update model card README.md
Browse files
README.md
CHANGED
@@ -23,16 +23,16 @@ model-index:
|
|
23 |
metrics:
|
24 |
- name: Precision
|
25 |
type: precision
|
26 |
-
value: 0.
|
27 |
- name: Recall
|
28 |
type: recall
|
29 |
-
value: 0.
|
30 |
- name: F1
|
31 |
type: f1
|
32 |
-
value: 0.
|
33 |
- name: Accuracy
|
34 |
type: accuracy
|
35 |
-
value: 0.
|
36 |
---
|
37 |
|
38 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -40,13 +40,13 @@ should probably proofread and complete it, then remove this comment. -->
|
|
40 |
|
41 |
# biobert-finetuned-ncbi
|
42 |
|
43 |
-
This model is a fine-tuned version of [dmis-lab/biobert-
|
44 |
It achieves the following results on the evaluation set:
|
45 |
-
- Loss: 0.
|
46 |
-
- Precision: 0.
|
47 |
-
- Recall: 0.
|
48 |
-
- F1: 0.
|
49 |
-
- Accuracy: 0.
|
50 |
|
51 |
## Model description
|
52 |
|
@@ -70,23 +70,21 @@ The following hyperparameters were used during training:
|
|
70 |
- eval_batch_size: 8
|
71 |
- seed: 42
|
72 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
73 |
-
- lr_scheduler_type:
|
74 |
-
- num_epochs:
|
75 |
|
76 |
### Training results
|
77 |
|
78 |
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
79 |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
80 |
-
| 0.
|
81 |
-
| 0.
|
82 |
-
| 0.
|
83 |
-
| 0.0051 | 4.0 | 2720 | 0.0800 | 0.8317 | 0.8602 | 0.8457 | 0.9848 |
|
84 |
-
| 0.0041 | 5.0 | 3400 | 0.0816 | 0.8309 | 0.8615 | 0.8459 | 0.9846 |
|
85 |
|
86 |
|
87 |
### Framework versions
|
88 |
|
89 |
- Transformers 4.25.1
|
90 |
-
- Pytorch 1.13.
|
91 |
- Datasets 2.8.0
|
92 |
- Tokenizers 0.13.2
|
|
|
23 |
metrics:
|
24 |
- name: Precision
|
25 |
type: precision
|
26 |
+
value: 0.8192771084337349
|
27 |
- name: Recall
|
28 |
type: recall
|
29 |
+
value: 0.8640406607369758
|
30 |
- name: F1
|
31 |
type: f1
|
32 |
+
value: 0.8410636982065552
|
33 |
- name: Accuracy
|
34 |
type: accuracy
|
35 |
+
value: 0.9856218100336114
|
36 |
---
|
37 |
|
38 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
40 |
|
41 |
# biobert-finetuned-ncbi
|
42 |
|
43 |
+
This model is a fine-tuned version of [dmis-lab/biobert-v1.1](https://huggingface.co/dmis-lab/biobert-v1.1) on the ncbi_disease dataset.
|
44 |
It achieves the following results on the evaluation set:
|
45 |
+
- Loss: 0.0590
|
46 |
+
- Precision: 0.8193
|
47 |
+
- Recall: 0.8640
|
48 |
+
- F1: 0.8411
|
49 |
+
- Accuracy: 0.9856
|
50 |
|
51 |
## Model description
|
52 |
|
|
|
70 |
- eval_batch_size: 8
|
71 |
- seed: 42
|
72 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
73 |
+
- lr_scheduler_type: linear
|
74 |
+
- num_epochs: 3
|
75 |
|
76 |
### Training results
|
77 |
|
78 |
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
79 |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
80 |
+
| 0.1049 | 1.0 | 680 | 0.0588 | 0.7826 | 0.7776 | 0.7801 | 0.9806 |
|
81 |
+
| 0.0362 | 2.0 | 1360 | 0.0539 | 0.8084 | 0.8577 | 0.8323 | 0.9852 |
|
82 |
+
| 0.0109 | 3.0 | 2040 | 0.0590 | 0.8193 | 0.8640 | 0.8411 | 0.9856 |
|
|
|
|
|
83 |
|
84 |
|
85 |
### Framework versions
|
86 |
|
87 |
- Transformers 4.25.1
|
88 |
+
- Pytorch 1.13.1+cu116
|
89 |
- Datasets 2.8.0
|
90 |
- Tokenizers 0.13.2
|