metadata
license: apache-2.0
base_model: google-t5/t5-base
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: t5_base_patent
results: []
t5_base_patent
This model is a fine-tuned version of google-t5/t5-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.9276
- Accuracy: 0.6776
- F1 Macro: 0.6237
- F1 Micro: 0.6776
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: 0.0005
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- total_train_batch_size: 32
- total_eval_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Micro |
|---|---|---|---|---|---|---|
| 1.3522 | 0.06 | 50 | 1.4202 | 0.5254 | 0.3609 | 0.5254 |
| 1.1693 | 0.13 | 100 | 1.1674 | 0.597 | 0.4695 | 0.597 |
| 1.171 | 0.19 | 150 | 1.1373 | 0.6052 | 0.4713 | 0.6052 |
| 1.048 | 0.26 | 200 | 1.0826 | 0.6286 | 0.5499 | 0.6286 |
| 0.9991 | 0.32 | 250 | 1.0599 | 0.638 | 0.5422 | 0.638 |
| 1.1814 | 0.38 | 300 | 1.0633 | 0.6332 | 0.5593 | 0.6332 |
| 1.0864 | 0.45 | 350 | 1.0400 | 0.6392 | 0.5678 | 0.6392 |
| 0.9748 | 0.51 | 400 | 1.0440 | 0.6424 | 0.5613 | 0.6424 |
| 1.0267 | 0.58 | 450 | 1.0116 | 0.6526 | 0.5818 | 0.6526 |
| 1.0052 | 0.64 | 500 | 0.9948 | 0.657 | 0.5787 | 0.657 |
| 0.9244 | 0.7 | 550 | 1.0002 | 0.657 | 0.5870 | 0.657 |
| 1.0172 | 0.77 | 600 | 0.9869 | 0.661 | 0.5889 | 0.661 |
| 1.032 | 0.83 | 650 | 0.9922 | 0.658 | 0.5967 | 0.658 |
| 0.9623 | 0.9 | 700 | 0.9955 | 0.6488 | 0.5863 | 0.6488 |
| 0.9257 | 0.96 | 750 | 0.9993 | 0.6556 | 0.5884 | 0.6556 |
| 0.7956 | 1.02 | 800 | 0.9737 | 0.6662 | 0.6148 | 0.6662 |
| 0.8475 | 1.09 | 850 | 1.0125 | 0.6544 | 0.5729 | 0.6544 |
| 0.8527 | 1.15 | 900 | 0.9999 | 0.6524 | 0.5897 | 0.6524 |
| 0.8587 | 1.21 | 950 | 1.0072 | 0.6576 | 0.5873 | 0.6576 |
| 0.8855 | 1.28 | 1000 | 0.9840 | 0.6592 | 0.6035 | 0.6592 |
| 0.7015 | 1.34 | 1050 | 0.9847 | 0.6682 | 0.5993 | 0.6682 |
| 0.8116 | 1.41 | 1100 | 0.9702 | 0.6678 | 0.6079 | 0.6678 |
| 0.8409 | 1.47 | 1150 | 0.9789 | 0.6606 | 0.6017 | 0.6606 |
| 0.7889 | 1.53 | 1200 | 0.9462 | 0.6818 | 0.6125 | 0.6818 |
| 0.8059 | 1.6 | 1250 | 0.9375 | 0.6694 | 0.6093 | 0.6694 |
| 0.7893 | 1.66 | 1300 | 0.9467 | 0.6762 | 0.6102 | 0.6762 |
| 0.8152 | 1.73 | 1350 | 0.9396 | 0.6822 | 0.6158 | 0.6822 |
| 0.7644 | 1.79 | 1400 | 0.9445 | 0.6798 | 0.6190 | 0.6798 |
| 0.7252 | 1.85 | 1450 | 0.9285 | 0.688 | 0.6209 | 0.688 |
| 1.0028 | 1.92 | 1500 | 0.9379 | 0.6702 | 0.6079 | 0.6702 |
| 0.8056 | 1.98 | 1550 | 0.9276 | 0.6776 | 0.6237 | 0.6776 |
| 0.5781 | 2.05 | 1600 | 0.9509 | 0.6864 | 0.6215 | 0.6864 |
| 0.5592 | 2.11 | 1650 | 0.9535 | 0.6866 | 0.6354 | 0.6866 |
| 0.6818 | 2.17 | 1700 | 0.9812 | 0.682 | 0.6203 | 0.682 |
| 0.6022 | 2.24 | 1750 | 0.9842 | 0.6822 | 0.6270 | 0.6822 |
| 0.5771 | 2.3 | 1800 | 1.0100 | 0.6832 | 0.6295 | 0.6832 |
| 0.596 | 2.37 | 1850 | 1.0079 | 0.6784 | 0.6280 | 0.6784 |
| 0.5209 | 2.43 | 1900 | 1.0118 | 0.6828 | 0.6257 | 0.6828 |
| 0.4842 | 2.49 | 1950 | 1.0165 | 0.68 | 0.6253 | 0.68 |
| 0.6581 | 2.56 | 2000 | 1.0119 | 0.6774 | 0.6234 | 0.6774 |
| 0.6417 | 2.62 | 2050 | 1.0035 | 0.6834 | 0.6345 | 0.6834 |
| 0.5388 | 2.69 | 2100 | 1.0133 | 0.681 | 0.6321 | 0.681 |
| 0.546 | 2.75 | 2150 | 1.0133 | 0.6808 | 0.6313 | 0.6808 |
| 0.5825 | 2.81 | 2200 | 1.0058 | 0.683 | 0.6316 | 0.683 |
| 0.6251 | 2.88 | 2250 | 1.0062 | 0.6848 | 0.6357 | 0.6848 |
| 0.619 | 2.94 | 2300 | 1.0014 | 0.6826 | 0.6307 | 0.6826 |
Framework versions
- Transformers 4.39.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2