metadata
tags:
- generated_from_keras_callback
model-index:
- name: madatnlp/ke-t5-scratch
results: []
madatnlp/ke-t5-scratch
This model is a fine-tuned version of madatnlp/ke-t5-math-py on an unknown dataset. It achieves the following results on the evaluation set:
- Train Loss: 1.8367
- Validation Loss: 1.5850
- Epoch: 88
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:
- optimizer: {'name': 'Adam', 'learning_rate': 1e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
- training_precision: float32
Training results
| Train Loss | Validation Loss | Epoch |
|---|---|---|
| 13.5076 | 11.8125 | 0 |
| 11.0983 | 9.4857 | 1 |
| 9.4413 | 7.9593 | 2 |
| 8.2675 | 6.9802 | 3 |
| 7.3769 | 6.1898 | 4 |
| 6.6978 | 5.6209 | 5 |
| 6.2266 | 5.1054 | 6 |
| 5.7871 | 4.9395 | 7 |
| 5.4937 | 4.6256 | 8 |
| 5.2013 | 4.4694 | 9 |
| 4.9649 | 4.1716 | 10 |
| 4.7273 | 4.0317 | 11 |
| 4.5237 | 3.7622 | 12 |
| 4.3581 | 3.4826 | 13 |
| 4.2078 | 3.4463 | 14 |
| 4.0755 | 3.2685 | 15 |
| 3.9494 | 3.1492 | 16 |
| 3.8338 | 3.1535 | 17 |
| 3.6767 | 2.8725 | 18 |
| 3.6546 | 3.1201 | 19 |
| 3.5395 | 3.0338 | 20 |
| 3.4086 | 2.9991 | 21 |
| 3.3886 | 2.8730 | 22 |
| 3.2900 | 2.8334 | 23 |
| 3.2906 | 2.6087 | 24 |
| 3.1844 | 2.6765 | 25 |
| 3.1672 | 2.6972 | 26 |
| 3.1023 | 2.5778 | 27 |
| 3.0528 | 2.5352 | 28 |
| 2.9885 | 2.5250 | 29 |
| 2.9455 | 2.6048 | 30 |
| 2.9025 | 2.3874 | 31 |
| 2.9228 | 2.4521 | 32 |
| 2.8160 | 2.2810 | 33 |
| 2.7895 | 2.3317 | 34 |
| 2.7372 | 2.3300 | 35 |
| 2.7494 | 2.3160 | 36 |
| 2.7219 | 2.3736 | 37 |
| 2.6818 | 2.3031 | 38 |
| 2.6464 | 2.2736 | 39 |
| 2.5834 | 2.2104 | 40 |
| 2.5779 | 2.0641 | 41 |
| 2.5577 | 2.0439 | 42 |
| 2.5212 | 2.0828 | 43 |
| 2.5029 | 2.1416 | 44 |
| 2.4391 | 2.0837 | 45 |
| 2.4556 | 2.0950 | 46 |
| 2.4138 | 1.8874 | 47 |
| 2.4138 | 1.9967 | 48 |
| 2.3698 | 2.0096 | 49 |
| 2.3776 | 1.9152 | 50 |
| 2.3011 | 2.0284 | 51 |
| 2.3454 | 2.0002 | 52 |
| 2.2767 | 1.9544 | 53 |
| 2.2332 | 1.8651 | 54 |
| 2.2900 | 1.9383 | 55 |
| 2.2442 | 1.8779 | 56 |
| 2.2183 | 1.8790 | 57 |
| 2.1824 | 1.7470 | 58 |
| 2.1648 | 1.7715 | 59 |
| 2.1859 | 1.8188 | 60 |
| 2.1529 | 1.7747 | 61 |
| 2.1343 | 1.8870 | 62 |
| 2.1344 | 1.8471 | 63 |
| 2.0876 | 1.8135 | 64 |
| 2.0775 | 1.7311 | 65 |
| 2.0557 | 1.8648 | 66 |
| 2.1017 | 1.6826 | 67 |
| 2.0649 | 1.7404 | 68 |
| 2.0505 | 1.6182 | 69 |
| 2.0084 | 1.6731 | 70 |
| 2.0143 | 1.6890 | 71 |
| 1.9882 | 1.6767 | 72 |
| 1.9759 | 1.5758 | 73 |
| 1.9800 | 1.7079 | 74 |
| 1.9602 | 1.6354 | 75 |
| 1.9580 | 1.6015 | 76 |
| 1.9401 | 1.5779 | 77 |
| 1.9070 | 1.5071 | 78 |
| 1.9304 | 1.5554 | 79 |
| 1.8987 | 1.5434 | 80 |
| 1.8927 | 1.6711 | 81 |
| 1.9044 | 1.5399 | 82 |
| 1.8664 | 1.5820 | 83 |
| 1.8860 | 1.5097 | 84 |
| 1.8043 | 1.5495 | 85 |
| 1.8571 | 1.5327 | 86 |
| 1.8285 | 1.5381 | 87 |
| 1.8367 | 1.5850 | 88 |
Framework versions
- Transformers 4.18.0
- TensorFlow 2.8.0
- Datasets 2.1.0
- Tokenizers 0.12.1