File size: 24,937 Bytes
002f360 |
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 |
2023-10-08 23:34:23,888 ----------------------------------------------------------------------------------------------------
2023-10-08 23:34:23,890 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()
)"
2023-10-08 23:34:23,890 ----------------------------------------------------------------------------------------------------
2023-10-08 23:34:23,890 MultiCorpus: 966 train + 219 dev + 204 test sentences
- NER_HIPE_2022 Corpus: 966 train + 219 dev + 204 test sentences - /app/.flair/datasets/ner_hipe_2022/v2.1/ajmc/fr/with_doc_seperator
2023-10-08 23:34:23,890 ----------------------------------------------------------------------------------------------------
2023-10-08 23:34:23,890 Train: 966 sentences
2023-10-08 23:34:23,890 (train_with_dev=False, train_with_test=False)
2023-10-08 23:34:23,890 ----------------------------------------------------------------------------------------------------
2023-10-08 23:34:23,890 Training Params:
2023-10-08 23:34:23,890 - learning_rate: "0.00015"
2023-10-08 23:34:23,890 - mini_batch_size: "8"
2023-10-08 23:34:23,890 - max_epochs: "10"
2023-10-08 23:34:23,890 - shuffle: "True"
2023-10-08 23:34:23,890 ----------------------------------------------------------------------------------------------------
2023-10-08 23:34:23,890 Plugins:
2023-10-08 23:34:23,891 - TensorboardLogger
2023-10-08 23:34:23,891 - LinearScheduler | warmup_fraction: '0.1'
2023-10-08 23:34:23,891 ----------------------------------------------------------------------------------------------------
2023-10-08 23:34:23,891 Final evaluation on model from best epoch (best-model.pt)
2023-10-08 23:34:23,891 - metric: "('micro avg', 'f1-score')"
2023-10-08 23:34:23,891 ----------------------------------------------------------------------------------------------------
2023-10-08 23:34:23,891 Computation:
2023-10-08 23:34:23,891 - compute on device: cuda:0
2023-10-08 23:34:23,891 - embedding storage: none
2023-10-08 23:34:23,891 ----------------------------------------------------------------------------------------------------
2023-10-08 23:34:23,891 Model training base path: "hmbench-ajmc/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs8-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-5"
2023-10-08 23:34:23,891 ----------------------------------------------------------------------------------------------------
2023-10-08 23:34:23,891 ----------------------------------------------------------------------------------------------------
2023-10-08 23:34:23,891 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-08 23:34:32,778 epoch 1 - iter 12/121 - loss 3.24017889 - time (sec): 8.89 - samples/sec: 258.16 - lr: 0.000014 - momentum: 0.000000
2023-10-08 23:34:42,632 epoch 1 - iter 24/121 - loss 3.23366481 - time (sec): 18.74 - samples/sec: 269.53 - lr: 0.000029 - momentum: 0.000000
2023-10-08 23:34:52,336 epoch 1 - iter 36/121 - loss 3.22367203 - time (sec): 28.44 - samples/sec: 271.42 - lr: 0.000043 - momentum: 0.000000
2023-10-08 23:35:01,746 epoch 1 - iter 48/121 - loss 3.20730354 - time (sec): 37.85 - samples/sec: 266.79 - lr: 0.000058 - momentum: 0.000000
2023-10-08 23:35:11,211 epoch 1 - iter 60/121 - loss 3.17435023 - time (sec): 47.32 - samples/sec: 268.14 - lr: 0.000073 - momentum: 0.000000
2023-10-08 23:35:20,140 epoch 1 - iter 72/121 - loss 3.12585276 - time (sec): 56.25 - samples/sec: 266.78 - lr: 0.000088 - momentum: 0.000000
2023-10-08 23:35:29,472 epoch 1 - iter 84/121 - loss 3.05911175 - time (sec): 65.58 - samples/sec: 268.07 - lr: 0.000103 - momentum: 0.000000
2023-10-08 23:35:38,423 epoch 1 - iter 96/121 - loss 2.98937366 - time (sec): 74.53 - samples/sec: 267.14 - lr: 0.000118 - momentum: 0.000000
2023-10-08 23:35:47,528 epoch 1 - iter 108/121 - loss 2.91543096 - time (sec): 83.64 - samples/sec: 265.16 - lr: 0.000133 - momentum: 0.000000
2023-10-08 23:35:56,942 epoch 1 - iter 120/121 - loss 2.83441714 - time (sec): 93.05 - samples/sec: 264.28 - lr: 0.000148 - momentum: 0.000000
2023-10-08 23:35:57,594 ----------------------------------------------------------------------------------------------------
2023-10-08 23:35:57,594 EPOCH 1 done: loss 2.8287 - lr: 0.000148
2023-10-08 23:36:03,894 DEV : loss 1.865909457206726 - f1-score (micro avg) 0.0
2023-10-08 23:36:03,899 ----------------------------------------------------------------------------------------------------
2023-10-08 23:36:13,707 epoch 2 - iter 12/121 - loss 1.82988856 - time (sec): 9.81 - samples/sec: 252.40 - lr: 0.000148 - momentum: 0.000000
2023-10-08 23:36:23,287 epoch 2 - iter 24/121 - loss 1.70061483 - time (sec): 19.39 - samples/sec: 260.65 - lr: 0.000147 - momentum: 0.000000
2023-10-08 23:36:32,140 epoch 2 - iter 36/121 - loss 1.62437798 - time (sec): 28.24 - samples/sec: 257.05 - lr: 0.000145 - momentum: 0.000000
2023-10-08 23:36:41,051 epoch 2 - iter 48/121 - loss 1.52599380 - time (sec): 37.15 - samples/sec: 257.31 - lr: 0.000144 - momentum: 0.000000
2023-10-08 23:36:50,363 epoch 2 - iter 60/121 - loss 1.43660909 - time (sec): 46.46 - samples/sec: 255.80 - lr: 0.000142 - momentum: 0.000000
2023-10-08 23:36:59,592 epoch 2 - iter 72/121 - loss 1.37023102 - time (sec): 55.69 - samples/sec: 255.46 - lr: 0.000140 - momentum: 0.000000
2023-10-08 23:37:08,394 epoch 2 - iter 84/121 - loss 1.30900136 - time (sec): 64.49 - samples/sec: 255.16 - lr: 0.000139 - momentum: 0.000000
2023-10-08 23:37:17,797 epoch 2 - iter 96/121 - loss 1.23447583 - time (sec): 73.90 - samples/sec: 256.89 - lr: 0.000137 - momentum: 0.000000
2023-10-08 23:37:27,844 epoch 2 - iter 108/121 - loss 1.16765337 - time (sec): 83.94 - samples/sec: 258.99 - lr: 0.000135 - momentum: 0.000000
2023-10-08 23:37:37,761 epoch 2 - iter 120/121 - loss 1.11224056 - time (sec): 93.86 - samples/sec: 262.29 - lr: 0.000134 - momentum: 0.000000
2023-10-08 23:37:38,273 ----------------------------------------------------------------------------------------------------
2023-10-08 23:37:38,273 EPOCH 2 done: loss 1.1093 - lr: 0.000134
2023-10-08 23:37:44,716 DEV : loss 0.6599521636962891 - f1-score (micro avg) 0.0
2023-10-08 23:37:44,723 ----------------------------------------------------------------------------------------------------
2023-10-08 23:37:53,805 epoch 3 - iter 12/121 - loss 0.66795242 - time (sec): 9.08 - samples/sec: 258.00 - lr: 0.000132 - momentum: 0.000000
2023-10-08 23:38:03,763 epoch 3 - iter 24/121 - loss 0.57765494 - time (sec): 19.04 - samples/sec: 267.13 - lr: 0.000130 - momentum: 0.000000
2023-10-08 23:38:13,177 epoch 3 - iter 36/121 - loss 0.58939285 - time (sec): 28.45 - samples/sec: 265.13 - lr: 0.000129 - momentum: 0.000000
2023-10-08 23:38:22,673 epoch 3 - iter 48/121 - loss 0.59008148 - time (sec): 37.95 - samples/sec: 266.23 - lr: 0.000127 - momentum: 0.000000
2023-10-08 23:38:32,149 epoch 3 - iter 60/121 - loss 0.59041910 - time (sec): 47.43 - samples/sec: 265.96 - lr: 0.000125 - momentum: 0.000000
2023-10-08 23:38:41,186 epoch 3 - iter 72/121 - loss 0.59060120 - time (sec): 56.46 - samples/sec: 263.95 - lr: 0.000124 - momentum: 0.000000
2023-10-08 23:38:50,402 epoch 3 - iter 84/121 - loss 0.56699050 - time (sec): 65.68 - samples/sec: 262.80 - lr: 0.000122 - momentum: 0.000000
2023-10-08 23:39:00,139 epoch 3 - iter 96/121 - loss 0.54734552 - time (sec): 75.42 - samples/sec: 263.58 - lr: 0.000120 - momentum: 0.000000
2023-10-08 23:39:08,891 epoch 3 - iter 108/121 - loss 0.53838536 - time (sec): 84.17 - samples/sec: 262.33 - lr: 0.000119 - momentum: 0.000000
2023-10-08 23:39:18,361 epoch 3 - iter 120/121 - loss 0.52998391 - time (sec): 93.64 - samples/sec: 263.15 - lr: 0.000117 - momentum: 0.000000
2023-10-08 23:39:18,899 ----------------------------------------------------------------------------------------------------
2023-10-08 23:39:18,899 EPOCH 3 done: loss 0.5291 - lr: 0.000117
2023-10-08 23:39:25,550 DEV : loss 0.3969781696796417 - f1-score (micro avg) 0.0
2023-10-08 23:39:25,557 ----------------------------------------------------------------------------------------------------
2023-10-08 23:39:35,472 epoch 4 - iter 12/121 - loss 0.38272330 - time (sec): 9.91 - samples/sec: 274.49 - lr: 0.000115 - momentum: 0.000000
2023-10-08 23:39:45,398 epoch 4 - iter 24/121 - loss 0.39630310 - time (sec): 19.84 - samples/sec: 273.45 - lr: 0.000114 - momentum: 0.000000
2023-10-08 23:39:54,512 epoch 4 - iter 36/121 - loss 0.36093487 - time (sec): 28.95 - samples/sec: 268.09 - lr: 0.000112 - momentum: 0.000000
2023-10-08 23:40:04,530 epoch 4 - iter 48/121 - loss 0.35342633 - time (sec): 38.97 - samples/sec: 266.02 - lr: 0.000110 - momentum: 0.000000
2023-10-08 23:40:14,229 epoch 4 - iter 60/121 - loss 0.34983137 - time (sec): 48.67 - samples/sec: 265.22 - lr: 0.000109 - momentum: 0.000000
2023-10-08 23:40:23,213 epoch 4 - iter 72/121 - loss 0.35500506 - time (sec): 57.65 - samples/sec: 263.88 - lr: 0.000107 - momentum: 0.000000
2023-10-08 23:40:31,803 epoch 4 - iter 84/121 - loss 0.34761176 - time (sec): 66.24 - samples/sec: 262.39 - lr: 0.000105 - momentum: 0.000000
2023-10-08 23:40:41,187 epoch 4 - iter 96/121 - loss 0.34047462 - time (sec): 75.63 - samples/sec: 262.83 - lr: 0.000104 - momentum: 0.000000
2023-10-08 23:40:50,246 epoch 4 - iter 108/121 - loss 0.33251375 - time (sec): 84.69 - samples/sec: 262.11 - lr: 0.000102 - momentum: 0.000000
2023-10-08 23:40:59,213 epoch 4 - iter 120/121 - loss 0.32254317 - time (sec): 93.65 - samples/sec: 262.27 - lr: 0.000101 - momentum: 0.000000
2023-10-08 23:40:59,856 ----------------------------------------------------------------------------------------------------
2023-10-08 23:40:59,857 EPOCH 4 done: loss 0.3228 - lr: 0.000101
2023-10-08 23:41:06,377 DEV : loss 0.2733049690723419 - f1-score (micro avg) 0.4662
2023-10-08 23:41:06,385 saving best model
2023-10-08 23:41:07,265 ----------------------------------------------------------------------------------------------------
2023-10-08 23:41:16,917 epoch 5 - iter 12/121 - loss 0.28853168 - time (sec): 9.65 - samples/sec: 263.94 - lr: 0.000099 - momentum: 0.000000
2023-10-08 23:41:25,961 epoch 5 - iter 24/121 - loss 0.23595621 - time (sec): 18.69 - samples/sec: 260.84 - lr: 0.000097 - momentum: 0.000000
2023-10-08 23:41:35,553 epoch 5 - iter 36/121 - loss 0.23653503 - time (sec): 28.29 - samples/sec: 258.97 - lr: 0.000095 - momentum: 0.000000
2023-10-08 23:41:44,553 epoch 5 - iter 48/121 - loss 0.22403941 - time (sec): 37.29 - samples/sec: 259.64 - lr: 0.000094 - momentum: 0.000000
2023-10-08 23:41:53,779 epoch 5 - iter 60/121 - loss 0.22717697 - time (sec): 46.51 - samples/sec: 260.49 - lr: 0.000092 - momentum: 0.000000
2023-10-08 23:42:02,805 epoch 5 - iter 72/121 - loss 0.23024853 - time (sec): 55.54 - samples/sec: 259.80 - lr: 0.000091 - momentum: 0.000000
2023-10-08 23:42:12,086 epoch 5 - iter 84/121 - loss 0.23713059 - time (sec): 64.82 - samples/sec: 259.99 - lr: 0.000089 - momentum: 0.000000
2023-10-08 23:42:21,895 epoch 5 - iter 96/121 - loss 0.23487341 - time (sec): 74.63 - samples/sec: 261.15 - lr: 0.000087 - momentum: 0.000000
2023-10-08 23:42:31,641 epoch 5 - iter 108/121 - loss 0.23401748 - time (sec): 84.37 - samples/sec: 261.88 - lr: 0.000086 - momentum: 0.000000
2023-10-08 23:42:40,945 epoch 5 - iter 120/121 - loss 0.22746030 - time (sec): 93.68 - samples/sec: 261.81 - lr: 0.000084 - momentum: 0.000000
2023-10-08 23:42:41,671 ----------------------------------------------------------------------------------------------------
2023-10-08 23:42:41,671 EPOCH 5 done: loss 0.2270 - lr: 0.000084
2023-10-08 23:42:48,196 DEV : loss 0.21299438178539276 - f1-score (micro avg) 0.6084
2023-10-08 23:42:48,202 saving best model
2023-10-08 23:42:52,580 ----------------------------------------------------------------------------------------------------
2023-10-08 23:43:01,768 epoch 6 - iter 12/121 - loss 0.20543145 - time (sec): 9.19 - samples/sec: 266.58 - lr: 0.000082 - momentum: 0.000000
2023-10-08 23:43:11,706 epoch 6 - iter 24/121 - loss 0.19076741 - time (sec): 19.12 - samples/sec: 271.32 - lr: 0.000081 - momentum: 0.000000
2023-10-08 23:43:20,980 epoch 6 - iter 36/121 - loss 0.19650215 - time (sec): 28.40 - samples/sec: 269.73 - lr: 0.000079 - momentum: 0.000000
2023-10-08 23:43:30,235 epoch 6 - iter 48/121 - loss 0.18904959 - time (sec): 37.65 - samples/sec: 267.23 - lr: 0.000077 - momentum: 0.000000
2023-10-08 23:43:40,393 epoch 6 - iter 60/121 - loss 0.18069900 - time (sec): 47.81 - samples/sec: 264.75 - lr: 0.000076 - momentum: 0.000000
2023-10-08 23:43:49,509 epoch 6 - iter 72/121 - loss 0.17923457 - time (sec): 56.93 - samples/sec: 267.01 - lr: 0.000074 - momentum: 0.000000
2023-10-08 23:43:59,118 epoch 6 - iter 84/121 - loss 0.17895962 - time (sec): 66.54 - samples/sec: 266.92 - lr: 0.000072 - momentum: 0.000000
2023-10-08 23:44:08,097 epoch 6 - iter 96/121 - loss 0.17558585 - time (sec): 75.52 - samples/sec: 266.34 - lr: 0.000071 - momentum: 0.000000
2023-10-08 23:44:16,813 epoch 6 - iter 108/121 - loss 0.17561514 - time (sec): 84.23 - samples/sec: 264.48 - lr: 0.000069 - momentum: 0.000000
2023-10-08 23:44:26,016 epoch 6 - iter 120/121 - loss 0.17486140 - time (sec): 93.43 - samples/sec: 263.55 - lr: 0.000067 - momentum: 0.000000
2023-10-08 23:44:26,515 ----------------------------------------------------------------------------------------------------
2023-10-08 23:44:26,516 EPOCH 6 done: loss 0.1750 - lr: 0.000067
2023-10-08 23:44:33,002 DEV : loss 0.1774875372648239 - f1-score (micro avg) 0.7266
2023-10-08 23:44:33,008 saving best model
2023-10-08 23:44:37,887 ----------------------------------------------------------------------------------------------------
2023-10-08 23:44:46,853 epoch 7 - iter 12/121 - loss 0.15820431 - time (sec): 8.97 - samples/sec: 262.34 - lr: 0.000066 - momentum: 0.000000
2023-10-08 23:44:55,909 epoch 7 - iter 24/121 - loss 0.16596847 - time (sec): 18.02 - samples/sec: 259.25 - lr: 0.000064 - momentum: 0.000000
2023-10-08 23:45:04,694 epoch 7 - iter 36/121 - loss 0.15533406 - time (sec): 26.81 - samples/sec: 258.67 - lr: 0.000062 - momentum: 0.000000
2023-10-08 23:45:14,414 epoch 7 - iter 48/121 - loss 0.14849612 - time (sec): 36.53 - samples/sec: 262.25 - lr: 0.000061 - momentum: 0.000000
2023-10-08 23:45:23,620 epoch 7 - iter 60/121 - loss 0.14917696 - time (sec): 45.73 - samples/sec: 264.06 - lr: 0.000059 - momentum: 0.000000
2023-10-08 23:45:33,084 epoch 7 - iter 72/121 - loss 0.14147734 - time (sec): 55.20 - samples/sec: 263.79 - lr: 0.000057 - momentum: 0.000000
2023-10-08 23:45:42,561 epoch 7 - iter 84/121 - loss 0.13644284 - time (sec): 64.67 - samples/sec: 262.09 - lr: 0.000056 - momentum: 0.000000
2023-10-08 23:45:52,135 epoch 7 - iter 96/121 - loss 0.13506282 - time (sec): 74.25 - samples/sec: 263.18 - lr: 0.000054 - momentum: 0.000000
2023-10-08 23:46:02,002 epoch 7 - iter 108/121 - loss 0.13473976 - time (sec): 84.11 - samples/sec: 262.94 - lr: 0.000052 - momentum: 0.000000
2023-10-08 23:46:11,175 epoch 7 - iter 120/121 - loss 0.13772286 - time (sec): 93.29 - samples/sec: 263.34 - lr: 0.000051 - momentum: 0.000000
2023-10-08 23:46:11,785 ----------------------------------------------------------------------------------------------------
2023-10-08 23:46:11,786 EPOCH 7 done: loss 0.1373 - lr: 0.000051
2023-10-08 23:46:18,296 DEV : loss 0.15338008105754852 - f1-score (micro avg) 0.842
2023-10-08 23:46:18,302 saving best model
2023-10-08 23:46:22,671 ----------------------------------------------------------------------------------------------------
2023-10-08 23:46:32,037 epoch 8 - iter 12/121 - loss 0.11148768 - time (sec): 9.36 - samples/sec: 262.38 - lr: 0.000049 - momentum: 0.000000
2023-10-08 23:46:41,192 epoch 8 - iter 24/121 - loss 0.11910156 - time (sec): 18.52 - samples/sec: 262.97 - lr: 0.000047 - momentum: 0.000000
2023-10-08 23:46:50,425 epoch 8 - iter 36/121 - loss 0.12923502 - time (sec): 27.75 - samples/sec: 261.49 - lr: 0.000046 - momentum: 0.000000
2023-10-08 23:46:59,623 epoch 8 - iter 48/121 - loss 0.12796695 - time (sec): 36.95 - samples/sec: 262.16 - lr: 0.000044 - momentum: 0.000000
2023-10-08 23:47:09,242 epoch 8 - iter 60/121 - loss 0.12207759 - time (sec): 46.57 - samples/sec: 263.01 - lr: 0.000042 - momentum: 0.000000
2023-10-08 23:47:18,465 epoch 8 - iter 72/121 - loss 0.11942277 - time (sec): 55.79 - samples/sec: 263.50 - lr: 0.000041 - momentum: 0.000000
2023-10-08 23:47:28,252 epoch 8 - iter 84/121 - loss 0.11435017 - time (sec): 65.58 - samples/sec: 263.90 - lr: 0.000039 - momentum: 0.000000
2023-10-08 23:47:38,022 epoch 8 - iter 96/121 - loss 0.11382818 - time (sec): 75.35 - samples/sec: 264.10 - lr: 0.000038 - momentum: 0.000000
2023-10-08 23:47:47,763 epoch 8 - iter 108/121 - loss 0.11198754 - time (sec): 85.09 - samples/sec: 263.20 - lr: 0.000036 - momentum: 0.000000
2023-10-08 23:47:56,645 epoch 8 - iter 120/121 - loss 0.11242524 - time (sec): 93.97 - samples/sec: 261.61 - lr: 0.000034 - momentum: 0.000000
2023-10-08 23:47:57,247 ----------------------------------------------------------------------------------------------------
2023-10-08 23:47:57,248 EPOCH 8 done: loss 0.1128 - lr: 0.000034
2023-10-08 23:48:03,852 DEV : loss 0.14283686876296997 - f1-score (micro avg) 0.8227
2023-10-08 23:48:03,858 ----------------------------------------------------------------------------------------------------
2023-10-08 23:48:13,211 epoch 9 - iter 12/121 - loss 0.10396974 - time (sec): 9.35 - samples/sec: 244.25 - lr: 0.000032 - momentum: 0.000000
2023-10-08 23:48:23,568 epoch 9 - iter 24/121 - loss 0.09710433 - time (sec): 19.71 - samples/sec: 261.11 - lr: 0.000031 - momentum: 0.000000
2023-10-08 23:48:32,954 epoch 9 - iter 36/121 - loss 0.09513445 - time (sec): 29.09 - samples/sec: 262.70 - lr: 0.000029 - momentum: 0.000000
2023-10-08 23:48:42,580 epoch 9 - iter 48/121 - loss 0.09058182 - time (sec): 38.72 - samples/sec: 261.16 - lr: 0.000028 - momentum: 0.000000
2023-10-08 23:48:52,022 epoch 9 - iter 60/121 - loss 0.09108372 - time (sec): 48.16 - samples/sec: 260.37 - lr: 0.000026 - momentum: 0.000000
2023-10-08 23:49:01,360 epoch 9 - iter 72/121 - loss 0.09190391 - time (sec): 57.50 - samples/sec: 261.56 - lr: 0.000024 - momentum: 0.000000
2023-10-08 23:49:10,170 epoch 9 - iter 84/121 - loss 0.09588537 - time (sec): 66.31 - samples/sec: 260.53 - lr: 0.000023 - momentum: 0.000000
2023-10-08 23:49:19,362 epoch 9 - iter 96/121 - loss 0.09700172 - time (sec): 75.50 - samples/sec: 259.33 - lr: 0.000021 - momentum: 0.000000
2023-10-08 23:49:28,627 epoch 9 - iter 108/121 - loss 0.09918562 - time (sec): 84.77 - samples/sec: 259.28 - lr: 0.000019 - momentum: 0.000000
2023-10-08 23:49:38,372 epoch 9 - iter 120/121 - loss 0.10076070 - time (sec): 94.51 - samples/sec: 260.41 - lr: 0.000018 - momentum: 0.000000
2023-10-08 23:49:38,893 ----------------------------------------------------------------------------------------------------
2023-10-08 23:49:38,893 EPOCH 9 done: loss 0.1003 - lr: 0.000018
2023-10-08 23:49:45,492 DEV : loss 0.1380492001771927 - f1-score (micro avg) 0.8354
2023-10-08 23:49:45,498 ----------------------------------------------------------------------------------------------------
2023-10-08 23:49:54,373 epoch 10 - iter 12/121 - loss 0.10280380 - time (sec): 8.87 - samples/sec: 260.00 - lr: 0.000016 - momentum: 0.000000
2023-10-08 23:50:03,805 epoch 10 - iter 24/121 - loss 0.09500688 - time (sec): 18.30 - samples/sec: 262.66 - lr: 0.000014 - momentum: 0.000000
2023-10-08 23:50:13,553 epoch 10 - iter 36/121 - loss 0.09699027 - time (sec): 28.05 - samples/sec: 263.07 - lr: 0.000013 - momentum: 0.000000
2023-10-08 23:50:23,171 epoch 10 - iter 48/121 - loss 0.09566332 - time (sec): 37.67 - samples/sec: 261.36 - lr: 0.000011 - momentum: 0.000000
2023-10-08 23:50:32,652 epoch 10 - iter 60/121 - loss 0.09384200 - time (sec): 47.15 - samples/sec: 262.26 - lr: 0.000009 - momentum: 0.000000
2023-10-08 23:50:42,319 epoch 10 - iter 72/121 - loss 0.09581268 - time (sec): 56.82 - samples/sec: 263.40 - lr: 0.000008 - momentum: 0.000000
2023-10-08 23:50:51,898 epoch 10 - iter 84/121 - loss 0.09664381 - time (sec): 66.40 - samples/sec: 263.82 - lr: 0.000006 - momentum: 0.000000
2023-10-08 23:51:00,278 epoch 10 - iter 96/121 - loss 0.09499539 - time (sec): 74.78 - samples/sec: 261.96 - lr: 0.000004 - momentum: 0.000000
2023-10-08 23:51:09,871 epoch 10 - iter 108/121 - loss 0.09536038 - time (sec): 84.37 - samples/sec: 260.50 - lr: 0.000003 - momentum: 0.000000
2023-10-08 23:51:19,558 epoch 10 - iter 120/121 - loss 0.09302861 - time (sec): 94.06 - samples/sec: 261.15 - lr: 0.000001 - momentum: 0.000000
2023-10-08 23:51:20,206 ----------------------------------------------------------------------------------------------------
2023-10-08 23:51:20,207 EPOCH 10 done: loss 0.0931 - lr: 0.000001
2023-10-08 23:51:26,721 DEV : loss 0.1366628259420395 - f1-score (micro avg) 0.8307
2023-10-08 23:51:27,596 ----------------------------------------------------------------------------------------------------
2023-10-08 23:51:27,597 Loading model from best epoch ...
2023-10-08 23:51:30,275 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
2023-10-08 23:51:36,836
Results:
- F-score (micro) 0.7862
- F-score (macro) 0.4704
- Accuracy 0.682
By class:
precision recall f1-score support
pers 0.8069 0.8417 0.8239 139
scope 0.7881 0.9225 0.8500 129
work 0.6186 0.7500 0.6780 80
loc 0.0000 0.0000 0.0000 9
date 0.0000 0.0000 0.0000 3
micro avg 0.7532 0.8222 0.7862 360
macro avg 0.4427 0.5028 0.4704 360
weighted avg 0.7314 0.8222 0.7734 360
2023-10-08 23:51:36,836 ----------------------------------------------------------------------------------------------------
|