--- license: mit tags: - generated_from_trainer datasets: - imdb metrics: - accuracy model-index: - name: IMDB_roBERTa_5E results: - task: name: Text Classification type: text-classification dataset: name: imdb type: imdb config: plain_text split: train args: plain_text metrics: - name: Accuracy type: accuracy value: 0.9466666666666667 --- # IMDB_roBERTa_5E This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on the imdb dataset. It achieves the following results on the evaluation set: - Loss: 0.2383 - Accuracy: 0.9467 ## 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: 16 - seed: 42 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 - lr_scheduler_type: linear - num_epochs: 5 ### Training results | Training Loss | Epoch | Step | Validation Loss | Accuracy | |:-------------:|:-----:|:----:|:---------------:|:--------:| | 0.5851 | 0.06 | 50 | 0.1789 | 0.94 | | 0.2612 | 0.13 | 100 | 0.1520 | 0.9533 | | 0.2339 | 0.19 | 150 | 0.1997 | 0.9267 | | 0.2349 | 0.26 | 200 | 0.1702 | 0.92 | | 0.207 | 0.32 | 250 | 0.1515 | 0.9333 | | 0.2222 | 0.38 | 300 | 0.1522 | 0.9467 | | 0.1916 | 0.45 | 350 | 0.1328 | 0.94 | | 0.1559 | 0.51 | 400 | 0.1676 | 0.94 | | 0.1621 | 0.58 | 450 | 0.1363 | 0.9467 | | 0.1663 | 0.64 | 500 | 0.1327 | 0.9533 | | 0.1841 | 0.7 | 550 | 0.1347 | 0.9467 | | 0.1742 | 0.77 | 600 | 0.1127 | 0.9533 | | 0.1559 | 0.83 | 650 | 0.1119 | 0.9467 | | 0.172 | 0.9 | 700 | 0.1123 | 0.9467 | | 0.1644 | 0.96 | 750 | 0.1326 | 0.96 | | 0.1524 | 1.02 | 800 | 0.1718 | 0.9467 | | 0.1456 | 1.09 | 850 | 0.1464 | 0.9467 | | 0.1271 | 1.15 | 900 | 0.1190 | 0.9533 | | 0.1412 | 1.21 | 950 | 0.1323 | 0.9533 | | 0.1114 | 1.28 | 1000 | 0.1475 | 0.9467 | | 0.1222 | 1.34 | 1050 | 0.1592 | 0.9467 | | 0.1164 | 1.41 | 1100 | 0.1345 | 0.96 | | 0.1126 | 1.47 | 1150 | 0.1325 | 0.9533 | | 0.1237 | 1.53 | 1200 | 0.1561 | 0.9533 | | 0.1385 | 1.6 | 1250 | 0.1225 | 0.9467 | | 0.1522 | 1.66 | 1300 | 0.1119 | 0.9533 | | 0.1154 | 1.73 | 1350 | 0.1231 | 0.96 | | 0.1182 | 1.79 | 1400 | 0.1366 | 0.96 | | 0.1415 | 1.85 | 1450 | 0.0972 | 0.96 | | 0.124 | 1.92 | 1500 | 0.1082 | 0.96 | | 0.1584 | 1.98 | 1550 | 0.1770 | 0.96 | | 0.0927 | 2.05 | 1600 | 0.1821 | 0.9533 | | 0.1065 | 2.11 | 1650 | 0.0999 | 0.9733 | | 0.0974 | 2.17 | 1700 | 0.1365 | 0.9533 | | 0.079 | 2.24 | 1750 | 0.1694 | 0.9467 | | 0.1217 | 2.3 | 1800 | 0.1564 | 0.9533 | | 0.0676 | 2.37 | 1850 | 0.2116 | 0.9467 | | 0.0832 | 2.43 | 1900 | 0.1779 | 0.9533 | | 0.0899 | 2.49 | 1950 | 0.0999 | 0.9667 | | 0.0898 | 2.56 | 2000 | 0.1502 | 0.9467 | | 0.0955 | 2.62 | 2050 | 0.1776 | 0.9467 | | 0.0989 | 2.69 | 2100 | 0.1279 | 0.9533 | | 0.102 | 2.75 | 2150 | 0.1005 | 0.9667 | | 0.0957 | 2.81 | 2200 | 0.1070 | 0.9667 | | 0.0786 | 2.88 | 2250 | 0.1881 | 0.9467 | | 0.0897 | 2.94 | 2300 | 0.1951 | 0.9533 | | 0.0801 | 3.01 | 2350 | 0.1827 | 0.9467 | | 0.0829 | 3.07 | 2400 | 0.1854 | 0.96 | | 0.0665 | 3.13 | 2450 | 0.1775 | 0.9533 | | 0.0838 | 3.2 | 2500 | 0.1700 | 0.96 | | 0.0441 | 3.26 | 2550 | 0.1810 | 0.96 | | 0.071 | 3.32 | 2600 | 0.2083 | 0.9533 | | 0.0655 | 3.39 | 2650 | 0.1943 | 0.96 | | 0.0565 | 3.45 | 2700 | 0.2486 | 0.9533 | | 0.0669 | 3.52 | 2750 | 0.2540 | 0.9533 | | 0.0671 | 3.58 | 2800 | 0.2140 | 0.9467 | | 0.0857 | 3.64 | 2850 | 0.1609 | 0.9533 | | 0.0585 | 3.71 | 2900 | 0.2067 | 0.9467 | | 0.0597 | 3.77 | 2950 | 0.2380 | 0.9467 | | 0.0932 | 3.84 | 3000 | 0.1727 | 0.9533 | | 0.0744 | 3.9 | 3050 | 0.2099 | 0.9467 | | 0.0679 | 3.96 | 3100 | 0.2034 | 0.9467 | | 0.0447 | 4.03 | 3150 | 0.2461 | 0.9533 | | 0.0486 | 4.09 | 3200 | 0.2032 | 0.9533 | | 0.0409 | 4.16 | 3250 | 0.2468 | 0.9467 | | 0.0605 | 4.22 | 3300 | 0.2422 | 0.9467 | | 0.0319 | 4.28 | 3350 | 0.2681 | 0.9467 | | 0.0483 | 4.35 | 3400 | 0.2222 | 0.9533 | | 0.0801 | 4.41 | 3450 | 0.2247 | 0.9533 | | 0.0333 | 4.48 | 3500 | 0.2190 | 0.9533 | | 0.0432 | 4.54 | 3550 | 0.2167 | 0.9533 | | 0.0381 | 4.6 | 3600 | 0.2507 | 0.9467 | | 0.0647 | 4.67 | 3650 | 0.2410 | 0.9533 | | 0.0427 | 4.73 | 3700 | 0.2447 | 0.9467 | | 0.0627 | 4.8 | 3750 | 0.2332 | 0.9533 | | 0.0569 | 4.86 | 3800 | 0.2358 | 0.9533 | | 0.069 | 4.92 | 3850 | 0.2379 | 0.9533 | | 0.0474 | 4.99 | 3900 | 0.2383 | 0.9467 | ### Framework versions - Transformers 4.23.1 - Pytorch 1.13.0 - Datasets 2.6.1 - Tokenizers 0.13.1