Commit
•
7322d1e
1
Parent(s):
8ee99bf
End of training
Browse files
README.md
CHANGED
@@ -5,9 +5,28 @@ tags:
|
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
7 |
- glue
|
|
|
|
|
|
|
8 |
model-index:
|
9 |
- name: bert-finetuned-mrpc
|
10 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
---
|
12 |
|
13 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
@@ -16,6 +35,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
16 |
# bert-finetuned-mrpc
|
17 |
|
18 |
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the glue dataset.
|
|
|
|
|
|
|
|
|
19 |
|
20 |
## Model description
|
21 |
|
@@ -44,6 +67,11 @@ The following hyperparameters were used during training:
|
|
44 |
|
45 |
### Training results
|
46 |
|
|
|
|
|
|
|
|
|
|
|
47 |
|
48 |
|
49 |
### Framework versions
|
|
|
5 |
- generated_from_trainer
|
6 |
datasets:
|
7 |
- glue
|
8 |
+
metrics:
|
9 |
+
- accuracy
|
10 |
+
- f1
|
11 |
model-index:
|
12 |
- name: bert-finetuned-mrpc
|
13 |
+
results:
|
14 |
+
- task:
|
15 |
+
name: Text Classification
|
16 |
+
type: text-classification
|
17 |
+
dataset:
|
18 |
+
name: glue
|
19 |
+
type: glue
|
20 |
+
config: mrpc
|
21 |
+
split: validation
|
22 |
+
args: mrpc
|
23 |
+
metrics:
|
24 |
+
- name: Accuracy
|
25 |
+
type: accuracy
|
26 |
+
value: 0.8480392156862745
|
27 |
+
- name: F1
|
28 |
+
type: f1
|
29 |
+
value: 0.8916083916083916
|
30 |
---
|
31 |
|
32 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
35 |
# bert-finetuned-mrpc
|
36 |
|
37 |
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on the glue dataset.
|
38 |
+
It achieves the following results on the evaluation set:
|
39 |
+
- Loss: 0.5841
|
40 |
+
- Accuracy: 0.8480
|
41 |
+
- F1: 0.8916
|
42 |
|
43 |
## Model description
|
44 |
|
|
|
67 |
|
68 |
### Training results
|
69 |
|
70 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|
71 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|
|
72 |
+
| No log | 1.0 | 459 | 0.4703 | 0.7794 | 0.8308 |
|
73 |
+
| 0.5511 | 2.0 | 918 | 0.6611 | 0.7451 | 0.8424 |
|
74 |
+
| 0.3958 | 3.0 | 1377 | 0.5841 | 0.8480 | 0.8916 |
|
75 |
|
76 |
|
77 |
### Framework versions
|