Commit ·
d7c2d54
1
Parent(s): 2f39777
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,38 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
---
|
| 4 |
+
BioLinkBERT-LitCovid-v1.0
|
| 5 |
+
This model is a fine-tuned version of michiyasunaga/BioLinkBERT-base on an unknown dataset. It achieves the following results on the evaluation set:
|
| 6 |
+
|
| 7 |
+
Loss: 0.1098
|
| 8 |
+
F1: 0.8992
|
| 9 |
+
Roc Auc: 0.9330
|
| 10 |
+
Accuracy: 0.7945
|
| 11 |
+
Model description
|
| 12 |
+
More information needed
|
| 13 |
+
|
| 14 |
+
Intended uses & limitations
|
| 15 |
+
More information needed
|
| 16 |
+
|
| 17 |
+
Training and evaluation data
|
| 18 |
+
More information needed
|
| 19 |
+
|
| 20 |
+
Training procedure
|
| 21 |
+
Training hyperparameters
|
| 22 |
+
The following hyperparameters were used during training:
|
| 23 |
+
|
| 24 |
+
learning_rate: 2e-05
|
| 25 |
+
train_batch_size: 8
|
| 26 |
+
eval_batch_size: 8
|
| 27 |
+
seed: 42
|
| 28 |
+
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 29 |
+
lr_scheduler_type: linear
|
| 30 |
+
num_epochs: 1
|
| 31 |
+
Training results
|
| 32 |
+
Training Loss Epoch Step Validation Loss F1 Roc Auc Accuracy
|
| 33 |
+
0.1172 1.0 3120 0.1098 0.8992 0.9330 0.7945
|
| 34 |
+
|
| 35 |
+
Transformers 4.28.0
|
| 36 |
+
Pytorch 2.0.1+cu118
|
| 37 |
+
Datasets 2.12.0
|
| 38 |
+
Tokenizers 0.13.3
|