Edit model card

bertbasecasedfinancialphrasebank

This model is a fine-tuned version of bert-base-cased on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5046
  • Accuracy: 0.8660

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 2e-05
  • train_batch_size: 32
  • eval_batch_size: 64
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.9921 0.04 5 0.9266 0.6082
0.8989 0.08 10 0.8833 0.6082
0.8563 0.12 15 0.8287 0.6371
0.8316 0.16 20 0.7626 0.6866
0.808 0.2 25 0.7398 0.6845
0.8219 0.25 30 0.7234 0.6969
0.7402 0.29 35 0.7080 0.6990
0.749 0.33 40 0.7261 0.6948
0.8777 0.37 45 0.7253 0.7072
0.7057 0.41 50 0.6736 0.7196
0.6968 0.45 55 0.6326 0.7320
0.6264 0.49 60 0.6121 0.7443
0.703 0.53 65 0.6008 0.7505
0.6788 0.57 70 0.6421 0.7278
0.6458 0.61 75 0.5801 0.7546
0.605 0.66 80 0.5601 0.7588
0.5428 0.7 85 0.5480 0.7546
0.6025 0.74 90 0.5632 0.7402
0.5761 0.78 95 0.5053 0.7794
0.4919 0.82 100 0.4745 0.8289
0.5217 0.86 105 0.4731 0.8268
0.5517 0.9 110 0.4512 0.8474
0.4131 0.94 115 0.4280 0.8392
0.5244 0.98 120 0.4222 0.8433
0.6407 1.02 125 0.4474 0.8392
0.3661 1.07 130 0.4101 0.8474
0.4148 1.11 135 0.3810 0.8619
0.296 1.15 140 0.4003 0.8371
0.3579 1.19 145 0.3738 0.8598
0.3718 1.23 150 0.3635 0.8598
0.307 1.27 155 0.3900 0.8515
0.3055 1.31 160 0.3881 0.8454
0.2789 1.35 165 0.3679 0.8495
0.2541 1.39 170 0.3709 0.8557
0.3263 1.43 175 0.3815 0.8557
0.305 1.48 180 0.3757 0.8495
0.2215 1.52 185 0.3643 0.8639
0.3642 1.56 190 0.3887 0.8495
0.4179 1.6 195 0.3736 0.8536
0.3126 1.64 200 0.3928 0.8392
0.2625 1.68 205 0.3648 0.8577
0.2155 1.72 210 0.3612 0.8577
0.2526 1.76 215 0.3781 0.8433
0.3033 1.8 220 0.4334 0.8412
0.2752 1.84 225 0.4039 0.8392
0.3466 1.89 230 0.3800 0.8412
0.4447 1.93 235 0.3575 0.8412
0.257 1.97 240 0.3732 0.8515
0.2647 2.01 245 0.3542 0.8454
0.1795 2.05 250 0.3599 0.8495
0.1697 2.09 255 0.3694 0.8515
0.1907 2.13 260 0.3985 0.8495
0.2515 2.17 265 0.3753 0.8660
0.1921 2.21 270 0.3674 0.8598
0.1342 2.25 275 0.3712 0.8577
0.1424 2.3 280 0.3595 0.8680
0.2071 2.34 285 0.3650 0.8680
0.2224 2.38 290 0.3881 0.8557
0.2556 2.42 295 0.3598 0.8536
0.1518 2.46 300 0.3506 0.8619
0.1427 2.5 305 0.3621 0.8639
0.1599 2.54 310 0.4495 0.8371
0.163 2.58 315 0.4812 0.8124
0.1462 2.62 320 0.4012 0.8680
0.2213 2.66 325 0.4047 0.8557
0.1967 2.7 330 0.4008 0.8701
0.2305 2.75 335 0.3838 0.8722
0.1097 2.79 340 0.3743 0.8577
0.1521 2.83 345 0.3700 0.8701
0.0833 2.87 350 0.3831 0.8680
0.2099 2.91 355 0.3947 0.8722
0.2086 2.95 360 0.4184 0.8515
0.1294 2.99 365 0.4214 0.8536
0.0992 3.03 370 0.4087 0.8639
0.0973 3.07 375 0.4073 0.8557
0.0519 3.11 380 0.4085 0.8639
0.1418 3.16 385 0.4109 0.8639
0.081 3.2 390 0.4426 0.8619
0.0506 3.24 395 0.4733 0.8454
0.0962 3.28 400 0.5100 0.8412
0.1045 3.32 405 0.4408 0.8639
0.1236 3.36 410 0.4395 0.8619
0.0962 3.4 415 0.4515 0.8557
0.1611 3.44 420 0.4819 0.8474
0.1044 3.48 425 0.4336 0.8598
0.0715 3.52 430 0.4574 0.8639
0.0302 3.57 435 0.4617 0.8619
0.1234 3.61 440 0.4710 0.8577
0.1439 3.65 445 0.5082 0.8433
0.1113 3.69 450 0.4596 0.8639
0.1336 3.73 455 0.4631 0.8598
0.072 3.77 460 0.4372 0.8660
0.0621 3.81 465 0.4381 0.8619
0.1703 3.85 470 0.4337 0.8680
0.0284 3.89 475 0.4321 0.8742
0.0647 3.93 480 0.4575 0.8680
0.0897 3.98 485 0.4676 0.8680
0.0646 4.02 490 0.4550 0.8722
0.0554 4.06 495 0.4721 0.8598
0.0259 4.1 500 0.5045 0.8454
0.0186 4.14 505 0.4635 0.8577
0.0391 4.18 510 0.5362 0.8639
0.0505 4.22 515 0.5122 0.8722
0.1101 4.26 520 0.5046 0.8660

Framework versions

  • Transformers 4.21.1
  • Pytorch 1.12.1+cu113
  • Datasets 2.4.0
  • Tokenizers 0.12.1
Downloads last month
5