sofia-todeschini
commited on
Commit
•
3f06a5c
1
Parent(s):
080878f
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
tags:
|
3 |
+
- generated_from_trainer
|
4 |
+
metrics:
|
5 |
+
- accuracy
|
6 |
+
model-index:
|
7 |
+
- name: BioELECTRA-LitCovid-v1.3h
|
8 |
+
results: []
|
9 |
+
---
|
10 |
+
|
11 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
12 |
+
should probably proofread and complete it, then remove this comment. -->
|
13 |
+
|
14 |
+
# BioELECTRA-LitCovid-v1.3h
|
15 |
+
|
16 |
+
This model is a fine-tuned version of [kamalkraj/bioelectra-base-discriminator-pubmed](https://huggingface.co/kamalkraj/bioelectra-base-discriminator-pubmed) on an unknown dataset.
|
17 |
+
It achieves the following results on the evaluation set:
|
18 |
+
- Loss: 0.7785
|
19 |
+
- Hamming loss: 0.0198
|
20 |
+
- F1 micro: 0.8361
|
21 |
+
- F1 macro: 0.3559
|
22 |
+
- F1 weighted: 0.8787
|
23 |
+
- F1 samples: 0.8727
|
24 |
+
- Precision micro: 0.7527
|
25 |
+
- Precision macro: 0.2860
|
26 |
+
- Precision weighted: 0.8332
|
27 |
+
- Precision samples: 0.8513
|
28 |
+
- Recall micro: 0.9403
|
29 |
+
- Recall macro: 0.7383
|
30 |
+
- Recall weighted: 0.9403
|
31 |
+
- Recall samples: 0.9483
|
32 |
+
- Roc Auc: 0.9614
|
33 |
+
- Accuracy: 0.6748
|
34 |
+
|
35 |
+
## Model description
|
36 |
+
|
37 |
+
More information needed
|
38 |
+
|
39 |
+
## Intended uses & limitations
|
40 |
+
|
41 |
+
More information needed
|
42 |
+
|
43 |
+
## Training and evaluation data
|
44 |
+
|
45 |
+
More information needed
|
46 |
+
|
47 |
+
## Training procedure
|
48 |
+
|
49 |
+
### Training hyperparameters
|
50 |
+
|
51 |
+
The following hyperparameters were used during training:
|
52 |
+
- learning_rate: 5e-05
|
53 |
+
- train_batch_size: 16
|
54 |
+
- eval_batch_size: 16
|
55 |
+
- seed: 42
|
56 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
57 |
+
- lr_scheduler_type: linear
|
58 |
+
- lr_scheduler_warmup_ratio: 0.1866747178469669
|
59 |
+
- num_epochs: 5
|
60 |
+
- mixed_precision_training: Native AMP
|
61 |
+
|
62 |
+
### Training results
|
63 |
+
|
64 |
+
| Training Loss | Epoch | Step | Validation Loss | Hamming loss | F1 micro | F1 macro | F1 weighted | F1 samples | Precision micro | Precision macro | Precision weighted | Precision samples | Recall micro | Recall macro | Recall weighted | Recall samples | Roc Auc | Accuracy |
|
65 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------------:|:--------:|:--------:|:-----------:|:----------:|:---------------:|:---------------:|:------------------:|:-----------------:|:------------:|:------------:|:---------------:|:--------------:|:-------:|:--------:|
|
66 |
+
| 1.5951 | 1.0 | 2272 | 0.6305 | 0.0577 | 0.6142 | 0.2231 | 0.7544 | 0.7325 | 0.4787 | 0.1794 | 0.7016 | 0.6957 | 0.8568 | 0.7122 | 0.8568 | 0.8824 | 0.9020 | 0.3878 |
|
67 |
+
| 1.1968 | 2.0 | 4544 | 0.4865 | 0.0353 | 0.7393 | 0.2825 | 0.8470 | 0.8156 | 0.6122 | 0.2305 | 0.7890 | 0.7790 | 0.9330 | 0.7538 | 0.9330 | 0.9449 | 0.9497 | 0.5560 |
|
68 |
+
| 0.9573 | 3.0 | 6816 | 0.5637 | 0.0247 | 0.8033 | 0.3292 | 0.8474 | 0.8430 | 0.7019 | 0.2574 | 0.7851 | 0.8066 | 0.9389 | 0.7380 | 0.9389 | 0.9470 | 0.9582 | 0.5918 |
|
69 |
+
| 0.7604 | 4.0 | 9088 | 0.6811 | 0.0206 | 0.8306 | 0.3558 | 0.8726 | 0.8675 | 0.7441 | 0.2835 | 0.8239 | 0.8438 | 0.9400 | 0.7574 | 0.9400 | 0.9483 | 0.9608 | 0.6561 |
|
70 |
+
| 0.4404 | 5.0 | 11360 | 0.7785 | 0.0198 | 0.8361 | 0.3559 | 0.8787 | 0.8727 | 0.7527 | 0.2860 | 0.8332 | 0.8513 | 0.9403 | 0.7383 | 0.9403 | 0.9483 | 0.9614 | 0.6748 |
|
71 |
+
|
72 |
+
|
73 |
+
### Framework versions
|
74 |
+
|
75 |
+
- Transformers 4.28.0
|
76 |
+
- Pytorch 2.1.0+cu121
|
77 |
+
- Datasets 2.18.0
|
78 |
+
- Tokenizers 0.13.3
|