Training complete
Browse files
README.md
CHANGED
@@ -11,7 +11,7 @@ metrics:
|
|
11 |
- f1
|
12 |
- accuracy
|
13 |
model-index:
|
14 |
-
- name: bert-finetuned-ner
|
15 |
results:
|
16 |
- task:
|
17 |
name: Token Classification
|
@@ -25,30 +25,30 @@ model-index:
|
|
25 |
metrics:
|
26 |
- name: Precision
|
27 |
type: precision
|
28 |
-
value: 0.
|
29 |
- name: Recall
|
30 |
type: recall
|
31 |
-
value: 0.
|
32 |
- name: F1
|
33 |
type: f1
|
34 |
-
value: 0.
|
35 |
- name: Accuracy
|
36 |
type: accuracy
|
37 |
-
value: 0.
|
38 |
---
|
39 |
|
40 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
41 |
should probably proofread and complete it, then remove this comment. -->
|
42 |
|
43 |
-
# bert-finetuned-ner
|
44 |
|
45 |
This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-cased) on the conll2003 dataset.
|
46 |
It achieves the following results on the evaluation set:
|
47 |
-
- Loss: 0.
|
48 |
-
- Precision: 0.
|
49 |
-
- Recall: 0.
|
50 |
-
- F1: 0.
|
51 |
-
- Accuracy: 0.
|
52 |
|
53 |
## Model description
|
54 |
|
@@ -79,9 +79,9 @@ The following hyperparameters were used during training:
|
|
79 |
|
80 |
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
81 |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
82 |
-
| 0.
|
83 |
-
| 0.
|
84 |
-
| 0.
|
85 |
|
86 |
|
87 |
### Framework versions
|
|
|
11 |
- f1
|
12 |
- accuracy
|
13 |
model-index:
|
14 |
+
- name: bert-finetuned-ner-accelerate
|
15 |
results:
|
16 |
- task:
|
17 |
name: Token Classification
|
|
|
25 |
metrics:
|
26 |
- name: Precision
|
27 |
type: precision
|
28 |
+
value: 0.9354304635761589
|
29 |
- name: Recall
|
30 |
type: recall
|
31 |
+
value: 0.9508582968697409
|
32 |
- name: F1
|
33 |
type: f1
|
34 |
+
value: 0.9430812885995661
|
35 |
- name: Accuracy
|
36 |
type: accuracy
|
37 |
+
value: 0.9866809913463237
|
38 |
---
|
39 |
|
40 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
41 |
should probably proofread and complete it, then remove this comment. -->
|
42 |
|
43 |
+
# bert-finetuned-ner-accelerate
|
44 |
|
45 |
This model is a fine-tuned version of [bert-base-cased](https://huggingface.co/bert-base-cased) on the conll2003 dataset.
|
46 |
It achieves the following results on the evaluation set:
|
47 |
+
- Loss: 0.0797
|
48 |
+
- Precision: 0.9354
|
49 |
+
- Recall: 0.9509
|
50 |
+
- F1: 0.9431
|
51 |
+
- Accuracy: 0.9867
|
52 |
|
53 |
## Model description
|
54 |
|
|
|
79 |
|
80 |
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
81 |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
82 |
+
| 0.0211 | 1.0 | 1756 | 0.0741 | 0.9254 | 0.9443 | 0.9348 | 0.9851 |
|
83 |
+
| 0.0126 | 2.0 | 3512 | 0.0741 | 0.9331 | 0.9485 | 0.9407 | 0.9862 |
|
84 |
+
| 0.0084 | 3.0 | 5268 | 0.0797 | 0.9354 | 0.9509 | 0.9431 | 0.9867 |
|
85 |
|
86 |
|
87 |
### Framework versions
|