<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
TSE_DistilBERT_5E
This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3301
- Accuracy: 0.9333
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 |
---|---|---|---|---|
0.6534 | 0.06 | 50 | 0.5269 | 0.8333 |
0.3926 | 0.12 | 100 | 0.2674 | 0.9133 |
0.275 | 0.17 | 150 | 0.2063 | 0.94 |
0.2341 | 0.23 | 200 | 0.1896 | 0.9333 |
0.2436 | 0.29 | 250 | 0.2132 | 0.9133 |
0.2561 | 0.35 | 300 | 0.2474 | 0.9 |
0.2536 | 0.4 | 350 | 0.2092 | 0.9267 |
0.2048 | 0.46 | 400 | 0.2135 | 0.92 |
0.2119 | 0.52 | 450 | 0.2382 | 0.9133 |
0.2152 | 0.58 | 500 | 0.2322 | 0.9267 |
0.2072 | 0.63 | 550 | 0.2182 | 0.9333 |
0.2134 | 0.69 | 600 | 0.2457 | 0.9133 |
0.2093 | 0.75 | 650 | 0.2476 | 0.92 |
0.2145 | 0.81 | 700 | 0.2489 | 0.9267 |
0.2191 | 0.87 | 750 | 0.2374 | 0.9267 |
0.2198 | 0.92 | 800 | 0.2347 | 0.92 |
0.2126 | 0.98 | 850 | 0.2015 | 0.9467 |
0.1373 | 1.04 | 900 | 0.2246 | 0.9467 |
0.1367 | 1.1 | 950 | 0.2875 | 0.9133 |
0.1726 | 1.15 | 1000 | 0.2641 | 0.94 |
0.1968 | 1.21 | 1050 | 0.2653 | 0.9333 |
0.1607 | 1.27 | 1100 | 0.2323 | 0.94 |
0.1437 | 1.33 | 1150 | 0.2900 | 0.9267 |
0.1707 | 1.38 | 1200 | 0.2430 | 0.94 |
0.1174 | 1.44 | 1250 | 0.2553 | 0.94 |
0.1662 | 1.5 | 1300 | 0.2442 | 0.9467 |
0.1374 | 1.56 | 1350 | 0.2365 | 0.9467 |
0.1632 | 1.61 | 1400 | 0.2794 | 0.9133 |
0.1558 | 1.67 | 1450 | 0.2428 | 0.94 |
0.1717 | 1.73 | 1500 | 0.2380 | 0.92 |
0.1301 | 1.79 | 1550 | 0.2006 | 0.94 |
0.1757 | 1.85 | 1600 | 0.2327 | 0.9467 |
0.1997 | 1.9 | 1650 | 0.2160 | 0.94 |
0.1611 | 1.96 | 1700 | 0.2797 | 0.92 |
0.1638 | 2.02 | 1750 | 0.2433 | 0.9333 |
0.1041 | 2.08 | 1800 | 0.2389 | 0.94 |
0.1172 | 2.13 | 1850 | 0.2381 | 0.9467 |
0.1332 | 2.19 | 1900 | 0.2650 | 0.94 |
0.1299 | 2.25 | 1950 | 0.2869 | 0.9333 |
0.0992 | 2.31 | 2000 | 0.2308 | 0.9533 |
0.1012 | 2.36 | 2050 | 0.2552 | 0.9467 |
0.0948 | 2.42 | 2100 | 0.2823 | 0.9267 |
0.1081 | 2.48 | 2150 | 0.2634 | 0.9467 |
0.1157 | 2.54 | 2200 | 0.2864 | 0.9333 |
0.1154 | 2.6 | 2250 | 0.2987 | 0.9267 |
0.1259 | 2.65 | 2300 | 0.2879 | 0.9333 |
0.1084 | 2.71 | 2350 | 0.2661 | 0.94 |
0.1342 | 2.77 | 2400 | 0.2711 | 0.94 |
0.12 | 2.83 | 2450 | 0.2362 | 0.9467 |
0.0839 | 2.88 | 2500 | 0.2712 | 0.9333 |
0.1546 | 2.94 | 2550 | 0.2433 | 0.9467 |
0.1321 | 3.0 | 2600 | 0.2421 | 0.9467 |
0.101 | 3.06 | 2650 | 0.2820 | 0.9333 |
0.061 | 3.11 | 2700 | 0.2990 | 0.9267 |
0.0608 | 3.17 | 2750 | 0.2512 | 0.9467 |
0.0983 | 3.23 | 2800 | 0.3033 | 0.9333 |
0.0806 | 3.29 | 2850 | 0.2621 | 0.9467 |
0.0788 | 3.34 | 2900 | 0.2672 | 0.9467 |
0.0827 | 3.4 | 2950 | 0.2797 | 0.9467 |
0.0912 | 3.46 | 3000 | 0.2802 | 0.9467 |
0.0771 | 3.52 | 3050 | 0.2693 | 0.9467 |
0.0842 | 3.58 | 3100 | 0.2758 | 0.9467 |
0.086 | 3.63 | 3150 | 0.2921 | 0.9333 |
0.1102 | 3.69 | 3200 | 0.3066 | 0.9333 |
0.1124 | 3.75 | 3250 | 0.2808 | 0.9333 |
0.0762 | 3.81 | 3300 | 0.2863 | 0.94 |
0.074 | 3.86 | 3350 | 0.3159 | 0.9333 |
0.062 | 3.92 | 3400 | 0.2977 | 0.9333 |
0.1027 | 3.98 | 3450 | 0.3449 | 0.9267 |
0.0734 | 4.04 | 3500 | 0.3165 | 0.9333 |
0.0375 | 4.09 | 3550 | 0.2960 | 0.9333 |
0.0377 | 4.15 | 3600 | 0.3245 | 0.9333 |
0.0661 | 4.21 | 3650 | 0.3262 | 0.9333 |
0.079 | 4.27 | 3700 | 0.3085 | 0.9333 |
0.0801 | 4.33 | 3750 | 0.3219 | 0.9333 |
0.0865 | 4.38 | 3800 | 0.3336 | 0.9267 |
0.058 | 4.44 | 3850 | 0.3083 | 0.9333 |
0.0689 | 4.5 | 3900 | 0.3351 | 0.9267 |
0.0345 | 4.56 | 3950 | 0.3412 | 0.9267 |
0.0557 | 4.61 | 4000 | 0.3236 | 0.9333 |
0.0758 | 4.67 | 4050 | 0.3224 | 0.9333 |
0.0682 | 4.73 | 4100 | 0.3241 | 0.9333 |
0.0534 | 4.79 | 4150 | 0.3349 | 0.9333 |
0.0707 | 4.84 | 4200 | 0.3254 | 0.9333 |
0.0672 | 4.9 | 4250 | 0.3277 | 0.9333 |
0.1033 | 4.96 | 4300 | 0.3301 | 0.9333 |
Framework versions
- Transformers 4.24.0
- Pytorch 1.13.0
- Datasets 2.3.2
- Tokenizers 0.13.1