Downloads · 30 days
11
13% of all-time downloads
bert-base-uncased-finetuned-squad
victoraavila/bert-base-uncased-finetuned-squad
bert-base-uncased-finetuned-squad is a question answering model from victoraavila. Use it when the input is a question plus a passage. It is set up for transformers. The card lists the license as apache-2.0.
should probably proofread and complete it, then remove this comment. --
All-time downloads
83
Public
Repo size
871 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin436 MB · 100%
At a glance
- Task
- Question Answering
- Library
- transformers
- License
- apache-2.0
- Model type
- bert
- Access
- Public
- Created
- Mar 2, 2022
- Updated
- Aug 25, 2021
- SHA
- c34524b1
Try a prompt
Datasets
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. -->
bert-base-uncased-finetuned-squad
This model is a fine-tuned version of bert-base-uncased on the SQuAD1.1 dataset. It was trained through Transformers' example Colab notebook on Question Answering, available here. It achieves the following results on the evaluation set:
- Loss: 1.0780
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. They are equal to the ones used to fine-tune distilbert-base-uncased for QA:
- 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: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.0706 | 1.0 | 5533 | 1.0250 |
| 0.7899 | 2.0 | 11066 | 1.0356 |
| 0.5991 | 3.0 | 16599 | 1.0780 |
Validation results
| EM | F1 |
|---|---|
| 80.3690 | 88.0110 |
Framework versions
- Transformers 4.9.2
- Pytorch 1.9.0+cu102
- Datasets 1.11.0
- Tokenizers 0.10.3
- Task
- Question Answering
- Library
- transformers
- Type
- bert
- License
- apache-2.0
- Created
- Mar 2, 2022
- Updated
- Aug 25, 2021