--- license: mit tags: - generated_from_trainer datasets: - tweet_eval metrics: - accuracy model-index: - name: TweetEval_roBERTa_5E results: - task: name: Text Classification type: text-classification dataset: name: tweet_eval type: tweet_eval config: sentiment split: train args: sentiment metrics: - name: Accuracy type: accuracy value: 0.9466666666666667 --- # TweetEval_roBERTa_5E This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on the tweet_eval dataset. It achieves the following results on the evaluation set: - Loss: 0.2770 - 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: 16 - eval_batch_size: 8 - 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 | |:-------------:|:-----:|:----:|:---------------:|:--------:| | 0.5967 | 0.04 | 50 | 0.4851 | 0.7333 | | 0.4085 | 0.08 | 100 | 0.2177 | 0.9333 | | 0.3449 | 0.12 | 150 | 0.2164 | 0.9333 | | 0.2739 | 0.16 | 200 | 0.2285 | 0.9267 | | 0.2588 | 0.2 | 250 | 0.2748 | 0.92 | | 0.3406 | 0.24 | 300 | 0.1956 | 0.9467 | | 0.2726 | 0.28 | 350 | 0.2285 | 0.92 | | 0.2645 | 0.32 | 400 | 0.2192 | 0.9267 | | 0.2549 | 0.37 | 450 | 0.2115 | 0.9333 | | 0.2387 | 0.41 | 500 | 0.2230 | 0.9333 | | 0.2415 | 0.45 | 550 | 0.2156 | 0.94 | | 0.2829 | 0.49 | 600 | 0.2575 | 0.9267 | | 0.2865 | 0.53 | 650 | 0.1572 | 0.9467 | | 0.2107 | 0.57 | 700 | 0.1437 | 0.9467 | | 0.2609 | 0.61 | 750 | 0.1595 | 0.94 | | 0.2234 | 0.65 | 800 | 0.2611 | 0.9333 | | 0.266 | 0.69 | 850 | 0.1544 | 0.9467 | | 0.2407 | 0.73 | 900 | 0.2145 | 0.9333 | | 0.2529 | 0.77 | 950 | 0.1861 | 0.9333 | | 0.2083 | 0.81 | 1000 | 0.1448 | 0.9533 | | 0.2942 | 0.85 | 1050 | 0.1703 | 0.9333 | | 0.1916 | 0.89 | 1100 | 0.1831 | 0.94 | | 0.2425 | 0.93 | 1150 | 0.2349 | 0.9333 | | 0.2521 | 0.97 | 1200 | 0.1268 | 0.94 | | 0.1742 | 1.01 | 1250 | 0.1782 | 0.9333 | | 0.172 | 1.06 | 1300 | 0.2636 | 0.9333 | | 0.1487 | 1.1 | 1350 | 0.1987 | 0.9467 | | 0.1805 | 1.14 | 1400 | 0.3030 | 0.9333 | | 0.1295 | 1.18 | 1450 | 0.2229 | 0.94 | | 0.2114 | 1.22 | 1500 | 0.1441 | 0.9467 | | 0.1714 | 1.26 | 1550 | 0.2157 | 0.9467 | | 0.1886 | 1.3 | 1600 | 0.2353 | 0.9267 | | 0.1666 | 1.34 | 1650 | 0.2572 | 0.94 | | 0.2254 | 1.38 | 1700 | 0.1569 | 0.9467 | | 0.1531 | 1.42 | 1750 | 0.2351 | 0.9333 | | 0.2174 | 1.46 | 1800 | 0.2137 | 0.9267 | | 0.2015 | 1.5 | 1850 | 0.2234 | 0.94 | | 0.1785 | 1.54 | 1900 | 0.1944 | 0.9333 | | 0.1954 | 1.58 | 1950 | 0.2013 | 0.9467 | | 0.1481 | 1.62 | 2000 | 0.2196 | 0.94 | | 0.1426 | 1.66 | 2050 | 0.2005 | 0.9467 | | 0.1951 | 1.7 | 2100 | 0.2281 | 0.9467 | | 0.1943 | 1.75 | 2150 | 0.1934 | 0.94 | | 0.2027 | 1.79 | 2200 | 0.1845 | 0.96 | | 0.2119 | 1.83 | 2250 | 0.1338 | 0.9533 | | 0.208 | 1.87 | 2300 | 0.1605 | 0.94 | | 0.1972 | 1.91 | 2350 | 0.1460 | 0.9533 | | 0.1876 | 1.95 | 2400 | 0.1488 | 0.9467 | | 0.1923 | 1.99 | 2450 | 0.2055 | 0.9533 | | 0.1391 | 2.03 | 2500 | 0.2245 | 0.9533 | | 0.1416 | 2.07 | 2550 | 0.2194 | 0.9533 | | 0.1521 | 2.11 | 2600 | 0.2234 | 0.9533 | | 0.0943 | 2.15 | 2650 | 0.2114 | 0.9533 | | 0.1452 | 2.19 | 2700 | 0.1772 | 0.9467 | | 0.1148 | 2.23 | 2750 | 0.2541 | 0.9333 | | 0.1706 | 2.27 | 2800 | 0.2151 | 0.9533 | | 0.12 | 2.31 | 2850 | 0.2521 | 0.9467 | | 0.181 | 2.35 | 2900 | 0.2518 | 0.9467 | | 0.1308 | 2.39 | 2950 | 0.2610 | 0.9533 | | 0.1482 | 2.44 | 3000 | 0.1789 | 0.9533 | | 0.1019 | 2.48 | 3050 | 0.2377 | 0.9467 | | 0.1474 | 2.52 | 3100 | 0.2468 | 0.94 | | 0.0843 | 2.56 | 3150 | 0.3056 | 0.94 | | 0.1521 | 2.6 | 3200 | 0.2067 | 0.96 | | 0.1333 | 2.64 | 3250 | 0.1921 | 0.94 | | 0.1318 | 2.68 | 3300 | 0.1699 | 0.96 | | 0.1503 | 2.72 | 3350 | 0.2186 | 0.94 | | 0.1242 | 2.76 | 3400 | 0.2322 | 0.94 | | 0.1179 | 2.8 | 3450 | 0.2313 | 0.9467 | | 0.1247 | 2.84 | 3500 | 0.2298 | 0.9467 | | 0.1289 | 2.88 | 3550 | 0.2502 | 0.94 | | 0.1597 | 2.92 | 3600 | 0.1875 | 0.9467 | | 0.1645 | 2.96 | 3650 | 0.2469 | 0.94 | | 0.1366 | 3.0 | 3700 | 0.2469 | 0.94 | | 0.1418 | 3.04 | 3750 | 0.2457 | 0.9467 | | 0.1146 | 3.08 | 3800 | 0.2188 | 0.9467 | | 0.091 | 3.12 | 3850 | 0.2476 | 0.94 | | 0.0972 | 3.17 | 3900 | 0.2791 | 0.94 | | 0.0976 | 3.21 | 3950 | 0.2933 | 0.9333 | | 0.0872 | 3.25 | 4000 | 0.2877 | 0.9467 | | 0.0857 | 3.29 | 4050 | 0.2664 | 0.9467 | | 0.1368 | 3.33 | 4100 | 0.2533 | 0.9467 | | 0.0713 | 3.37 | 4150 | 0.2855 | 0.9467 | | 0.1101 | 3.41 | 4200 | 0.2716 | 0.9533 | | 0.0871 | 3.45 | 4250 | 0.2654 | 0.9467 | | 0.1152 | 3.49 | 4300 | 0.2449 | 0.9467 | | 0.0441 | 3.53 | 4350 | 0.2904 | 0.9467 | | 0.1503 | 3.57 | 4400 | 0.2784 | 0.9467 | | 0.0763 | 3.61 | 4450 | 0.2804 | 0.9467 | | 0.083 | 3.65 | 4500 | 0.3278 | 0.94 | | 0.1111 | 3.69 | 4550 | 0.2899 | 0.9333 | | 0.0791 | 3.73 | 4600 | 0.3137 | 0.9333 | | 0.0837 | 3.77 | 4650 | 0.2799 | 0.9467 | | 0.1048 | 3.81 | 4700 | 0.2496 | 0.9533 | | 0.1031 | 3.86 | 4750 | 0.2689 | 0.9533 | | 0.0837 | 3.9 | 4800 | 0.2753 | 0.9533 | | 0.0929 | 3.94 | 4850 | 0.2357 | 0.9467 | | 0.0856 | 3.98 | 4900 | 0.2615 | 0.9467 | | 0.0619 | 4.02 | 4950 | 0.2983 | 0.9467 | | 0.0974 | 4.06 | 5000 | 0.2706 | 0.9533 | | 0.0548 | 4.1 | 5050 | 0.2978 | 0.9467 | | 0.0425 | 4.14 | 5100 | 0.3217 | 0.9333 | | 0.0808 | 4.18 | 5150 | 0.3054 | 0.94 | | 0.0466 | 4.22 | 5200 | 0.3142 | 0.94 | | 0.0593 | 4.26 | 5250 | 0.3193 | 0.9267 | | 0.0551 | 4.3 | 5300 | 0.3017 | 0.9333 | | 0.0493 | 4.34 | 5350 | 0.2954 | 0.94 | | 0.0897 | 4.38 | 5400 | 0.2912 | 0.9467 | | 0.0529 | 4.42 | 5450 | 0.2956 | 0.94 | | 0.0924 | 4.46 | 5500 | 0.2858 | 0.94 | | 0.1018 | 4.5 | 5550 | 0.2826 | 0.94 | | 0.1137 | 4.55 | 5600 | 0.2711 | 0.94 | | 0.0667 | 4.59 | 5650 | 0.2776 | 0.94 | | 0.0521 | 4.63 | 5700 | 0.2955 | 0.94 | | 0.0334 | 4.67 | 5750 | 0.2972 | 0.94 | | 0.0298 | 4.71 | 5800 | 0.3133 | 0.94 | | 0.1261 | 4.75 | 5850 | 0.2891 | 0.9467 | | 0.0514 | 4.79 | 5900 | 0.2804 | 0.9467 | | 0.0416 | 4.83 | 5950 | 0.2809 | 0.94 | | 0.0745 | 4.87 | 6000 | 0.2774 | 0.9467 | | 0.1134 | 4.91 | 6050 | 0.2715 | 0.9467 | | 0.0446 | 4.95 | 6100 | 0.2748 | 0.9467 | | 0.0581 | 4.99 | 6150 | 0.2770 | 0.9467 | ### Framework versions - Transformers 4.24.0 - Pytorch 1.13.0 - Datasets 2.3.2 - Tokenizers 0.13.2