Downloads · 30 days
13
24% of all-time downloads
FYP_sq2_mrqa_adqa_quac
kxx-kkk/FYP_sq2_mrqa_adqa_quac
FYP_sq2_mrqa_adqa_quac is a question answering model from kxx-kkk. Use it when the input is a question plus a passage. It is set up for transformers. The card lists the license as cc-by-4.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
55
Public
Parameters
184M
8.8 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors735 MB · 99%
At a glance
- Task
- Question Answering
- Library
- transformers
- License
- cc-by-4.0
- Model type
- deberta-v2
- Access
- Public
- Created
- Apr 5, 2024
- Updated
- Apr 5, 2024
- SHA
- 037bc94a
Try a prompt
Base models
Model card
From the Hugging Face model README
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
FYP_sq2_mrqa_adqa_quac
This model is a fine-tuned version of kxx-kkk/FYP_sq2_mrqa_adqa on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.2206
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: 1e-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: 2
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.8023 | 0.48 | 500 | 3.4618 |
| 2.4772 | 0.95 | 1000 | 3.3024 |
| 2.4056 | 1.43 | 1500 | 3.2717 |
| 2.3162 | 1.91 | 2000 | 3.2206 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Task
- Question Answering
- Library
- transformers
- Type
- deberta-v2
- License
- cc-by-4.0
- Created
- Apr 5, 2024
- Updated
- Apr 5, 2024