# Load model directly
from transformers import AutoTokenizer, AutoModelForQuestionAnswering
tokenizer = AutoTokenizer.from_pretrained("seungkim1313/qa_model")
model = AutoModelForQuestionAnswering.from_pretrained("seungkim1313/qa_model", device_map="auto")Quick Links
qa_model
This model is a fine-tuned version of deepset/minilm-uncased-squad2 on the squad_kor_v1 dataset. It achieves the following results on the evaluation set:
- Loss: 3.2803
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: 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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.4482 | 1.0 | 25 | 3.8476 |
| 4.1886 | 2.0 | 50 | 3.3495 |
| 2.8781 | 3.0 | 75 | 3.2032 |
| 3.5417 | 4.0 | 100 | 3.3601 |
| 2.1682 | 5.0 | 125 | 3.2218 |
| 3.1787 | 6.0 | 150 | 3.3264 |
| 2.814 | 7.0 | 175 | 3.3053 |
| 2.7755 | 8.0 | 200 | 3.2801 |
| 1.9859 | 9.0 | 225 | 3.4267 |
| 2.1119 | 10.0 | 250 | 3.2803 |
Framework versions
- Transformers 4.32.0
- Pytorch 2.0.1+cu118
- Datasets 2.14.4
- Tokenizers 0.13.3
- Downloads last month
- 7
Model tree for seungkim1313/qa_model
Base model
deepset/minilm-uncased-squad2
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("question-answering", model="seungkim1313/qa_model")