# Load model directly
from transformers import AutoTokenizer, AutoModelForQuestionAnswering
tokenizer = AutoTokenizer.from_pretrained("hung200504/bert-covidqa-3")
model = AutoModelForQuestionAnswering.from_pretrained("hung200504/bert-covidqa-3", device_map="auto")Quick Links
bert-covidqa-3
This model is a fine-tuned version of deepset/bert-base-uncased-squad2 on the covid_qa_deepset dataset. It achieves the following results on the evaluation set:
- Loss: 0.3717
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: 3e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.6653 | 0.04 | 5 | 0.4879 |
| 0.2392 | 0.09 | 10 | 0.4815 |
| 0.4918 | 0.13 | 15 | 0.4405 |
| 0.3634 | 0.18 | 20 | 0.4156 |
| 0.6494 | 0.22 | 25 | 0.3953 |
| 0.2573 | 0.26 | 30 | 0.3845 |
| 0.3645 | 0.31 | 35 | 0.3737 |
| 0.5168 | 0.35 | 40 | 0.3656 |
| 0.5341 | 0.39 | 45 | 0.3680 |
| 0.4362 | 0.44 | 50 | 0.3774 |
| 0.5495 | 0.48 | 55 | 0.3692 |
| 0.5316 | 0.53 | 60 | 0.3496 |
| 0.4068 | 0.57 | 65 | 0.3414 |
| 0.4793 | 0.61 | 70 | 0.3470 |
| 0.7173 | 0.66 | 75 | 0.3517 |
| 0.5335 | 0.7 | 80 | 0.3646 |
| 0.7152 | 0.75 | 85 | 0.3848 |
| 0.7003 | 0.79 | 90 | 0.3962 |
| 0.2466 | 0.83 | 95 | 0.3971 |
| 0.415 | 0.88 | 100 | 0.3879 |
| 0.4797 | 0.92 | 105 | 0.3767 |
| 0.7039 | 0.96 | 110 | 0.3717 |
Framework versions
- Transformers 4.34.1
- Pytorch 2.1.0+cu118
- Datasets 2.14.5
- Tokenizers 0.14.1
- Downloads last month
- 3
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("question-answering", model="hung200504/bert-covidqa-3")