Downloads · 30 days
11
3% of all-time downloads
biogpt
j1username/biogpt
biogpt is a token classification model from j1username. Use it when you need labels on individual words, such as names. It is set up for transformers.
should probably proofread and complete it, then remove this comment. --
All-time downloads
321
Public
Repo size
5.6 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin1.4 GB · 100%
At a glance
- Task
- Token Classification
- Library
- transformers
- Model type
- gpt2
- Access
- Public
- Created
- Apr 15, 2023
- Updated
- Apr 16, 2023
- SHA
- 1c88c783
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. -->
biogpt
This model was trained from scratch on the ncbi_disease dataset. It achieves the following results on the evaluation set:
- Loss: 0.1599
- Precision: 0.5170
- Recall: 0.6014
- F1: 0.5560
- Accuracy: 0.9556
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: 0.0001
- 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 | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 340 | 0.1765 | 0.3914 | 0.5946 | 0.4720 | 0.9425 |
| 0.2426 | 2.0 | 680 | 0.1538 | 0.4769 | 0.6091 | 0.5350 | 0.9514 |
| 0.0881 | 3.0 | 1020 | 0.1599 | 0.5170 | 0.6014 | 0.5560 | 0.9556 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3
- Task
- Token Classification
- Library
- transformers
- Type
- gpt2
- Created
- Apr 15, 2023
- Updated
- Apr 16, 2023