update model card README.md
Browse files
README.md
CHANGED
@@ -1,11 +1,24 @@
|
|
1 |
---
|
2 |
tags:
|
3 |
- generated_from_trainer
|
|
|
|
|
4 |
metrics:
|
5 |
- pearsonr
|
6 |
model-index:
|
7 |
- name: roberta-base-finetuned-sts
|
8 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
---
|
10 |
|
11 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -13,10 +26,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
13 |
|
14 |
# roberta-base-finetuned-sts
|
15 |
|
16 |
-
This model is a fine-tuned version of [klue/roberta-base](https://huggingface.co/klue/roberta-base) on the
|
17 |
It achieves the following results on the evaluation set:
|
18 |
-
- Loss: 0.
|
19 |
-
- Pearsonr: 0.
|
20 |
|
21 |
## Model description
|
22 |
|
@@ -35,12 +48,13 @@ More information needed
|
|
35 |
### Training hyperparameters
|
36 |
|
37 |
The following hyperparameters were used during training:
|
38 |
-
- learning_rate:
|
39 |
-
- train_batch_size:
|
40 |
-
- eval_batch_size:
|
41 |
- seed: 42
|
42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
43 |
- lr_scheduler_type: linear
|
|
|
44 |
- num_epochs: 15
|
45 |
- mixed_precision_training: Native AMP
|
46 |
|
@@ -48,21 +62,21 @@ The following hyperparameters were used during training:
|
|
48 |
|
49 |
| Training Loss | Epoch | Step | Validation Loss | Pearsonr |
|
50 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
51 |
-
| No log | 1.0 |
|
52 |
-
|
|
53 |
-
|
|
54 |
-
|
|
55 |
-
|
|
56 |
-
|
|
57 |
-
| 0.
|
58 |
-
| 0.
|
59 |
-
| 0.
|
60 |
-
| 0.
|
61 |
-
| 0.
|
62 |
-
| 0.
|
63 |
-
| 0.
|
64 |
-
| 0.
|
65 |
-
| 0.
|
66 |
|
67 |
|
68 |
### Framework versions
|
1 |
---
|
2 |
tags:
|
3 |
- generated_from_trainer
|
4 |
+
datasets:
|
5 |
+
- klue
|
6 |
metrics:
|
7 |
- pearsonr
|
8 |
model-index:
|
9 |
- name: roberta-base-finetuned-sts
|
10 |
+
results:
|
11 |
+
- task:
|
12 |
+
name: Text Classification
|
13 |
+
type: text-classification
|
14 |
+
dataset:
|
15 |
+
name: klue
|
16 |
+
type: klue
|
17 |
+
args: sts
|
18 |
+
metrics:
|
19 |
+
- name: Pearsonr
|
20 |
+
type: pearsonr
|
21 |
+
value: 0.956039443806831
|
22 |
---
|
23 |
|
24 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
26 |
|
27 |
# roberta-base-finetuned-sts
|
28 |
|
29 |
+
This model is a fine-tuned version of [klue/roberta-base](https://huggingface.co/klue/roberta-base) on the klue dataset.
|
30 |
It achieves the following results on the evaluation set:
|
31 |
+
- Loss: 0.1999
|
32 |
+
- Pearsonr: 0.9560
|
33 |
|
34 |
## Model description
|
35 |
|
48 |
### Training hyperparameters
|
49 |
|
50 |
The following hyperparameters were used during training:
|
51 |
+
- learning_rate: 1e-05
|
52 |
+
- train_batch_size: 32
|
53 |
+
- eval_batch_size: 8
|
54 |
- seed: 42
|
55 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
56 |
- lr_scheduler_type: linear
|
57 |
+
- lr_scheduler_warmup_steps: 200
|
58 |
- num_epochs: 15
|
59 |
- mixed_precision_training: Native AMP
|
60 |
|
62 |
|
63 |
| Training Loss | Epoch | Step | Validation Loss | Pearsonr |
|
64 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
65 |
+
| No log | 1.0 | 329 | 0.2462 | 0.9478 |
|
66 |
+
| 1.2505 | 2.0 | 658 | 0.1671 | 0.9530 |
|
67 |
+
| 1.2505 | 3.0 | 987 | 0.1890 | 0.9525 |
|
68 |
+
| 0.133 | 4.0 | 1316 | 0.2360 | 0.9548 |
|
69 |
+
| 0.0886 | 5.0 | 1645 | 0.2265 | 0.9528 |
|
70 |
+
| 0.0886 | 6.0 | 1974 | 0.2097 | 0.9518 |
|
71 |
+
| 0.0687 | 7.0 | 2303 | 0.2281 | 0.9523 |
|
72 |
+
| 0.0539 | 8.0 | 2632 | 0.2212 | 0.9542 |
|
73 |
+
| 0.0539 | 9.0 | 2961 | 0.1843 | 0.9532 |
|
74 |
+
| 0.045 | 10.0 | 3290 | 0.1999 | 0.9560 |
|
75 |
+
| 0.0378 | 11.0 | 3619 | 0.2357 | 0.9533 |
|
76 |
+
| 0.0378 | 12.0 | 3948 | 0.2134 | 0.9541 |
|
77 |
+
| 0.033 | 13.0 | 4277 | 0.2273 | 0.9540 |
|
78 |
+
| 0.03 | 14.0 | 4606 | 0.2148 | 0.9533 |
|
79 |
+
| 0.03 | 15.0 | 4935 | 0.2207 | 0.9534 |
|
80 |
|
81 |
|
82 |
### Framework versions
|