simonycl's picture
update model card README.md
7b717ef
|
raw
history blame
10.6 kB
metadata
license: mit
base_model: roberta-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: best_model-yelp_polarity-64-100
    results: []

best_model-yelp_polarity-64-100

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

  • Loss: 0.6421
  • Accuracy: 0.9453

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: 1e-05
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • num_epochs: 150

Training results

Training Loss Epoch Step Validation Loss Accuracy
No log 1.0 4 0.6227 0.9219
No log 2.0 8 0.6400 0.9219
0.151 3.0 12 0.6723 0.9219
0.151 4.0 16 0.6894 0.9219
0.1653 5.0 20 0.6886 0.9219
0.1653 6.0 24 0.6817 0.9219
0.1653 7.0 28 0.6861 0.9219
0.1389 8.0 32 0.6867 0.9219
0.1389 9.0 36 0.6738 0.9219
0.1365 10.0 40 0.6662 0.9219
0.1365 11.0 44 0.6472 0.9219
0.1365 12.0 48 0.6332 0.9219
0.058 13.0 52 0.6449 0.9297
0.058 14.0 56 0.6604 0.9297
0.0532 15.0 60 0.6461 0.9297
0.0532 16.0 64 0.5893 0.9219
0.0532 17.0 68 0.5498 0.9297
0.0062 18.0 72 0.5295 0.9375
0.0062 19.0 76 0.5262 0.9375
0.0045 20.0 80 0.5270 0.9375
0.0045 21.0 84 0.5261 0.9453
0.0045 22.0 88 0.5272 0.9453
0.0 23.0 92 0.5281 0.9453
0.0 24.0 96 0.5295 0.9453
0.001 25.0 100 0.5315 0.9453
0.001 26.0 104 0.5375 0.9375
0.001 27.0 108 0.5524 0.9297
0.0 28.0 112 0.5820 0.9219
0.0 29.0 116 0.6046 0.9297
0.0 30.0 120 0.6323 0.9297
0.0 31.0 124 0.6660 0.9297
0.0 32.0 128 0.6793 0.9297
0.0 33.0 132 0.6855 0.9297
0.0 34.0 136 0.6604 0.9297
0.0 35.0 140 0.5577 0.9297
0.0 36.0 144 0.5515 0.9453
0.0 37.0 148 0.5494 0.9453
0.0 38.0 152 0.5492 0.9453
0.0 39.0 156 0.5491 0.9453
0.0 40.0 160 0.5493 0.9453
0.0 41.0 164 0.5671 0.9453
0.0 42.0 168 0.5708 0.9453
0.006 43.0 172 0.5740 0.9375
0.006 44.0 176 0.5883 0.9297
0.0 45.0 180 0.6010 0.9297
0.0 46.0 184 0.6081 0.9297
0.0 47.0 188 0.6122 0.9297
0.0 48.0 192 0.6149 0.9297
0.0 49.0 196 0.6166 0.9297
0.0 50.0 200 0.6177 0.9297
0.0 51.0 204 0.6205 0.9297
0.0 52.0 208 0.6229 0.9297
0.0 53.0 212 0.6242 0.9297
0.0 54.0 216 0.6251 0.9297
0.0 55.0 220 0.6205 0.9297
0.0 56.0 224 0.6152 0.9297
0.0 57.0 228 0.6106 0.9297
0.0 58.0 232 0.6068 0.9297
0.0 59.0 236 0.6041 0.9297
0.0 60.0 240 0.6025 0.9297
0.0 61.0 244 0.6008 0.9297
0.0 62.0 248 0.5988 0.9297
0.0 63.0 252 0.5965 0.9297
0.0 64.0 256 0.5944 0.9297
0.0 65.0 260 0.5928 0.9297
0.0 66.0 264 0.5920 0.9453
0.0 67.0 268 0.5914 0.9453
0.0 68.0 272 0.5914 0.9453
0.0 69.0 276 0.5916 0.9453
0.0 70.0 280 0.5919 0.9453
0.0 71.0 284 0.5923 0.9453
0.0 72.0 288 0.5927 0.9453
0.0 73.0 292 0.5931 0.9453
0.0 74.0 296 0.5935 0.9453
0.0 75.0 300 0.5938 0.9453
0.0 76.0 304 0.5942 0.9453
0.0 77.0 308 0.5946 0.9453
0.0 78.0 312 0.5950 0.9453
0.0 79.0 316 0.5954 0.9453
0.0 80.0 320 0.5959 0.9453
0.0 81.0 324 0.5868 0.9453
0.0 82.0 328 0.6180 0.9375
0.0005 83.0 332 0.6404 0.9453
0.0005 84.0 336 0.6560 0.9453
0.0 85.0 340 0.6606 0.9453
0.0 86.0 344 0.6415 0.9453
0.0 87.0 348 0.5770 0.9453
0.0247 88.0 352 0.5282 0.9375
0.0247 89.0 356 0.5532 0.9453
0.0 90.0 360 0.5550 0.9453
0.0 91.0 364 0.5455 0.9453
0.0 92.0 368 0.5395 0.9375
0.0 93.0 372 0.5358 0.9375
0.0 94.0 376 0.5333 0.9453
0.0 95.0 380 0.5314 0.9453
0.0 96.0 384 0.5303 0.9453
0.0 97.0 388 0.5295 0.9453
0.0 98.0 392 0.5288 0.9453
0.0 99.0 396 0.5279 0.9453
0.0 100.0 400 0.5270 0.9453
0.0 101.0 404 0.5264 0.9453
0.0 102.0 408 0.5260 0.9453
0.0 103.0 412 0.5257 0.9453
0.0 104.0 416 0.5256 0.9453
0.0 105.0 420 0.5255 0.9453
0.0 106.0 424 0.5255 0.9453
0.0 107.0 428 0.5256 0.9453
0.0 108.0 432 0.5257 0.9453
0.0 109.0 436 0.5258 0.9453
0.0 110.0 440 0.5259 0.9453
0.0 111.0 444 0.5262 0.9453
0.0 112.0 448 0.5264 0.9453
0.0 113.0 452 0.5266 0.9453
0.0 114.0 456 0.5265 0.9453
0.0 115.0 460 0.5266 0.9453
0.0 116.0 464 0.5268 0.9453
0.0 117.0 468 0.5263 0.9453
0.0 118.0 472 0.5401 0.9453
0.0 119.0 476 0.5557 0.9453
0.0 120.0 480 0.5663 0.9453
0.0 121.0 484 0.5731 0.9453
0.0 122.0 488 0.5776 0.9453
0.0 123.0 492 0.5804 0.9453
0.0 124.0 496 0.5823 0.9453
0.0 125.0 500 0.5836 0.9453
0.0 126.0 504 0.5842 0.9453
0.0 127.0 508 0.5844 0.9453
0.0 128.0 512 0.5844 0.9453
0.0 129.0 516 0.5826 0.9453
0.0 130.0 520 0.5813 0.9453
0.0 131.0 524 0.5805 0.9453
0.0 132.0 528 0.5800 0.9453
0.0 133.0 532 0.5796 0.9453
0.0 134.0 536 0.6100 0.9453
0.0 135.0 540 0.6275 0.9453
0.0 136.0 544 0.6351 0.9453
0.0 137.0 548 0.6393 0.9453
0.0 138.0 552 0.6424 0.9453
0.0 139.0 556 0.6445 0.9453
0.0 140.0 560 0.6439 0.9453
0.0 141.0 564 0.6435 0.9453
0.0 142.0 568 0.6431 0.9453
0.0 143.0 572 0.6428 0.9453
0.0 144.0 576 0.6425 0.9453
0.0 145.0 580 0.6423 0.9453
0.0 146.0 584 0.6422 0.9453
0.0 147.0 588 0.6422 0.9453
0.0 148.0 592 0.6421 0.9453
0.0 149.0 596 0.6421 0.9453
0.0 150.0 600 0.6421 0.9453

Framework versions

  • Transformers 4.32.0.dev0
  • Pytorch 2.0.1+cu118
  • Datasets 2.4.0
  • Tokenizers 0.13.3