Hugging Face's logo

sheepy928
/
default

Transformers
PyTorch
roberta
Generated from Trainer
Inference Endpoints
Model card Files Community
default / README.md
sheepy928's picture
sheepy928
Model save
3763ed9
preview code
|
raw
history blame contribute delete
10.6 kB
metadata
license: mit
base_model: roberta-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - recall
  - precision
model-index:
  - name: default
    results: []

default

This model is a fine-tuned version of roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1946
  • Accuracy: 0.9395
  • F1: 0.9398
  • Recall: 0.9395
  • Precision: 0.9413
  • Combined Score: 0.9400

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: 0.0005
  • train_batch_size: 128
  • eval_batch_size: 512
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 512
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 10000
  • num_epochs: 5

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 Recall Precision Combined Score
1.0664 0.06 20 1.0651 0.4448 0.2739 0.4448 0.1978 0.3403
1.0423 0.12 40 1.0188 0.5034 0.4138 0.5034 0.4328 0.4633
1.0137 0.18 60 0.9871 0.5279 0.4596 0.5279 0.4219 0.4843
1.0027 0.24 80 0.9889 0.5308 0.4653 0.5308 0.4196 0.4866
0.9914 0.3 100 0.9763 0.5308 0.4666 0.5308 0.4176 0.4864
0.9826 0.36 120 0.9713 0.5388 0.4711 0.5388 0.4292 0.4945
0.9788 0.42 140 0.9766 0.5313 0.4674 0.5313 0.4186 0.4871
0.984 0.48 160 0.9590 0.5398 0.4751 0.5398 0.4243 0.4948
0.9694 0.54 180 0.9535 0.5423 0.4772 0.5423 0.4269 0.4972
0.9676 0.6 200 0.9274 0.5672 0.4991 0.5672 0.4467 0.5200
0.9753 0.66 220 0.9126 0.5736 0.5026 0.5736 0.4616 0.5279
0.9557 0.72 240 0.9053 0.5760 0.5069 0.5760 0.4532 0.5280
0.9508 0.78 260 0.9179 0.5767 0.5018 0.5767 0.4811 0.5341
0.9355 0.84 280 0.8937 0.5892 0.5183 0.5892 0.4662 0.5407
0.9 0.9 300 0.8469 0.6130 0.5579 0.6130 0.5855 0.5924
0.993 0.96 320 0.8615 0.6047 0.5352 0.6047 0.6175 0.5905
0.8527 1.02 340 0.7896 0.6439 0.6212 0.6439 0.6240 0.6333
0.966 1.08 360 1.0124 0.5316 0.4944 0.5316 0.6428 0.5501
0.8441 1.14 380 0.7911 0.6489 0.6371 0.6489 0.6584 0.6483
0.8226 1.2 400 0.7472 0.6700 0.6459 0.6700 0.6625 0.6621
0.7948 1.26 420 0.7664 0.6581 0.6095 0.6581 0.6796 0.6513
0.7428 1.32 440 0.6994 0.6992 0.6533 0.6992 0.7104 0.6905
0.7109 1.38 460 0.6511 0.7284 0.6890 0.7284 0.7345 0.7201
0.6882 1.44 480 0.5988 0.7577 0.7336 0.7577 0.7570 0.7515
0.7296 1.5 500 0.5993 0.7564 0.7669 0.7564 0.7902 0.7675
0.5677 1.57 520 0.5068 0.8126 0.7942 0.8126 0.8225 0.8105
0.5096 1.63 540 0.4273 0.8520 0.8449 0.8520 0.8511 0.8500
0.4452 1.69 560 0.3796 0.8722 0.8699 0.8722 0.8703 0.8711
0.3836 1.75 580 0.3855 0.8757 0.8758 0.8757 0.8778 0.8762
0.3783 1.81 600 0.3586 0.8894 0.8883 0.8894 0.8885 0.8889
0.3496 1.87 620 0.3210 0.8972 0.8973 0.8972 0.8974 0.8973
0.3585 1.93 640 0.3006 0.9035 0.9031 0.9035 0.9033 0.9034
0.345 1.99 660 0.3054 0.9014 0.9025 0.9014 0.9052 0.9026
0.3327 2.05 680 0.3174 0.8913 0.8866 0.8913 0.8955 0.8912
0.2962 2.11 700 0.2770 0.9122 0.9125 0.9122 0.9130 0.9125
0.3032 2.17 720 0.2979 0.9062 0.9055 0.9062 0.9093 0.9068
0.27 2.23 740 0.2973 0.8998 0.8971 0.8998 0.9045 0.9003
0.2912 2.29 760 0.2467 0.9222 0.9221 0.9222 0.9223 0.9222
0.2412 2.35 780 0.2761 0.9113 0.9128 0.9113 0.9173 0.9132
0.2746 2.41 800 0.2410 0.9260 0.9257 0.9260 0.9260 0.9259
0.2637 2.47 820 0.2447 0.9221 0.9213 0.9221 0.9228 0.9221
0.2605 2.53 840 0.2475 0.9237 0.9232 0.9237 0.9254 0.9240
0.2517 2.59 860 0.2590 0.9265 0.9259 0.9265 0.9272 0.9265
0.2453 2.65 880 0.2248 0.9300 0.9305 0.9300 0.9315 0.9305
0.2247 2.71 900 0.2285 0.9273 0.9281 0.9273 0.9299 0.9282
0.2402 2.77 920 0.2304 0.9306 0.9310 0.9306 0.9317 0.9310
0.2033 2.83 940 0.2228 0.9319 0.9316 0.9319 0.9325 0.9320
0.2315 2.89 960 0.2275 0.9271 0.9281 0.9271 0.9311 0.9283
0.2231 2.95 980 0.2115 0.9343 0.9345 0.9343 0.9350 0.9345
0.2061 3.01 1000 0.2156 0.9355 0.9356 0.9355 0.9361 0.9357
0.238 3.07 1020 0.2359 0.9252 0.9262 0.9252 0.9294 0.9265
0.1959 3.13 1040 0.2052 0.9343 0.9343 0.9343 0.9351 0.9345
0.2013 3.19 1060 0.2114 0.9337 0.9336 0.9337 0.9349 0.9340
0.2055 3.25 1080 0.1985 0.9354 0.9354 0.9354 0.9361 0.9356
0.1851 3.31 1100 0.2098 0.9340 0.9338 0.9340 0.9352 0.9342
0.2091 3.37 1120 0.2002 0.9350 0.9347 0.9350 0.9355 0.9350
0.1933 3.43 1140 0.2045 0.9367 0.9371 0.9367 0.9377 0.9370
0.2011 3.49 1160 0.2072 0.9350 0.9345 0.9350 0.9353 0.9349
0.2085 3.55 1180 0.2124 0.9327 0.9337 0.9327 0.9362 0.9338
0.1991 3.61 1200 0.1880 0.9396 0.9398 0.9396 0.9406 0.9399
0.1972 3.67 1220 0.2123 0.9347 0.9344 0.9347 0.9351 0.9347
0.1905 3.73 1240 0.2056 0.9351 0.9352 0.9351 0.9368 0.9355
0.1974 3.79 1260 0.2432 0.9285 0.9280 0.9285 0.9308 0.9289
0.1831 3.85 1280 0.1881 0.9392 0.9390 0.9392 0.9402 0.9394
0.1883 3.91 1300 0.1948 0.9413 0.9415 0.9413 0.9423 0.9416
0.2023 3.97 1320 0.2041 0.9335 0.9343 0.9335 0.9369 0.9346
0.1694 4.03 1340 0.2007 0.9360 0.9360 0.9360 0.9372 0.9363
0.1727 4.09 1360 0.1990 0.9411 0.9414 0.9411 0.9422 0.9414
0.1666 4.15 1380 0.1952 0.9397 0.9399 0.9397 0.9403 0.9399
0.2011 4.21 1400 0.1953 0.9384 0.9388 0.9384 0.9401 0.9389
0.1798 4.27 1420 0.2032 0.9325 0.9318 0.9325 0.9345 0.9328
0.1808 4.33 1440 0.2085 0.9308 0.9319 0.9308 0.9357 0.9323
0.1552 4.39 1460 0.2010 0.9371 0.9377 0.9371 0.9391 0.9377
0.1608 4.45 1480 0.2090 0.9342 0.9349 0.9342 0.9372 0.9351
0.1683 4.51 1500 0.2005 0.9374 0.9371 0.9374 0.9384 0.9376
0.1484 4.57 1520 0.1893 0.9380 0.9386 0.9380 0.9402 0.9387
0.1561 4.64 1540 0.1922 0.9401 0.9405 0.9401 0.9417 0.9406
0.1753 4.7 1560 0.1905 0.9416 0.9418 0.9416 0.9428 0.9419
0.1668 4.76 1580 0.1767 0.9420 0.9419 0.9420 0.9426 0.9421
0.1651 4.82 1600 0.1772 0.9429 0.9429 0.9429 0.9431 0.9429
0.1601 4.88 1620 0.1778 0.9433 0.9434 0.9433 0.9436 0.9434
0.1686 4.94 1640 0.1943 0.9414 0.9415 0.9414 0.9423 0.9416
0.1342 5.0 1660 0.1946 0.9395 0.9398 0.9395 0.9413 0.9400

Framework versions

  • Transformers 4.34.1
  • Pytorch 2.1.0
  • Datasets 2.14.5
  • Tokenizers 0.14.1