Downloads · 30 days
62
4% of all-time downloads
MRR-sqv2-long
am-infoweb/MRR-sqv2-long
MRR-sqv2-long is a question answering model from am-infoweb. Use it when the input is a question plus a passage. It is set up for transformers.
should probably proofread and complete it, then remove this comment. --
All-time downloads
1.4K
Public
Parameters
148M
310 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin592 MB · 50%
At a glance
- Task
- Question Answering
- Library
- transformers
- Model type
- longformer
- Access
- Public
- Created
- Aug 3, 2023
- Updated
- Sep 18, 2023
- SHA
- bbf7d0db
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. -->
MRR-sqv2-long
This model is a fine-tuned version of mrm8488/longformer-base-4096-finetuned-squadv2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.2581
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: 1
- eval_batch_size: 1
- 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 |
|---|---|---|---|
| 1.215 | 1.0 | 11811 | 1.0751 |
| 0.8617 | 2.0 | 23622 | 0.9498 |
| 0.8639 | 3.0 | 35433 | 1.0341 |
| 0.7073 | 4.0 | 47244 | 0.9740 |
| 0.6156 | 5.0 | 59055 | 1.0663 |
| 0.5489 | 6.0 | 70866 | 1.0665 |
| 0.4932 | 7.0 | 82677 | 1.2496 |
| 0.527 | 8.0 | 94488 | 1.1351 |
| 0.3722 | 9.0 | 106299 | 1.1568 |
| 0.3599 | 10.0 | 118110 | 1.2581 |
Framework versions
- Transformers 4.32.0.dev0
- Pytorch 2.0.1+cu117
- Datasets 2.13.1
- Tokenizers 0.13.3
- Task
- Question Answering
- Library
- transformers
- Type
- longformer
- Created
- Aug 3, 2023
- Updated
- Sep 18, 2023