File size: 24,064 Bytes
b696a52 |
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 |
2023-10-23 15:09:00,085 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:00,086 Model: "SequenceTagger(
(embeddings): TransformerWordEmbeddings(
(model): BertModel(
(embeddings): BertEmbeddings(
(word_embeddings): Embedding(64001, 768)
(position_embeddings): Embedding(512, 768)
(token_type_embeddings): Embedding(2, 768)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(encoder): BertEncoder(
(layer): ModuleList(
(0-11): 12 x BertLayer(
(attention): BertAttention(
(self): BertSelfAttention(
(query): Linear(in_features=768, out_features=768, bias=True)
(key): Linear(in_features=768, out_features=768, bias=True)
(value): Linear(in_features=768, out_features=768, bias=True)
(dropout): Dropout(p=0.1, inplace=False)
)
(output): BertSelfOutput(
(dense): Linear(in_features=768, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
(intermediate): BertIntermediate(
(dense): Linear(in_features=768, out_features=3072, bias=True)
(intermediate_act_fn): GELUActivation()
)
(output): BertOutput(
(dense): Linear(in_features=3072, out_features=768, bias=True)
(LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)
(dropout): Dropout(p=0.1, inplace=False)
)
)
)
)
(pooler): BertPooler(
(dense): Linear(in_features=768, out_features=768, bias=True)
(activation): Tanh()
)
)
)
(locked_dropout): LockedDropout(p=0.5)
(linear): Linear(in_features=768, out_features=25, bias=True)
(loss_function): CrossEntropyLoss()
)"
2023-10-23 15:09:00,086 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:00,086 MultiCorpus: 1100 train + 206 dev + 240 test sentences
- NER_HIPE_2022 Corpus: 1100 train + 206 dev + 240 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/ajmc/de/with_doc_seperator
2023-10-23 15:09:00,086 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:00,086 Train: 1100 sentences
2023-10-23 15:09:00,086 (train_with_dev=False, train_with_test=False)
2023-10-23 15:09:00,086 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:00,086 Training Params:
2023-10-23 15:09:00,086 - learning_rate: "5e-05"
2023-10-23 15:09:00,086 - mini_batch_size: "8"
2023-10-23 15:09:00,086 - max_epochs: "10"
2023-10-23 15:09:00,086 - shuffle: "True"
2023-10-23 15:09:00,087 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:00,087 Plugins:
2023-10-23 15:09:00,087 - TensorboardLogger
2023-10-23 15:09:00,087 - LinearScheduler | warmup_fraction: '0.1'
2023-10-23 15:09:00,087 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:00,087 Final evaluation on model from best epoch (best-model.pt)
2023-10-23 15:09:00,087 - metric: "('micro avg', 'f1-score')"
2023-10-23 15:09:00,087 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:00,087 Computation:
2023-10-23 15:09:00,087 - compute on device: cuda:0
2023-10-23 15:09:00,087 - embedding storage: none
2023-10-23 15:09:00,087 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:00,087 Model training base path: "hmbench-ajmc/de-dbmdz/bert-base-historic-multilingual-64k-td-cased-bs8-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-2"
2023-10-23 15:09:00,087 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:00,087 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:00,087 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-23 15:09:00,807 epoch 1 - iter 13/138 - loss 3.30817675 - time (sec): 0.72 - samples/sec: 2987.42 - lr: 0.000004 - momentum: 0.000000
2023-10-23 15:09:01,521 epoch 1 - iter 26/138 - loss 2.71086038 - time (sec): 1.43 - samples/sec: 2914.66 - lr: 0.000009 - momentum: 0.000000
2023-10-23 15:09:02,222 epoch 1 - iter 39/138 - loss 2.15628855 - time (sec): 2.13 - samples/sec: 2855.54 - lr: 0.000014 - momentum: 0.000000
2023-10-23 15:09:02,927 epoch 1 - iter 52/138 - loss 1.82507197 - time (sec): 2.84 - samples/sec: 2838.66 - lr: 0.000018 - momentum: 0.000000
2023-10-23 15:09:03,650 epoch 1 - iter 65/138 - loss 1.56174745 - time (sec): 3.56 - samples/sec: 2905.86 - lr: 0.000023 - momentum: 0.000000
2023-10-23 15:09:04,366 epoch 1 - iter 78/138 - loss 1.35697247 - time (sec): 4.28 - samples/sec: 2967.13 - lr: 0.000028 - momentum: 0.000000
2023-10-23 15:09:05,087 epoch 1 - iter 91/138 - loss 1.21045319 - time (sec): 5.00 - samples/sec: 2982.54 - lr: 0.000033 - momentum: 0.000000
2023-10-23 15:09:05,806 epoch 1 - iter 104/138 - loss 1.09402921 - time (sec): 5.72 - samples/sec: 2997.08 - lr: 0.000037 - momentum: 0.000000
2023-10-23 15:09:06,524 epoch 1 - iter 117/138 - loss 0.99931724 - time (sec): 6.44 - samples/sec: 2994.86 - lr: 0.000042 - momentum: 0.000000
2023-10-23 15:09:07,232 epoch 1 - iter 130/138 - loss 0.91695124 - time (sec): 7.14 - samples/sec: 3022.26 - lr: 0.000047 - momentum: 0.000000
2023-10-23 15:09:07,663 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:07,663 EPOCH 1 done: loss 0.8835 - lr: 0.000047
2023-10-23 15:09:08,085 DEV : loss 0.16610857844352722 - f1-score (micro avg) 0.7589
2023-10-23 15:09:08,092 saving best model
2023-10-23 15:09:08,500 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:09,217 epoch 2 - iter 13/138 - loss 0.11222475 - time (sec): 0.72 - samples/sec: 2927.76 - lr: 0.000050 - momentum: 0.000000
2023-10-23 15:09:09,933 epoch 2 - iter 26/138 - loss 0.13490147 - time (sec): 1.43 - samples/sec: 2920.96 - lr: 0.000049 - momentum: 0.000000
2023-10-23 15:09:10,649 epoch 2 - iter 39/138 - loss 0.14882689 - time (sec): 2.15 - samples/sec: 2875.93 - lr: 0.000048 - momentum: 0.000000
2023-10-23 15:09:11,364 epoch 2 - iter 52/138 - loss 0.16367161 - time (sec): 2.86 - samples/sec: 2939.80 - lr: 0.000048 - momentum: 0.000000
2023-10-23 15:09:12,242 epoch 2 - iter 65/138 - loss 0.17017491 - time (sec): 3.74 - samples/sec: 2808.03 - lr: 0.000047 - momentum: 0.000000
2023-10-23 15:09:12,958 epoch 2 - iter 78/138 - loss 0.16368057 - time (sec): 4.46 - samples/sec: 2805.45 - lr: 0.000047 - momentum: 0.000000
2023-10-23 15:09:13,671 epoch 2 - iter 91/138 - loss 0.15206442 - time (sec): 5.17 - samples/sec: 2810.79 - lr: 0.000046 - momentum: 0.000000
2023-10-23 15:09:14,401 epoch 2 - iter 104/138 - loss 0.15146141 - time (sec): 5.90 - samples/sec: 2873.90 - lr: 0.000046 - momentum: 0.000000
2023-10-23 15:09:15,108 epoch 2 - iter 117/138 - loss 0.14489742 - time (sec): 6.61 - samples/sec: 2898.70 - lr: 0.000045 - momentum: 0.000000
2023-10-23 15:09:15,825 epoch 2 - iter 130/138 - loss 0.14753327 - time (sec): 7.32 - samples/sec: 2923.39 - lr: 0.000045 - momentum: 0.000000
2023-10-23 15:09:16,264 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:16,264 EPOCH 2 done: loss 0.1500 - lr: 0.000045
2023-10-23 15:09:16,800 DEV : loss 0.11995560675859451 - f1-score (micro avg) 0.8565
2023-10-23 15:09:16,806 saving best model
2023-10-23 15:09:17,358 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:18,121 epoch 3 - iter 13/138 - loss 0.08208827 - time (sec): 0.76 - samples/sec: 2882.91 - lr: 0.000044 - momentum: 0.000000
2023-10-23 15:09:18,861 epoch 3 - iter 26/138 - loss 0.07035186 - time (sec): 1.50 - samples/sec: 2825.43 - lr: 0.000043 - momentum: 0.000000
2023-10-23 15:09:19,577 epoch 3 - iter 39/138 - loss 0.06559420 - time (sec): 2.21 - samples/sec: 2741.33 - lr: 0.000043 - momentum: 0.000000
2023-10-23 15:09:20,313 epoch 3 - iter 52/138 - loss 0.07919367 - time (sec): 2.95 - samples/sec: 2853.16 - lr: 0.000042 - momentum: 0.000000
2023-10-23 15:09:21,082 epoch 3 - iter 65/138 - loss 0.07261362 - time (sec): 3.72 - samples/sec: 2848.48 - lr: 0.000042 - momentum: 0.000000
2023-10-23 15:09:21,845 epoch 3 - iter 78/138 - loss 0.08353706 - time (sec): 4.48 - samples/sec: 2897.77 - lr: 0.000041 - momentum: 0.000000
2023-10-23 15:09:22,591 epoch 3 - iter 91/138 - loss 0.08212751 - time (sec): 5.23 - samples/sec: 2878.72 - lr: 0.000041 - momentum: 0.000000
2023-10-23 15:09:23,357 epoch 3 - iter 104/138 - loss 0.09248064 - time (sec): 5.99 - samples/sec: 2855.22 - lr: 0.000040 - momentum: 0.000000
2023-10-23 15:09:24,117 epoch 3 - iter 117/138 - loss 0.08840392 - time (sec): 6.75 - samples/sec: 2827.73 - lr: 0.000040 - momentum: 0.000000
2023-10-23 15:09:24,861 epoch 3 - iter 130/138 - loss 0.08729436 - time (sec): 7.50 - samples/sec: 2877.59 - lr: 0.000039 - momentum: 0.000000
2023-10-23 15:09:25,304 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:25,304 EPOCH 3 done: loss 0.0879 - lr: 0.000039
2023-10-23 15:09:25,849 DEV : loss 0.12923850119113922 - f1-score (micro avg) 0.8419
2023-10-23 15:09:25,854 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:26,585 epoch 4 - iter 13/138 - loss 0.05200905 - time (sec): 0.73 - samples/sec: 2915.79 - lr: 0.000038 - momentum: 0.000000
2023-10-23 15:09:27,337 epoch 4 - iter 26/138 - loss 0.05412447 - time (sec): 1.48 - samples/sec: 2866.05 - lr: 0.000038 - momentum: 0.000000
2023-10-23 15:09:28,101 epoch 4 - iter 39/138 - loss 0.07019529 - time (sec): 2.25 - samples/sec: 2997.69 - lr: 0.000037 - momentum: 0.000000
2023-10-23 15:09:28,853 epoch 4 - iter 52/138 - loss 0.07814313 - time (sec): 3.00 - samples/sec: 2964.04 - lr: 0.000037 - momentum: 0.000000
2023-10-23 15:09:29,613 epoch 4 - iter 65/138 - loss 0.07463506 - time (sec): 3.76 - samples/sec: 2908.71 - lr: 0.000036 - momentum: 0.000000
2023-10-23 15:09:30,359 epoch 4 - iter 78/138 - loss 0.07111250 - time (sec): 4.50 - samples/sec: 2913.80 - lr: 0.000036 - momentum: 0.000000
2023-10-23 15:09:31,089 epoch 4 - iter 91/138 - loss 0.06615343 - time (sec): 5.23 - samples/sec: 2894.17 - lr: 0.000035 - momentum: 0.000000
2023-10-23 15:09:31,812 epoch 4 - iter 104/138 - loss 0.06073453 - time (sec): 5.96 - samples/sec: 2904.56 - lr: 0.000035 - momentum: 0.000000
2023-10-23 15:09:32,567 epoch 4 - iter 117/138 - loss 0.05896126 - time (sec): 6.71 - samples/sec: 2914.25 - lr: 0.000034 - momentum: 0.000000
2023-10-23 15:09:33,332 epoch 4 - iter 130/138 - loss 0.05591755 - time (sec): 7.48 - samples/sec: 2896.16 - lr: 0.000034 - momentum: 0.000000
2023-10-23 15:09:33,792 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:33,792 EPOCH 4 done: loss 0.0535 - lr: 0.000034
2023-10-23 15:09:34,333 DEV : loss 0.1422549933195114 - f1-score (micro avg) 0.8493
2023-10-23 15:09:34,338 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:35,067 epoch 5 - iter 13/138 - loss 0.08297898 - time (sec): 0.73 - samples/sec: 2802.41 - lr: 0.000033 - momentum: 0.000000
2023-10-23 15:09:35,799 epoch 5 - iter 26/138 - loss 0.04678987 - time (sec): 1.46 - samples/sec: 2966.01 - lr: 0.000032 - momentum: 0.000000
2023-10-23 15:09:36,520 epoch 5 - iter 39/138 - loss 0.04701843 - time (sec): 2.18 - samples/sec: 2955.73 - lr: 0.000032 - momentum: 0.000000
2023-10-23 15:09:37,234 epoch 5 - iter 52/138 - loss 0.04528808 - time (sec): 2.89 - samples/sec: 3000.09 - lr: 0.000031 - momentum: 0.000000
2023-10-23 15:09:37,955 epoch 5 - iter 65/138 - loss 0.05278775 - time (sec): 3.62 - samples/sec: 2919.24 - lr: 0.000031 - momentum: 0.000000
2023-10-23 15:09:38,701 epoch 5 - iter 78/138 - loss 0.04949446 - time (sec): 4.36 - samples/sec: 2975.21 - lr: 0.000030 - momentum: 0.000000
2023-10-23 15:09:39,429 epoch 5 - iter 91/138 - loss 0.04668619 - time (sec): 5.09 - samples/sec: 2995.38 - lr: 0.000030 - momentum: 0.000000
2023-10-23 15:09:40,151 epoch 5 - iter 104/138 - loss 0.04299450 - time (sec): 5.81 - samples/sec: 2975.06 - lr: 0.000029 - momentum: 0.000000
2023-10-23 15:09:40,878 epoch 5 - iter 117/138 - loss 0.04191106 - time (sec): 6.54 - samples/sec: 2967.52 - lr: 0.000029 - momentum: 0.000000
2023-10-23 15:09:41,613 epoch 5 - iter 130/138 - loss 0.04395596 - time (sec): 7.27 - samples/sec: 2986.45 - lr: 0.000028 - momentum: 0.000000
2023-10-23 15:09:42,057 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:42,057 EPOCH 5 done: loss 0.0433 - lr: 0.000028
2023-10-23 15:09:42,593 DEV : loss 0.13214845955371857 - f1-score (micro avg) 0.8717
2023-10-23 15:09:42,599 saving best model
2023-10-23 15:09:43,130 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:43,866 epoch 6 - iter 13/138 - loss 0.04858694 - time (sec): 0.73 - samples/sec: 3252.27 - lr: 0.000027 - momentum: 0.000000
2023-10-23 15:09:44,604 epoch 6 - iter 26/138 - loss 0.03040304 - time (sec): 1.47 - samples/sec: 3209.63 - lr: 0.000027 - momentum: 0.000000
2023-10-23 15:09:45,327 epoch 6 - iter 39/138 - loss 0.04288248 - time (sec): 2.19 - samples/sec: 3173.55 - lr: 0.000026 - momentum: 0.000000
2023-10-23 15:09:46,037 epoch 6 - iter 52/138 - loss 0.03874496 - time (sec): 2.90 - samples/sec: 2985.77 - lr: 0.000026 - momentum: 0.000000
2023-10-23 15:09:46,758 epoch 6 - iter 65/138 - loss 0.03459841 - time (sec): 3.63 - samples/sec: 2977.06 - lr: 0.000025 - momentum: 0.000000
2023-10-23 15:09:47,483 epoch 6 - iter 78/138 - loss 0.03051650 - time (sec): 4.35 - samples/sec: 3023.41 - lr: 0.000025 - momentum: 0.000000
2023-10-23 15:09:48,199 epoch 6 - iter 91/138 - loss 0.03324664 - time (sec): 5.07 - samples/sec: 3042.35 - lr: 0.000024 - momentum: 0.000000
2023-10-23 15:09:48,924 epoch 6 - iter 104/138 - loss 0.02975550 - time (sec): 5.79 - samples/sec: 3034.95 - lr: 0.000024 - momentum: 0.000000
2023-10-23 15:09:49,650 epoch 6 - iter 117/138 - loss 0.02891134 - time (sec): 6.52 - samples/sec: 3013.07 - lr: 0.000023 - momentum: 0.000000
2023-10-23 15:09:50,388 epoch 6 - iter 130/138 - loss 0.02886668 - time (sec): 7.26 - samples/sec: 3003.23 - lr: 0.000023 - momentum: 0.000000
2023-10-23 15:09:50,823 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:50,823 EPOCH 6 done: loss 0.0283 - lr: 0.000023
2023-10-23 15:09:51,359 DEV : loss 0.15722046792507172 - f1-score (micro avg) 0.8843
2023-10-23 15:09:51,365 saving best model
2023-10-23 15:09:51,903 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:52,637 epoch 7 - iter 13/138 - loss 0.02399450 - time (sec): 0.73 - samples/sec: 2634.31 - lr: 0.000022 - momentum: 0.000000
2023-10-23 15:09:53,387 epoch 7 - iter 26/138 - loss 0.01265243 - time (sec): 1.48 - samples/sec: 2845.36 - lr: 0.000021 - momentum: 0.000000
2023-10-23 15:09:54,122 epoch 7 - iter 39/138 - loss 0.01612770 - time (sec): 2.22 - samples/sec: 2921.04 - lr: 0.000021 - momentum: 0.000000
2023-10-23 15:09:54,837 epoch 7 - iter 52/138 - loss 0.02357778 - time (sec): 2.93 - samples/sec: 2882.53 - lr: 0.000020 - momentum: 0.000000
2023-10-23 15:09:55,565 epoch 7 - iter 65/138 - loss 0.02038982 - time (sec): 3.66 - samples/sec: 2980.57 - lr: 0.000020 - momentum: 0.000000
2023-10-23 15:09:56,302 epoch 7 - iter 78/138 - loss 0.02129536 - time (sec): 4.40 - samples/sec: 2976.73 - lr: 0.000019 - momentum: 0.000000
2023-10-23 15:09:57,021 epoch 7 - iter 91/138 - loss 0.01919314 - time (sec): 5.11 - samples/sec: 2954.22 - lr: 0.000019 - momentum: 0.000000
2023-10-23 15:09:57,758 epoch 7 - iter 104/138 - loss 0.02367637 - time (sec): 5.85 - samples/sec: 2975.61 - lr: 0.000018 - momentum: 0.000000
2023-10-23 15:09:58,482 epoch 7 - iter 117/138 - loss 0.02211340 - time (sec): 6.57 - samples/sec: 2993.43 - lr: 0.000018 - momentum: 0.000000
2023-10-23 15:09:59,196 epoch 7 - iter 130/138 - loss 0.02251227 - time (sec): 7.29 - samples/sec: 2967.52 - lr: 0.000017 - momentum: 0.000000
2023-10-23 15:09:59,625 ----------------------------------------------------------------------------------------------------
2023-10-23 15:09:59,625 EPOCH 7 done: loss 0.0223 - lr: 0.000017
2023-10-23 15:10:00,164 DEV : loss 0.17066876590251923 - f1-score (micro avg) 0.8961
2023-10-23 15:10:00,170 saving best model
2023-10-23 15:10:00,706 ----------------------------------------------------------------------------------------------------
2023-10-23 15:10:01,456 epoch 8 - iter 13/138 - loss 0.00845146 - time (sec): 0.75 - samples/sec: 2999.37 - lr: 0.000016 - momentum: 0.000000
2023-10-23 15:10:02,198 epoch 8 - iter 26/138 - loss 0.01002596 - time (sec): 1.49 - samples/sec: 3004.22 - lr: 0.000016 - momentum: 0.000000
2023-10-23 15:10:02,953 epoch 8 - iter 39/138 - loss 0.01643886 - time (sec): 2.24 - samples/sec: 2893.18 - lr: 0.000015 - momentum: 0.000000
2023-10-23 15:10:03,725 epoch 8 - iter 52/138 - loss 0.01342550 - time (sec): 3.02 - samples/sec: 2885.82 - lr: 0.000015 - momentum: 0.000000
2023-10-23 15:10:04,471 epoch 8 - iter 65/138 - loss 0.01584849 - time (sec): 3.76 - samples/sec: 2906.44 - lr: 0.000014 - momentum: 0.000000
2023-10-23 15:10:05,192 epoch 8 - iter 78/138 - loss 0.01788111 - time (sec): 4.48 - samples/sec: 2945.74 - lr: 0.000014 - momentum: 0.000000
2023-10-23 15:10:05,918 epoch 8 - iter 91/138 - loss 0.01544449 - time (sec): 5.21 - samples/sec: 2977.28 - lr: 0.000013 - momentum: 0.000000
2023-10-23 15:10:06,651 epoch 8 - iter 104/138 - loss 0.01385370 - time (sec): 5.94 - samples/sec: 2938.58 - lr: 0.000013 - momentum: 0.000000
2023-10-23 15:10:07,380 epoch 8 - iter 117/138 - loss 0.01384517 - time (sec): 6.67 - samples/sec: 2917.16 - lr: 0.000012 - momentum: 0.000000
2023-10-23 15:10:08,133 epoch 8 - iter 130/138 - loss 0.01338048 - time (sec): 7.42 - samples/sec: 2919.81 - lr: 0.000012 - momentum: 0.000000
2023-10-23 15:10:08,603 ----------------------------------------------------------------------------------------------------
2023-10-23 15:10:08,603 EPOCH 8 done: loss 0.0157 - lr: 0.000012
2023-10-23 15:10:09,143 DEV : loss 0.16071170568466187 - f1-score (micro avg) 0.8942
2023-10-23 15:10:09,148 ----------------------------------------------------------------------------------------------------
2023-10-23 15:10:09,892 epoch 9 - iter 13/138 - loss 0.00315447 - time (sec): 0.74 - samples/sec: 2571.89 - lr: 0.000011 - momentum: 0.000000
2023-10-23 15:10:10,619 epoch 9 - iter 26/138 - loss 0.00728140 - time (sec): 1.47 - samples/sec: 2837.31 - lr: 0.000010 - momentum: 0.000000
2023-10-23 15:10:11,367 epoch 9 - iter 39/138 - loss 0.00488290 - time (sec): 2.22 - samples/sec: 2920.25 - lr: 0.000010 - momentum: 0.000000
2023-10-23 15:10:12,098 epoch 9 - iter 52/138 - loss 0.00382886 - time (sec): 2.95 - samples/sec: 2844.34 - lr: 0.000009 - momentum: 0.000000
2023-10-23 15:10:12,813 epoch 9 - iter 65/138 - loss 0.00530259 - time (sec): 3.66 - samples/sec: 2875.94 - lr: 0.000009 - momentum: 0.000000
2023-10-23 15:10:13,547 epoch 9 - iter 78/138 - loss 0.00503857 - time (sec): 4.40 - samples/sec: 2891.82 - lr: 0.000008 - momentum: 0.000000
2023-10-23 15:10:14,281 epoch 9 - iter 91/138 - loss 0.00434295 - time (sec): 5.13 - samples/sec: 2916.36 - lr: 0.000008 - momentum: 0.000000
2023-10-23 15:10:15,018 epoch 9 - iter 104/138 - loss 0.00705956 - time (sec): 5.87 - samples/sec: 2923.38 - lr: 0.000007 - momentum: 0.000000
2023-10-23 15:10:15,738 epoch 9 - iter 117/138 - loss 0.00942999 - time (sec): 6.59 - samples/sec: 2921.28 - lr: 0.000007 - momentum: 0.000000
2023-10-23 15:10:16,469 epoch 9 - iter 130/138 - loss 0.00981395 - time (sec): 7.32 - samples/sec: 2941.85 - lr: 0.000006 - momentum: 0.000000
2023-10-23 15:10:16,929 ----------------------------------------------------------------------------------------------------
2023-10-23 15:10:16,929 EPOCH 9 done: loss 0.0108 - lr: 0.000006
2023-10-23 15:10:17,469 DEV : loss 0.1666877120733261 - f1-score (micro avg) 0.8972
2023-10-23 15:10:17,475 saving best model
2023-10-23 15:10:18,012 ----------------------------------------------------------------------------------------------------
2023-10-23 15:10:18,766 epoch 10 - iter 13/138 - loss 0.00901088 - time (sec): 0.75 - samples/sec: 2829.11 - lr: 0.000005 - momentum: 0.000000
2023-10-23 15:10:19,511 epoch 10 - iter 26/138 - loss 0.00467408 - time (sec): 1.49 - samples/sec: 2764.82 - lr: 0.000005 - momentum: 0.000000
2023-10-23 15:10:20,284 epoch 10 - iter 39/138 - loss 0.00779833 - time (sec): 2.26 - samples/sec: 2815.40 - lr: 0.000004 - momentum: 0.000000
2023-10-23 15:10:21,075 epoch 10 - iter 52/138 - loss 0.00602233 - time (sec): 3.06 - samples/sec: 2790.53 - lr: 0.000004 - momentum: 0.000000
2023-10-23 15:10:21,857 epoch 10 - iter 65/138 - loss 0.00598930 - time (sec): 3.84 - samples/sec: 2759.35 - lr: 0.000003 - momentum: 0.000000
2023-10-23 15:10:22,596 epoch 10 - iter 78/138 - loss 0.00629598 - time (sec): 4.58 - samples/sec: 2871.17 - lr: 0.000003 - momentum: 0.000000
2023-10-23 15:10:23,361 epoch 10 - iter 91/138 - loss 0.00709990 - time (sec): 5.34 - samples/sec: 2935.94 - lr: 0.000002 - momentum: 0.000000
2023-10-23 15:10:24,089 epoch 10 - iter 104/138 - loss 0.00705259 - time (sec): 6.07 - samples/sec: 2940.51 - lr: 0.000002 - momentum: 0.000000
2023-10-23 15:10:24,821 epoch 10 - iter 117/138 - loss 0.00652926 - time (sec): 6.80 - samples/sec: 2893.97 - lr: 0.000001 - momentum: 0.000000
2023-10-23 15:10:25,588 epoch 10 - iter 130/138 - loss 0.00843776 - time (sec): 7.57 - samples/sec: 2861.39 - lr: 0.000000 - momentum: 0.000000
2023-10-23 15:10:26,048 ----------------------------------------------------------------------------------------------------
2023-10-23 15:10:26,048 EPOCH 10 done: loss 0.0082 - lr: 0.000000
2023-10-23 15:10:26,585 DEV : loss 0.1702803522348404 - f1-score (micro avg) 0.8951
2023-10-23 15:10:26,997 ----------------------------------------------------------------------------------------------------
2023-10-23 15:10:26,998 Loading model from best epoch ...
2023-10-23 15:10:28,749 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
2023-10-23 15:10:29,406
Results:
- F-score (micro) 0.9194
- F-score (macro) 0.8501
- Accuracy 0.8593
By class:
precision recall f1-score support
scope 0.9023 0.8920 0.8971 176
pers 0.9921 0.9766 0.9843 128
work 0.8873 0.8514 0.8690 74
object 1.0000 1.0000 1.0000 2
loc 0.5000 0.5000 0.5000 2
micro avg 0.9280 0.9110 0.9194 382
macro avg 0.8563 0.8440 0.8501 382
weighted avg 0.9279 0.9110 0.9193 382
2023-10-23 15:10:29,407 ----------------------------------------------------------------------------------------------------
|