pos-morph-analysis-eng
This model is a fine-tuned version of om-ashish-soni/pos-morph-analysis-eng on the universal_dependencies dataset. It achieves the following results on the evaluation set:
- Loss: 0.2354
- Precision: 0.9547
- Recall: 0.9594
- F1: 0.9571
- Accuracy: 0.9574
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 8
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
No log | 1.0 | 99 | 0.2425 | 0.9476 | 0.9523 | 0.9499 | 0.9505 |
No log | 1.99 | 198 | 0.2253 | 0.9504 | 0.9553 | 0.9528 | 0.9540 |
No log | 2.99 | 297 | 0.2273 | 0.9511 | 0.9565 | 0.9538 | 0.9548 |
No log | 4.0 | 397 | 0.2348 | 0.9512 | 0.9559 | 0.9536 | 0.9541 |
No log | 5.0 | 496 | 0.2294 | 0.9539 | 0.9586 | 0.9562 | 0.9574 |
0.0728 | 5.99 | 595 | 0.2319 | 0.9547 | 0.9594 | 0.9570 | 0.9574 |
0.0728 | 6.99 | 694 | 0.2405 | 0.9540 | 0.9585 | 0.9562 | 0.9566 |
0.0728 | 7.98 | 792 | 0.2354 | 0.9547 | 0.9594 | 0.9571 | 0.9574 |
Framework versions
- Transformers 4.33.1
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3
- Downloads last month
- 1
Finetuned from
Dataset used to train om-ashish-soni/pos-morph-analysis-eng
Evaluation results
- Precision on universal_dependenciesvalidation set self-reported0.955
- Recall on universal_dependenciesvalidation set self-reported0.959
- F1 on universal_dependenciesvalidation set self-reported0.957
- Accuracy on universal_dependenciesvalidation set self-reported0.957