File size: 25,131 Bytes
d3ddcdc |
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 |
2023-10-08 19:26:47,941 ----------------------------------------------------------------------------------------------------
2023-10-08 19:26:47,942 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 19:26:47,942 ----------------------------------------------------------------------------------------------------
2023-10-08 19:26:47,942 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 19:26:47,943 ----------------------------------------------------------------------------------------------------
2023-10-08 19:26:47,943 Train: 966 sentences
2023-10-08 19:26:47,943 (train_with_dev=False, train_with_test=False)
2023-10-08 19:26:47,943 ----------------------------------------------------------------------------------------------------
2023-10-08 19:26:47,943 Training Params:
2023-10-08 19:26:47,943 - learning_rate: "0.00015"
2023-10-08 19:26:47,943 - mini_batch_size: "4"
2023-10-08 19:26:47,943 - max_epochs: "10"
2023-10-08 19:26:47,943 - shuffle: "True"
2023-10-08 19:26:47,943 ----------------------------------------------------------------------------------------------------
2023-10-08 19:26:47,943 Plugins:
2023-10-08 19:26:47,943 - TensorboardLogger
2023-10-08 19:26:47,943 - LinearScheduler | warmup_fraction: '0.1'
2023-10-08 19:26:47,943 ----------------------------------------------------------------------------------------------------
2023-10-08 19:26:47,943 Final evaluation on model from best epoch (best-model.pt)
2023-10-08 19:26:47,943 - metric: "('micro avg', 'f1-score')"
2023-10-08 19:26:47,943 ----------------------------------------------------------------------------------------------------
2023-10-08 19:26:47,943 Computation:
2023-10-08 19:26:47,943 - compute on device: cuda:0
2023-10-08 19:26:47,943 - embedding storage: none
2023-10-08 19:26:47,944 ----------------------------------------------------------------------------------------------------
2023-10-08 19:26:47,944 Model training base path: "hmbench-ajmc/fr-hmbyt5-preliminary/byt5-small-historic-multilingual-span20-flax-bs4-wsFalse-e10-lr0.00015-poolingfirst-layers-1-crfFalse-2"
2023-10-08 19:26:47,944 ----------------------------------------------------------------------------------------------------
2023-10-08 19:26:47,944 ----------------------------------------------------------------------------------------------------
2023-10-08 19:26:47,944 Logging anything other than scalars to TensorBoard is currently not supported.
2023-10-08 19:26:58,455 epoch 1 - iter 24/242 - loss 3.23018118 - time (sec): 10.51 - samples/sec: 248.15 - lr: 0.000014 - momentum: 0.000000
2023-10-08 19:27:09,462 epoch 1 - iter 48/242 - loss 3.21975537 - time (sec): 21.52 - samples/sec: 251.42 - lr: 0.000029 - momentum: 0.000000
2023-10-08 19:27:19,308 epoch 1 - iter 72/242 - loss 3.20292151 - time (sec): 31.36 - samples/sec: 246.41 - lr: 0.000044 - momentum: 0.000000
2023-10-08 19:27:28,606 epoch 1 - iter 96/242 - loss 3.16766256 - time (sec): 40.66 - samples/sec: 243.55 - lr: 0.000059 - momentum: 0.000000
2023-10-08 19:27:38,504 epoch 1 - iter 120/242 - loss 3.09045996 - time (sec): 50.56 - samples/sec: 242.69 - lr: 0.000074 - momentum: 0.000000
2023-10-08 19:27:48,252 epoch 1 - iter 144/242 - loss 2.99581041 - time (sec): 60.31 - samples/sec: 242.03 - lr: 0.000089 - momentum: 0.000000
2023-10-08 19:27:58,214 epoch 1 - iter 168/242 - loss 2.88863965 - time (sec): 70.27 - samples/sec: 243.22 - lr: 0.000104 - momentum: 0.000000
2023-10-08 19:28:08,332 epoch 1 - iter 192/242 - loss 2.77484047 - time (sec): 80.39 - samples/sec: 243.98 - lr: 0.000118 - momentum: 0.000000
2023-10-08 19:28:18,853 epoch 1 - iter 216/242 - loss 2.64647879 - time (sec): 90.91 - samples/sec: 245.39 - lr: 0.000133 - momentum: 0.000000
2023-10-08 19:28:28,397 epoch 1 - iter 240/242 - loss 2.53515433 - time (sec): 100.45 - samples/sec: 244.51 - lr: 0.000148 - momentum: 0.000000
2023-10-08 19:28:29,065 ----------------------------------------------------------------------------------------------------
2023-10-08 19:28:29,065 EPOCH 1 done: loss 2.5262 - lr: 0.000148
2023-10-08 19:28:35,558 DEV : loss 1.1793770790100098 - f1-score (micro avg) 0.0
2023-10-08 19:28:35,564 ----------------------------------------------------------------------------------------------------
2023-10-08 19:28:45,269 epoch 2 - iter 24/242 - loss 1.12998825 - time (sec): 9.70 - samples/sec: 239.30 - lr: 0.000148 - momentum: 0.000000
2023-10-08 19:28:55,077 epoch 2 - iter 48/242 - loss 1.02715380 - time (sec): 19.51 - samples/sec: 238.73 - lr: 0.000147 - momentum: 0.000000
2023-10-08 19:29:05,508 epoch 2 - iter 72/242 - loss 0.92096229 - time (sec): 29.94 - samples/sec: 241.96 - lr: 0.000145 - momentum: 0.000000
2023-10-08 19:29:15,624 epoch 2 - iter 96/242 - loss 0.85340030 - time (sec): 40.06 - samples/sec: 242.02 - lr: 0.000143 - momentum: 0.000000
2023-10-08 19:29:25,887 epoch 2 - iter 120/242 - loss 0.79456575 - time (sec): 50.32 - samples/sec: 241.19 - lr: 0.000142 - momentum: 0.000000
2023-10-08 19:29:35,994 epoch 2 - iter 144/242 - loss 0.75737283 - time (sec): 60.43 - samples/sec: 243.11 - lr: 0.000140 - momentum: 0.000000
2023-10-08 19:29:45,897 epoch 2 - iter 168/242 - loss 0.73321781 - time (sec): 70.33 - samples/sec: 243.22 - lr: 0.000139 - momentum: 0.000000
2023-10-08 19:29:55,775 epoch 2 - iter 192/242 - loss 0.70288592 - time (sec): 80.21 - samples/sec: 243.35 - lr: 0.000137 - momentum: 0.000000
2023-10-08 19:30:05,342 epoch 2 - iter 216/242 - loss 0.67740262 - time (sec): 89.78 - samples/sec: 242.26 - lr: 0.000135 - momentum: 0.000000
2023-10-08 19:30:16,100 epoch 2 - iter 240/242 - loss 0.63844289 - time (sec): 100.53 - samples/sec: 243.56 - lr: 0.000134 - momentum: 0.000000
2023-10-08 19:30:17,027 ----------------------------------------------------------------------------------------------------
2023-10-08 19:30:17,028 EPOCH 2 done: loss 0.6350 - lr: 0.000134
2023-10-08 19:30:23,505 DEV : loss 0.4186439514160156 - f1-score (micro avg) 0.0
2023-10-08 19:30:23,511 ----------------------------------------------------------------------------------------------------
2023-10-08 19:30:33,891 epoch 3 - iter 24/242 - loss 0.34404676 - time (sec): 10.38 - samples/sec: 256.89 - lr: 0.000132 - momentum: 0.000000
2023-10-08 19:30:44,601 epoch 3 - iter 48/242 - loss 0.34909213 - time (sec): 21.09 - samples/sec: 252.93 - lr: 0.000130 - momentum: 0.000000
2023-10-08 19:30:54,764 epoch 3 - iter 72/242 - loss 0.34216476 - time (sec): 31.25 - samples/sec: 249.40 - lr: 0.000128 - momentum: 0.000000
2023-10-08 19:31:04,335 epoch 3 - iter 96/242 - loss 0.33454166 - time (sec): 40.82 - samples/sec: 245.13 - lr: 0.000127 - momentum: 0.000000
2023-10-08 19:31:13,985 epoch 3 - iter 120/242 - loss 0.31976787 - time (sec): 50.47 - samples/sec: 245.10 - lr: 0.000125 - momentum: 0.000000
2023-10-08 19:31:23,325 epoch 3 - iter 144/242 - loss 0.31813013 - time (sec): 59.81 - samples/sec: 242.72 - lr: 0.000124 - momentum: 0.000000
2023-10-08 19:31:33,554 epoch 3 - iter 168/242 - loss 0.31672677 - time (sec): 70.04 - samples/sec: 243.18 - lr: 0.000122 - momentum: 0.000000
2023-10-08 19:31:44,595 epoch 3 - iter 192/242 - loss 0.30297284 - time (sec): 81.08 - samples/sec: 243.89 - lr: 0.000120 - momentum: 0.000000
2023-10-08 19:31:54,407 epoch 3 - iter 216/242 - loss 0.29839779 - time (sec): 90.89 - samples/sec: 243.92 - lr: 0.000119 - momentum: 0.000000
2023-10-08 19:32:04,662 epoch 3 - iter 240/242 - loss 0.29490268 - time (sec): 101.15 - samples/sec: 243.17 - lr: 0.000117 - momentum: 0.000000
2023-10-08 19:32:05,329 ----------------------------------------------------------------------------------------------------
2023-10-08 19:32:05,329 EPOCH 3 done: loss 0.2953 - lr: 0.000117
2023-10-08 19:32:11,891 DEV : loss 0.24679391086101532 - f1-score (micro avg) 0.5088
2023-10-08 19:32:11,897 saving best model
2023-10-08 19:32:12,776 ----------------------------------------------------------------------------------------------------
2023-10-08 19:32:23,503 epoch 4 - iter 24/242 - loss 0.17282296 - time (sec): 10.72 - samples/sec: 247.74 - lr: 0.000115 - momentum: 0.000000
2023-10-08 19:32:34,252 epoch 4 - iter 48/242 - loss 0.17292848 - time (sec): 21.47 - samples/sec: 245.60 - lr: 0.000113 - momentum: 0.000000
2023-10-08 19:32:43,965 epoch 4 - iter 72/242 - loss 0.18281416 - time (sec): 31.19 - samples/sec: 242.82 - lr: 0.000112 - momentum: 0.000000
2023-10-08 19:32:53,689 epoch 4 - iter 96/242 - loss 0.18598185 - time (sec): 40.91 - samples/sec: 240.67 - lr: 0.000110 - momentum: 0.000000
2023-10-08 19:33:03,446 epoch 4 - iter 120/242 - loss 0.19378349 - time (sec): 50.67 - samples/sec: 241.14 - lr: 0.000109 - momentum: 0.000000
2023-10-08 19:33:13,583 epoch 4 - iter 144/242 - loss 0.19081445 - time (sec): 60.81 - samples/sec: 242.81 - lr: 0.000107 - momentum: 0.000000
2023-10-08 19:33:24,027 epoch 4 - iter 168/242 - loss 0.19299211 - time (sec): 71.25 - samples/sec: 243.83 - lr: 0.000105 - momentum: 0.000000
2023-10-08 19:33:34,232 epoch 4 - iter 192/242 - loss 0.18829297 - time (sec): 81.45 - samples/sec: 244.29 - lr: 0.000104 - momentum: 0.000000
2023-10-08 19:33:43,781 epoch 4 - iter 216/242 - loss 0.18864321 - time (sec): 91.00 - samples/sec: 243.12 - lr: 0.000102 - momentum: 0.000000
2023-10-08 19:33:54,027 epoch 4 - iter 240/242 - loss 0.18335743 - time (sec): 101.25 - samples/sec: 243.20 - lr: 0.000100 - momentum: 0.000000
2023-10-08 19:33:54,562 ----------------------------------------------------------------------------------------------------
2023-10-08 19:33:54,562 EPOCH 4 done: loss 0.1836 - lr: 0.000100
2023-10-08 19:34:01,066 DEV : loss 0.16814066469669342 - f1-score (micro avg) 0.8
2023-10-08 19:34:01,072 saving best model
2023-10-08 19:34:05,615 ----------------------------------------------------------------------------------------------------
2023-10-08 19:34:15,235 epoch 5 - iter 24/242 - loss 0.16839716 - time (sec): 9.62 - samples/sec: 241.31 - lr: 0.000098 - momentum: 0.000000
2023-10-08 19:34:25,705 epoch 5 - iter 48/242 - loss 0.15370578 - time (sec): 20.09 - samples/sec: 250.39 - lr: 0.000097 - momentum: 0.000000
2023-10-08 19:34:35,106 epoch 5 - iter 72/242 - loss 0.13876158 - time (sec): 29.49 - samples/sec: 246.30 - lr: 0.000095 - momentum: 0.000000
2023-10-08 19:34:45,685 epoch 5 - iter 96/242 - loss 0.13675425 - time (sec): 40.07 - samples/sec: 248.52 - lr: 0.000094 - momentum: 0.000000
2023-10-08 19:34:55,753 epoch 5 - iter 120/242 - loss 0.13381720 - time (sec): 50.14 - samples/sec: 245.91 - lr: 0.000092 - momentum: 0.000000
2023-10-08 19:35:06,439 epoch 5 - iter 144/242 - loss 0.12482283 - time (sec): 60.82 - samples/sec: 246.67 - lr: 0.000090 - momentum: 0.000000
2023-10-08 19:35:16,629 epoch 5 - iter 168/242 - loss 0.12576448 - time (sec): 71.01 - samples/sec: 246.30 - lr: 0.000089 - momentum: 0.000000
2023-10-08 19:35:26,601 epoch 5 - iter 192/242 - loss 0.12561961 - time (sec): 80.98 - samples/sec: 244.64 - lr: 0.000087 - momentum: 0.000000
2023-10-08 19:35:36,252 epoch 5 - iter 216/242 - loss 0.12596156 - time (sec): 90.64 - samples/sec: 243.32 - lr: 0.000085 - momentum: 0.000000
2023-10-08 19:35:46,683 epoch 5 - iter 240/242 - loss 0.12429901 - time (sec): 101.07 - samples/sec: 243.79 - lr: 0.000084 - momentum: 0.000000
2023-10-08 19:35:47,237 ----------------------------------------------------------------------------------------------------
2023-10-08 19:35:47,238 EPOCH 5 done: loss 0.1242 - lr: 0.000084
2023-10-08 19:35:53,752 DEV : loss 0.14256200194358826 - f1-score (micro avg) 0.8099
2023-10-08 19:35:53,758 saving best model
2023-10-08 19:35:58,133 ----------------------------------------------------------------------------------------------------
2023-10-08 19:36:07,737 epoch 6 - iter 24/242 - loss 0.07318159 - time (sec): 9.60 - samples/sec: 232.64 - lr: 0.000082 - momentum: 0.000000
2023-10-08 19:36:17,776 epoch 6 - iter 48/242 - loss 0.09547560 - time (sec): 19.64 - samples/sec: 237.75 - lr: 0.000080 - momentum: 0.000000
2023-10-08 19:36:28,058 epoch 6 - iter 72/242 - loss 0.10091463 - time (sec): 29.92 - samples/sec: 241.65 - lr: 0.000079 - momentum: 0.000000
2023-10-08 19:36:37,251 epoch 6 - iter 96/242 - loss 0.09709764 - time (sec): 39.12 - samples/sec: 239.95 - lr: 0.000077 - momentum: 0.000000
2023-10-08 19:36:47,708 epoch 6 - iter 120/242 - loss 0.09869626 - time (sec): 49.57 - samples/sec: 241.46 - lr: 0.000075 - momentum: 0.000000
2023-10-08 19:36:57,206 epoch 6 - iter 144/242 - loss 0.09876091 - time (sec): 59.07 - samples/sec: 241.10 - lr: 0.000074 - momentum: 0.000000
2023-10-08 19:37:07,362 epoch 6 - iter 168/242 - loss 0.09649570 - time (sec): 69.23 - samples/sec: 241.90 - lr: 0.000072 - momentum: 0.000000
2023-10-08 19:37:18,206 epoch 6 - iter 192/242 - loss 0.09659304 - time (sec): 80.07 - samples/sec: 243.96 - lr: 0.000070 - momentum: 0.000000
2023-10-08 19:37:28,339 epoch 6 - iter 216/242 - loss 0.09258049 - time (sec): 90.20 - samples/sec: 243.64 - lr: 0.000069 - momentum: 0.000000
2023-10-08 19:37:39,031 epoch 6 - iter 240/242 - loss 0.09096943 - time (sec): 100.90 - samples/sec: 243.94 - lr: 0.000067 - momentum: 0.000000
2023-10-08 19:37:39,654 ----------------------------------------------------------------------------------------------------
2023-10-08 19:37:39,654 EPOCH 6 done: loss 0.0910 - lr: 0.000067
2023-10-08 19:37:46,246 DEV : loss 0.1273980289697647 - f1-score (micro avg) 0.8089
2023-10-08 19:37:46,252 ----------------------------------------------------------------------------------------------------
2023-10-08 19:37:57,076 epoch 7 - iter 24/242 - loss 0.06871926 - time (sec): 10.82 - samples/sec: 245.24 - lr: 0.000065 - momentum: 0.000000
2023-10-08 19:38:07,511 epoch 7 - iter 48/242 - loss 0.07730926 - time (sec): 21.26 - samples/sec: 248.48 - lr: 0.000064 - momentum: 0.000000
2023-10-08 19:38:17,953 epoch 7 - iter 72/242 - loss 0.07575126 - time (sec): 31.70 - samples/sec: 248.53 - lr: 0.000062 - momentum: 0.000000
2023-10-08 19:38:28,429 epoch 7 - iter 96/242 - loss 0.07265611 - time (sec): 42.17 - samples/sec: 249.94 - lr: 0.000060 - momentum: 0.000000
2023-10-08 19:38:37,550 epoch 7 - iter 120/242 - loss 0.07085807 - time (sec): 51.30 - samples/sec: 250.31 - lr: 0.000059 - momentum: 0.000000
2023-10-08 19:38:47,618 epoch 7 - iter 144/242 - loss 0.07431527 - time (sec): 61.36 - samples/sec: 253.47 - lr: 0.000057 - momentum: 0.000000
2023-10-08 19:38:57,283 epoch 7 - iter 168/242 - loss 0.07243790 - time (sec): 71.03 - samples/sec: 254.16 - lr: 0.000055 - momentum: 0.000000
2023-10-08 19:39:06,063 epoch 7 - iter 192/242 - loss 0.06893922 - time (sec): 79.81 - samples/sec: 253.08 - lr: 0.000054 - momentum: 0.000000
2023-10-08 19:39:15,560 epoch 7 - iter 216/242 - loss 0.06858497 - time (sec): 89.31 - samples/sec: 253.08 - lr: 0.000052 - momentum: 0.000000
2023-10-08 19:39:24,274 epoch 7 - iter 240/242 - loss 0.06623735 - time (sec): 98.02 - samples/sec: 251.23 - lr: 0.000050 - momentum: 0.000000
2023-10-08 19:39:24,809 ----------------------------------------------------------------------------------------------------
2023-10-08 19:39:24,809 EPOCH 7 done: loss 0.0663 - lr: 0.000050
2023-10-08 19:39:30,675 DEV : loss 0.12801626324653625 - f1-score (micro avg) 0.8209
2023-10-08 19:39:30,681 saving best model
2023-10-08 19:39:35,384 ----------------------------------------------------------------------------------------------------
2023-10-08 19:39:44,381 epoch 8 - iter 24/242 - loss 0.05398769 - time (sec): 9.00 - samples/sec: 256.90 - lr: 0.000049 - momentum: 0.000000
2023-10-08 19:39:53,342 epoch 8 - iter 48/242 - loss 0.05644280 - time (sec): 17.96 - samples/sec: 257.96 - lr: 0.000047 - momentum: 0.000000
2023-10-08 19:40:02,980 epoch 8 - iter 72/242 - loss 0.05735985 - time (sec): 27.59 - samples/sec: 261.97 - lr: 0.000045 - momentum: 0.000000
2023-10-08 19:40:12,506 epoch 8 - iter 96/242 - loss 0.05615888 - time (sec): 37.12 - samples/sec: 262.09 - lr: 0.000044 - momentum: 0.000000
2023-10-08 19:40:22,214 epoch 8 - iter 120/242 - loss 0.05155077 - time (sec): 46.83 - samples/sec: 262.58 - lr: 0.000042 - momentum: 0.000000
2023-10-08 19:40:30,895 epoch 8 - iter 144/242 - loss 0.05489311 - time (sec): 55.51 - samples/sec: 260.68 - lr: 0.000040 - momentum: 0.000000
2023-10-08 19:40:40,449 epoch 8 - iter 168/242 - loss 0.05635034 - time (sec): 65.06 - samples/sec: 260.91 - lr: 0.000039 - momentum: 0.000000
2023-10-08 19:40:50,423 epoch 8 - iter 192/242 - loss 0.05629495 - time (sec): 75.04 - samples/sec: 262.65 - lr: 0.000037 - momentum: 0.000000
2023-10-08 19:40:59,910 epoch 8 - iter 216/242 - loss 0.05612159 - time (sec): 84.52 - samples/sec: 262.30 - lr: 0.000035 - momentum: 0.000000
2023-10-08 19:41:09,170 epoch 8 - iter 240/242 - loss 0.05374507 - time (sec): 93.79 - samples/sec: 261.34 - lr: 0.000034 - momentum: 0.000000
2023-10-08 19:41:09,952 ----------------------------------------------------------------------------------------------------
2023-10-08 19:41:09,952 EPOCH 8 done: loss 0.0535 - lr: 0.000034
2023-10-08 19:41:15,913 DEV : loss 0.13416925072669983 - f1-score (micro avg) 0.8165
2023-10-08 19:41:15,919 ----------------------------------------------------------------------------------------------------
2023-10-08 19:41:25,146 epoch 9 - iter 24/242 - loss 0.06955313 - time (sec): 9.23 - samples/sec: 262.11 - lr: 0.000032 - momentum: 0.000000
2023-10-08 19:41:34,589 epoch 9 - iter 48/242 - loss 0.05268402 - time (sec): 18.67 - samples/sec: 261.24 - lr: 0.000030 - momentum: 0.000000
2023-10-08 19:41:44,695 epoch 9 - iter 72/242 - loss 0.04529954 - time (sec): 28.77 - samples/sec: 266.66 - lr: 0.000029 - momentum: 0.000000
2023-10-08 19:41:54,189 epoch 9 - iter 96/242 - loss 0.04194193 - time (sec): 38.27 - samples/sec: 265.41 - lr: 0.000027 - momentum: 0.000000
2023-10-08 19:42:03,055 epoch 9 - iter 120/242 - loss 0.04197999 - time (sec): 47.13 - samples/sec: 263.16 - lr: 0.000025 - momentum: 0.000000
2023-10-08 19:42:12,389 epoch 9 - iter 144/242 - loss 0.04208191 - time (sec): 56.47 - samples/sec: 261.83 - lr: 0.000024 - momentum: 0.000000
2023-10-08 19:42:21,985 epoch 9 - iter 168/242 - loss 0.03968387 - time (sec): 66.06 - samples/sec: 260.79 - lr: 0.000022 - momentum: 0.000000
2023-10-08 19:42:31,126 epoch 9 - iter 192/242 - loss 0.04070292 - time (sec): 75.21 - samples/sec: 260.71 - lr: 0.000020 - momentum: 0.000000
2023-10-08 19:42:40,599 epoch 9 - iter 216/242 - loss 0.04310635 - time (sec): 84.68 - samples/sec: 260.77 - lr: 0.000019 - momentum: 0.000000
2023-10-08 19:42:50,301 epoch 9 - iter 240/242 - loss 0.04357004 - time (sec): 94.38 - samples/sec: 260.87 - lr: 0.000017 - momentum: 0.000000
2023-10-08 19:42:50,830 ----------------------------------------------------------------------------------------------------
2023-10-08 19:42:50,830 EPOCH 9 done: loss 0.0434 - lr: 0.000017
2023-10-08 19:42:56,751 DEV : loss 0.13480180501937866 - f1-score (micro avg) 0.8331
2023-10-08 19:42:56,758 saving best model
2023-10-08 19:43:01,147 ----------------------------------------------------------------------------------------------------
2023-10-08 19:43:10,893 epoch 10 - iter 24/242 - loss 0.03780505 - time (sec): 9.74 - samples/sec: 272.37 - lr: 0.000015 - momentum: 0.000000
2023-10-08 19:43:19,978 epoch 10 - iter 48/242 - loss 0.03456896 - time (sec): 18.83 - samples/sec: 265.59 - lr: 0.000014 - momentum: 0.000000
2023-10-08 19:43:29,159 epoch 10 - iter 72/242 - loss 0.03411245 - time (sec): 28.01 - samples/sec: 260.93 - lr: 0.000012 - momentum: 0.000000
2023-10-08 19:43:38,241 epoch 10 - iter 96/242 - loss 0.03540146 - time (sec): 37.09 - samples/sec: 255.58 - lr: 0.000010 - momentum: 0.000000
2023-10-08 19:43:48,044 epoch 10 - iter 120/242 - loss 0.03421286 - time (sec): 46.90 - samples/sec: 256.51 - lr: 0.000009 - momentum: 0.000000
2023-10-08 19:43:56,972 epoch 10 - iter 144/242 - loss 0.03425299 - time (sec): 55.82 - samples/sec: 255.16 - lr: 0.000007 - momentum: 0.000000
2023-10-08 19:44:06,515 epoch 10 - iter 168/242 - loss 0.03345010 - time (sec): 65.37 - samples/sec: 254.63 - lr: 0.000005 - momentum: 0.000000
2023-10-08 19:44:16,793 epoch 10 - iter 192/242 - loss 0.03486005 - time (sec): 75.64 - samples/sec: 255.06 - lr: 0.000004 - momentum: 0.000000
2023-10-08 19:44:27,150 epoch 10 - iter 216/242 - loss 0.03874650 - time (sec): 86.00 - samples/sec: 256.60 - lr: 0.000002 - momentum: 0.000000
2023-10-08 19:44:37,226 epoch 10 - iter 240/242 - loss 0.03982743 - time (sec): 96.08 - samples/sec: 255.16 - lr: 0.000000 - momentum: 0.000000
2023-10-08 19:44:38,048 ----------------------------------------------------------------------------------------------------
2023-10-08 19:44:38,048 EPOCH 10 done: loss 0.0397 - lr: 0.000000
2023-10-08 19:44:44,215 DEV : loss 0.1376410871744156 - f1-score (micro avg) 0.8352
2023-10-08 19:44:44,221 saving best model
2023-10-08 19:44:49,445 ----------------------------------------------------------------------------------------------------
2023-10-08 19:44:49,446 Loading model from best epoch ...
2023-10-08 19:44:53,145 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 19:44:59,472
Results:
- F-score (micro) 0.7936
- F-score (macro) 0.4008
- Accuracy 0.6866
By class:
precision recall f1-score support
pers 0.8207 0.8561 0.8380 139
scope 0.8357 0.9070 0.8699 129
work 0.6327 0.7750 0.6966 80
loc 0.0000 0.0000 0.0000 9
object 0.0000 0.0000 0.0000 0
date 0.0000 0.0000 0.0000 3
micro avg 0.7621 0.8278 0.7936 360
macro avg 0.3815 0.4230 0.4008 360
weighted avg 0.7569 0.8278 0.7901 360
2023-10-08 19:44:59,472 ----------------------------------------------------------------------------------------------------
|