time (sec): 18.70 - samples/sec: 484.48 - lr: 0.000104 - momentum: 0.000000
2023-10-10 22:08:02,953 epoch 5 - iter 42/146 - loss 0.32225237 - time (sec): 27.44 - samples/sec: 484.25 - lr: 0.000102 - momentum: 0.000000
2023-10-10 22:08:11,715 epoch 5 - iter 56/146 - loss 0.29084459 - time (sec): 36.20 - samples/sec: 478.55 - lr: 0.000100 - momentum: 0.000000
2023-10-10 22:08:21,054 epoch 5 - iter 70/146 - loss 0.27706780 - time (sec): 45.54 - samples/sec: 481.60 - lr: 0.000099 - momentum: 0.000000
2023-10-10 22:08:30,720 epoch 5 - iter 84/146 - loss 0.26855614 - time (sec): 55.21 - samples/sec: 469.63 - lr: 0.000097 - momentum: 0.000000
2023-10-10 22:08:40,750 epoch 5 - iter 98/146 - loss 0.26372697 - time (sec): 65.24 - samples/sec: 463.02 - lr: 0.000095 - momentum: 0.000000
2023-10-10 22:08:51,290 epoch 5 - iter 112/146 - loss 0.26061632 - time (sec): 75.78 - samples/sec: 458.22 - lr: 0.000093 - momentum: 0.000000
2023-10-10 22:09:00,154 epoch 5 - iter 126/146 - loss 0.25838387 - time (sec): 84.64 - samples/sec: 456.78 - lr: 0.000092 - momentum: 0.000000
2023-10-10 22:09:08,987 epoch 5 - iter 140/146 - loss 0.25502563 - time (sec): 93.48 - samples/sec: 456.41 - lr: 0.000090 - momentum: 0.000000
2023-10-10 22:09:12,868 ----------------------------------------------------------------------------------------------------
2023-10-10 22:09:12,869 EPOCH 5 done: loss 0.2561 - lr: 0.000090
2023-10-10 22:09:18,945 DEV : loss 0.20666354894638062 - f1-score (micro avg) 0.4602
2023-10-10 22:09:18,955 saving best model
2023-10-10 22:09:25,879 ----------------------------------------------------------------------------------------------------
2023-10-10 22:09:35,820 epoch 6 - iter 14/146 - loss 0.21070114 - time (sec): 9.94 - samples/sec: 504.06 - lr: 0.000088 - momentum: 0.000000
2023-10-10 22:09:44,414 epoch 6 - iter 28/146 - loss 0.22631495 - time (sec): 18.53 - samples/sec: 483.31 - lr: 0.000086 - momentum: 0.000000
2023-10-10 22:09:53,224 epoch 6 - iter 42/146 - loss 0.21485580 - time (sec): 27.34 - samples/sec: 480.19 - lr: 0.000084 - momentum: 0.000000
2023-10-10 22:10:02,417 epoch 6 - iter 56/146 - loss 0.20592699 - time (sec): 36.53 - samples/sec: 482.40 - lr: 0.000083 - momentum: 0.000000
2023-10-10 22:10:11,772 epoch 6 - iter 70/146 - loss 0.20663057 - time (sec): 45.89 - samples/sec: 485.72 - lr: 0.000081 - momentum: 0.000000
2023-10-10 22:10:20,029 epoch 6 - iter 84/146 - loss 0.20369501 - time (sec): 54.15 - samples/sec: 478.12 - lr: 0.000079 - momentum: 0.000000
2023-10-10 22:10:28,856 epoch 6 - iter 98/146 - loss 0.20747226 - time (sec): 62.97 - samples/sec: 474.76 - lr: 0.000077 - momentum: 0.000000
2023-10-10 22:10:37,595 epoch 6 - iter 112/146 - loss 0.20151474 - time (sec): 71.71 - samples/sec: 474.56 - lr: 0.000076 - momentum: 0.000000
2023-10-10 22:10:46,686 epoch 6 - iter 126/146 - loss 0.20676841 - time (sec): 80.80 - samples/sec: 478.66 - lr: 0.000074 - momentum: 0.000000
2023-10-10 22:10:54,676 epoch 6 - iter 140/146 - loss 0.20652055 - time (sec): 88.79 - samples/sec: 474.05 - lr: 0.000072 - momentum: 0.000000
2023-10-10 22:10:58,983 ----------------------------------------------------------------------------------------------------
2023-10-10 22:10:58,983 EPOCH 6 done: loss 0.2048 - lr: 0.000072
2023-10-10 22:11:05,038 DEV : loss 0.18372002243995667 - f1-score (micro avg) 0.5195
2023-10-10 22:11:05,048 saving best model
2023-10-10 22:11:14,179 ----------------------------------------------------------------------------------------------------
2023-10-10 22:11:23,472 epoch 7 - iter 14/146 - loss 0.16630834 - time (sec): 9.29 - samples/sec: 515.70 - lr: 0.000070 - momentum: 0.000000
2023-10-10 22:11:31,585 epoch 7 - iter 28/146 - loss 0.15910112 - time (sec): 17.40 - samples/sec: 494.00 - lr: 0.000068 - momentum: 0.000000
2023-10-10 22:11:39,899 epoch 7 - iter 42/146 - loss 0.18159621 - time (sec): 25.72 - samples/sec: 496.78 - lr: 0.000067 - momentum: 0.000000
2023-10-10 22:11:48,970 epoch 7 - iter 56/146 - loss 0.16818074 - time (sec): 34.79 - samples/sec: 507.13 - lr: 0.000065 - momentum: 0.000000
2023-10-10 22:11:57,416 epoch 7 - iter 70/146 - loss 0.16285809 - time (sec): 43.23 - samples/sec: 506.75 - lr: 0.000063 - momentum: 0.000000
2023-10-10 22:12:05,581 epoch 7 - iter 84/146 - loss 0.16942539 - time (sec): 51.40 - samples/sec: 493.58 - lr: 0.000061 - momentum: 0.000000
2023-10-10 22:12:14,887 epoch 7 - iter 98/146 - loss 0.16837468 - time (sec): 60.70 - samples/sec: 487.16 - lr: 0.000060 - momentum: 0.000000
2023-10-10 22:12:24,198 epoch 7 - iter 112/146 - loss 0.16411960 - time (sec): 70.01 - samples/sec: 491.76 - lr: 0.000058 - momentum: 0.000000
2023-10-10 22:12:33,243 epoch 7 - iter 126/146 - loss 0.16260578 - time (sec): 79.06 - samples/sec: 490.77 - lr: 0.000056 - momentum: 0.000000
2023-10-10 22:12:42,356 epoch 7 - iter 140/146 - loss 0.16658286 - time (sec): 88.17 - samples/sec: 480.73 - lr: 0.000055 - momentum: 0.000000
2023-10-10 22:12:46,633 ----------------------------------------------------------------------------------------------------
2023-10-10 22:12:46,633 EPOCH 7 done: loss 0.1690 - lr: 0.000055
2023-10-10 22:12:52,591 DEV : loss 0.17367814481258392 - f1-score (micro avg) 0.604
2023-10-10 22:12:52,600 saving best model
2023-10-10 22:13:00,680 ----------------------------------------------------------------------------------------------------
2023-10-10 22:13:10,905 epoch 8 - iter 14/146 - loss 0.14405019 - time (sec): 10.21 - samples/sec: 481.65 - lr: 0.000052 - momentum: 0.000000
2023-10-10 22:13:20,300 epoch 8 - iter 28/146 - loss 0.15973513 - time (sec): 19.61 - samples/sec: 494.79 - lr: 0.000051 - momentum: 0.000000
2023-10-10 22:13:29,217 epoch 8 - iter 42/146 - loss 0.16052807 - time (sec): 28.53 - samples/sec: 486.59 - lr: 0.000049 - momentum: 0.000000
2023-10-10 22:13:38,038 epoch 8 - iter 56/146 - loss 0.15210947 - time (sec): 37.35 - samples/sec: 471.97 - lr: 0.000047 - momentum: 0.000000
2023-10-10 22:13:47,341 epoch 8 - iter 70/146 - loss 0.15143169 - time (sec): 46.65 - samples/sec: 474.18 - lr: 0.000045 - momentum: 0.000000
2023-10-10 22:13:56,442 epoch 8 - iter 84/146 - loss 0.15707453 - time (sec): 55.75 - samples/sec: 466.59 - lr: 0.000044 - momentum: 0.000000
2023-10-10 22:14:05,085 epoch 8 - iter 98/146 - loss 0.15161172 - time (sec): 64.39 - samples/sec: 465.47 - lr: 0.000042 - momentum: 0.000000
2023-10-10 22:14:14,292 epoch 8 - iter 112/146 - loss 0.14901872 - time (sec): 73.60 - samples/sec: 469.22 - lr: 0.000040 - momentum: 0.000000
2023-10-10 22:14:23,650 epoch 8 - iter 126/146 - loss 0.14657533 - time (sec): 82.96 - samples/sec: 464.50 - lr: 0.000039 - momentum: 0.000000
2023-10-10 22:14:32,792 epoch 8 - iter 140/146 - loss 0.14114562 - time (sec): 92.10 - samples/sec: 465.03 - lr: 0.000037 - momentum: 0.000000
2023-10-10 22:14:36,412 ----------------------------------------------------------------------------------------------------
2023-10-10 22:14:36,412 EPOCH 8 done: loss 0.1447 - lr: 0.000037
2023-10-10 22:14:42,561 DEV : loss 0.16653960943222046 - f1-score (micro avg) 0.6522
2023-10-10 22:14:42,570 saving best model
2023-10-10 22:14:52,366 ----------------------------------------------------------------------------------------------------
2023-10-10 22:15:01,070 epoch 9 - iter 14/146 - loss 0.10564763 - time (sec): 8.70 - samples/sec: 483.88 - lr: 0.000035 - momentum: 0.000000
2023-10-10 22:15:10,860 epoch 9 - iter 28/146 - loss 0.13546293 - time (sec): 18.49 - samples/sec: 494.59 - lr: 0.000033 - momentum: 0.000000
2023-10-10 22:15:19,875 epoch 9 - iter 42/146 - loss 0.13424335 - time (sec): 27.50 - samples/sec: 482.35 - lr: 0.000031 - momentum: 0.000000
2023-10-10 22:15:28,801 epoch 9 - iter 56/146 - loss 0.12517220 - time (sec): 36.43 - samples/sec: 481.07 - lr: 0.000029 - momentum: 0.000000
2023-10-10 22:15:37,873 epoch 9 - iter 70/146 - loss 0.12523915 - time (sec): 45.50 - samples/sec: 480.35 - lr: 0.000028 - momentum: 0.000000
2023-10-10 22:15:46,336 epoch 9 - iter 84/146 - loss 0.12571857 - time (sec): 53.97 - samples/sec: 479.53 - lr: 0.000026 - momentum: 0.000000
2023-10-10 22:15:55,269 epoch 9 - iter 98/146 - loss 0.13040790 - time (sec): 62.90 - samples/sec: 480.15 - lr: 0.000024 - momentum: 0.000000
2023-10-10 22:16:03,942 epoch 9 - iter 112/146 - loss 0.12849922 - time (sec): 71.57 - samples/sec: 479.88 - lr: 0.000023 - momentum: 0.000000
2023-10-10 22:16:13,165 epoch 9 - iter 126/146 - loss 0.12666524 - time (sec): 80.79 - samples/sec: 484.20 - lr: 0.000021 - momentum: 0.000000
2023-10-10 22:16:21,714 epoch 9 - iter 140/146 - loss 0.12857980 - time (sec): 89.34 - samples/sec: 479.99 - lr: 0.000019 - momentum: 0.000000
2023-10-10 22:16:25,263 ----------------------------------------------------------------------------------------------------
2023-10-10 22:16:25,264 EPOCH 9 done: loss 0.1282 - lr: 0.000019
2023-10-10 22:16:31,347 DEV : loss 0.1624951958656311 - f1-score (micro avg) 0.6638
2023-10-10 22:16:31,356 saving best model
2023-10-10 22:16:39,276 ----------------------------------------------------------------------------------------------------
2023-10-10 22:16:48,522 epoch 10 - iter 14/146 - loss 0.11168896 - time (sec): 9.24 - samples/sec: 485.28 - lr: 0.000017 - momentum: 0.000000
2023-10-10 22:16:57,095 epoch 10 - iter 28/146 - loss 0.12877886 - time (sec): 17.81 - samples/sec: 457.50 - lr: 0.000015 - momentum: 0.000000
2023-10-10 22:17:05,914 epoch 10 - iter 42/146 - loss 0.11898550 - time (sec): 26.63 - samples/sec: 463.39 - lr: 0.000013 - momentum: 0.000000
2023-10-10 22:17:15,949 epoch 10 - iter 56/146 - loss 0.10834056 - time (sec): 36.67 - samples/sec: 479.49 - lr: 0.000012 - momentum: 0.000000
2023-10-10 22:17:25,541 epoch 10 - iter 70/146 - loss 0.10628809 - time (sec): 46.26 - samples/sec: 483.48 - lr: 0.000010 - momentum: 0.000000
2023-10-10 22:17:34,310 epoch 10 - iter 84/146 - loss 0.10553800 - time (sec): 55.03 - samples/sec: 477.65 - lr: 0.000008 - momentum: 0.000000
2023-10-10 22:17:43,365 epoch 10 - iter 98/146 - loss 0.10631791 - time (sec): 64.08 - samples/sec: 474.69 - lr: 0.000007 - momentum: 0.000000
2023-10-10 22:17:52,344 epoch 10 - iter 112/146 - loss 0.10970536 - time (sec): 73.06 - samples/sec: 475.00 - lr: 0.000005 - momentum: 0.000000
2023-10-10 22:18:01,109 epoch 10 - iter 126/146 - loss 0.11345018 - time (sec): 81.83 - samples/sec: 473.11 - lr: 0.000003 - momentum: 0.000000
2023-10-10 22:18:10,282 epoch 10 - iter 140/146 - loss 0.11718872 - time (sec): 91.00 - samples/sec: 472.88 - lr: 0.000002 - momentum: 0.000000
2023-10-10 22:18:13,713 ----------------------------------------------------------------------------------------------------
2023-10-10 22:18:13,714 EPOCH 10 done: loss 0.1203 - lr: 0.000002
2023-10-10 22:18:19,724 DEV : loss 0.16049127280712128 - f1-score (micro avg) 0.6566
2023-10-10 22:18:20,641 ----------------------------------------------------------------------------------------------------
2023-10-10 22:18:20,643 Loading model from best epoch ...
2023-10-10 22:18:24,871 SequenceTagger predicts: Dictionary with 17 tags: O, S-LOC, B-LOC, E-LOC, I-LOC, S-PER, B-PER, E-PER, I-PER, S-ORG, B-ORG, E-ORG, I-ORG, S-HumanProd, B-HumanProd, E-HumanProd, I-HumanProd
2023-10-10 22:18:37,888
Results:
- F-score (micro) 0.6941
- F-score (macro) 0.5894
- Accuracy 0.5673
By class:
precision recall f1-score support
PER 0.7945 0.7443 0.7685 348
LOC 0.5994 0.7854 0.6799 261
ORG 0.3191 0.2885 0.3030 52
HumanProd 0.9091 0.4545 0.6061 22
micro avg 0.6736 0.7160 0.6941 683
macro avg 0.6555 0.5682 0.5894 683
weighted avg 0.6874 0.7160 0.6940 683
2023-10-10 22:18:37,888 ----------------------------------------------------------------------------------------------------
|