bert-large-uncased-finetuned-docvqa
This model is a fine-tuned version of bert-large-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.6367
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: 2e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 250500
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.5228 | 0.05 | 1000 | 2.6645 |
| 2.4909 | 0.1 | 2000 | 2.8985 |
| 2.1679 | 0.16 | 3000 | 2.3551 |
| 1.9451 | 0.21 | 4000 | 2.2226 |
| 1.6814 | 0.26 | 5000 | 2.1590 |
| 1.8868 | 0.31 | 6000 | 2.6197 |
| 1.6618 | 0.36 | 7000 | 2.3632 |
| 1.8313 | 0.41 | 8000 | 2.4519 |
| 1.7017 | 0.47 | 9000 | 2.2682 |
| 1.8169 | 0.52 | 10000 | 2.4486 |
| 1.7074 | 0.57 | 11000 | 2.3862 |
| 1.7674 | 0.62 | 12000 | 2.1801 |
| 1.8134 | 0.67 | 13000 | 2.3032 |
| 1.8334 | 0.73 | 14000 | 2.4205 |
| 1.6819 | 0.78 | 15000 | 2.2398 |
| 1.5846 | 0.83 | 16000 | 2.3834 |
| 1.6758 | 0.88 | 17000 | 1.9683 |
| 1.6303 | 0.93 | 18000 | 2.3297 |
| 1.5652 | 0.98 | 19000 | 2.0581 |
| 1.3045 | 1.04 | 20000 | 2.4950 |
| 1.2393 | 1.09 | 21000 | 2.6622 |
| 1.1526 | 1.14 | 22000 | 2.3749 |
| 1.2631 | 1.19 | 23000 | 2.3915 |
| 1.1846 | 1.24 | 24000 | 2.2592 |
| 1.2731 | 1.3 | 25000 | 2.4239 |
| 1.3057 | 1.35 | 26000 | 2.2920 |
| 1.134 | 1.4 | 27000 | 2.3107 |
| 1.2017 | 1.45 | 28000 | 2.4271 |
| 1.2202 | 1.5 | 29000 | 2.1814 |
| 1.2179 | 1.56 | 30000 | 2.3365 |
| 1.2359 | 1.61 | 31000 | 2.1256 |
| 1.1964 | 1.66 | 32000 | 2.1720 |
| 1.269 | 1.71 | 33000 | 2.4363 |
| 1.1812 | 1.76 | 34000 | 2.2372 |
| 1.2187 | 1.81 | 35000 | 2.2318 |
| 1.1805 | 1.87 | 36000 | 2.3693 |
| 1.1458 | 1.92 | 37000 | 2.5128 |
| 1.1958 | 1.97 | 38000 | 2.1311 |
| 0.8924 | 2.02 | 39000 | 2.4635 |
| 0.869 | 2.07 | 40000 | 2.8231 |
| 0.8333 | 2.13 | 41000 | 2.6762 |
| 0.9194 | 2.18 | 42000 | 2.4588 |
| 0.8089 | 2.23 | 43000 | 2.6443 |
| 0.8612 | 2.28 | 44000 | 2.4300 |
| 0.7981 | 2.33 | 45000 | 2.7418 |
| 0.9765 | 2.38 | 46000 | 2.6543 |
| 0.8646 | 2.44 | 47000 | 2.5990 |
| 1.0316 | 2.49 | 48000 | 2.4625 |
| 0.9862 | 2.54 | 49000 | 2.4691 |
| 1.027 | 2.59 | 50000 | 2.4156 |
| 0.9412 | 2.64 | 51000 | 2.4204 |
| 0.9353 | 2.7 | 52000 | 2.4933 |
| 0.9509 | 2.75 | 53000 | 2.4708 |
| 0.9351 | 2.8 | 54000 | 2.5351 |
| 0.9968 | 2.85 | 55000 | 2.2506 |
| 1.025 | 2.9 | 56000 | 2.6317 |
| 1.627 | 2.95 | 57000 | 2.7843 |
| 0.9294 | 3.01 | 58000 | 2.9396 |
| 0.6043 | 3.06 | 59000 | 3.1560 |
| 0.7903 | 3.11 | 60000 | 2.8330 |
| 0.7373 | 3.16 | 61000 | 2.9422 |
| 0.6499 | 3.21 | 62000 | 3.0948 |
| 0.6411 | 3.27 | 63000 | 2.7900 |
| 0.625 | 3.32 | 64000 | 2.5268 |
| 0.6264 | 3.37 | 65000 | 2.8701 |
| 0.6143 | 3.42 | 66000 | 3.2544 |
| 0.6286 | 3.47 | 67000 | 2.6208 |
| 0.739 | 3.53 | 68000 | 2.8107 |
| 0.5981 | 3.58 | 69000 | 2.8073 |
| 0.6502 | 3.63 | 70000 | 2.6293 |
| 0.6548 | 3.68 | 71000 | 2.9501 |
| 0.7243 | 3.73 | 72000 | 2.7917 |
| 0.598 | 3.78 | 73000 | 2.9341 |
| 0.6159 | 3.84 | 74000 | 2.7629 |
| 0.5905 | 3.89 | 75000 | 2.6441 |
| 0.6393 | 3.94 | 76000 | 2.6660 |
| 0.677 | 3.99 | 77000 | 2.7616 |
| 0.3281 | 4.04 | 78000 | 3.6873 |
| 0.4524 | 4.1 | 79000 | 3.3441 |
| 0.3994 | 4.15 | 80000 | 3.3129 |
| 0.4686 | 4.2 | 81000 | 3.1813 |
| 0.5293 | 4.25 | 82000 | 2.9088 |
| 0.3961 | 4.3 | 83000 | 3.0765 |
| 0.4406 | 4.35 | 84000 | 3.1254 |
| 0.401 | 4.41 | 85000 | 3.2415 |
| 0.4594 | 4.46 | 86000 | 3.0691 |
| 0.4523 | 4.51 | 87000 | 3.0493 |
| 0.4719 | 4.56 | 88000 | 3.1352 |
| 0.4895 | 4.61 | 89000 | 2.8991 |
| 0.423 | 4.67 | 90000 | 3.1738 |
| 0.3984 | 4.72 | 91000 | 3.1862 |
| 0.4206 | 4.77 | 92000 | 3.1213 |
| 0.4587 | 4.82 | 93000 | 3.0030 |
| 0.381 | 4.87 | 94000 | 3.3218 |
| 0.4138 | 4.92 | 95000 | 3.1529 |
| 0.4003 | 4.98 | 96000 | 3.1375 |
| 0.2098 | 5.03 | 97000 | 3.7443 |
| 0.2334 | 5.08 | 98000 | 3.7359 |
| 0.2534 | 5.13 | 99000 | 3.7814 |
| 0.3067 | 5.18 | 100000 | 3.7128 |
| 0.2363 | 5.24 | 101000 | 3.6091 |
| 0.2652 | 5.29 | 102000 | 3.4015 |
| 0.3311 | 5.34 | 103000 | 3.4793 |
| 0.2344 | 5.39 | 104000 | 3.6792 |
| 0.2741 | 5.44 | 105000 | 3.5385 |
| 0.2896 | 5.5 | 106000 | 3.8118 |
| 0.2071 | 5.55 | 107000 | 3.8690 |
| 0.3023 | 5.6 | 108000 | 3.7087 |
| 0.3299 | 5.65 | 109000 | 3.4925 |
| 0.1943 | 5.7 | 110000 | 3.6739 |
| 0.2488 | 5.75 | 111000 | 3.7614 |
| 0.3138 | 5.81 | 112000 | 3.5156 |
| 0.2555 | 5.86 | 113000 | 3.6056 |
| 0.2918 | 5.91 | 114000 | 3.6533 |
| 0.2751 | 5.96 | 115000 | 3.6367 |
Framework versions
- Transformers 4.10.0
- Pytorch 1.8.0+cu101
- Datasets 1.11.0
- Tokenizers 0.10.3
New
Select AutoNLP in the “Train” menu to fine-tune this model automatically.
- Downloads last month
- 4