File size: 2,217 Bytes
843a927 |
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 70 71 72 73 74 75 76 77 78 79 80 81 |
---
license: mit
tags:
- generated_from_trainer
datasets:
- pritamdeka/cord-19-abstract
metrics:
- accuracy
model-index:
- name: pubmedbert-abstract-cord19
results:
- task:
name: Masked Language Modeling
type: fill-mask
dataset:
name: pritamdeka/cord-19-abstract
type: pritamdeka/cord-19-abstract
args: fulltext
metrics:
- name: Accuracy
type: accuracy
value: 0.7246798699728464
---
<!-- 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. -->
# pubmedbert-abstract-cord19
This model is a fine-tuned version of [microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext](https://huggingface.co./microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext) on the pritamdeka/cord-19-abstract dataset.
It achieves the following results on the evaluation set:
- Loss: 1.2371
- Accuracy: 0.7247
## 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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 10000
- num_epochs: 4.0
- mixed_precision_training: Native AMP
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
| 1.27 | 0.53 | 5000 | 1.2425 | 0.7236 |
| 1.2634 | 1.06 | 10000 | 1.3123 | 0.7141 |
| 1.3041 | 1.59 | 15000 | 1.3583 | 0.7072 |
| 1.3829 | 2.12 | 20000 | 1.3590 | 0.7121 |
| 1.3069 | 2.65 | 25000 | 1.3506 | 0.7154 |
| 1.2921 | 3.18 | 30000 | 1.3448 | 0.7160 |
| 1.2731 | 3.7 | 35000 | 1.3375 | 0.7178 |
### Framework versions
- Transformers 4.17.0.dev0
- Pytorch 1.10.0+cu111
- Datasets 1.18.3
- Tokenizers 0.11.0
|