File size: 1,807 Bytes
637cec8 870ac1a 637cec8 870ac1a 637cec8 3a46a82 637cec8 3b5f3fc 637cec8 026dd9a 637cec8 3a46a82 637cec8 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 |
---
license: apache-2.0
base_model: distilbert-base-uncased
tags:
- question-answering
- nlp
- generated_from_trainer
model-index:
- name: distilbert-qa-mash-covid
results: []
---
<!-- 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. -->
# distilbert-qa-mash-covid
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the mashqa_dataset dataset.
It achieves the following results on the evaluation set:
- Loss: 1.6691
## 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 |
|:-------------:|:-----:|:----:|:---------------:|
| 1.9281 | 1.0 | 657 | 1.4079 |
| 1.4703 | 2.0 | 1314 | 1.2640 |
| 1.3062 | 3.0 | 1971 | 1.2448 |
| 1.0172 | 4.0 | 2628 | 1.2780 |
| 0.8991 | 5.0 | 3285 | 1.3009 |
| 0.8004 | 6.0 | 3942 | 1.4255 |
| 0.6502 | 7.0 | 4599 | 1.4567 |
| 0.5806 | 8.0 | 5256 | 1.5352 |
| 0.5117 | 9.0 | 5913 | 1.6334 |
| 0.4538 | 10.0 | 6570 | 1.6691 |
### Framework versions
- Transformers 4.39.3
- Pytorch 2.1.2
- Datasets 2.18.0
- Tokenizers 0.15.2
|