update model card README.md
Browse files
README.md
CHANGED
@@ -13,13 +13,13 @@ should probably proofread and complete it, then remove this comment. -->
|
|
13 |
|
14 |
This model is a fine-tuned version of [HueyNemud/das22-10-camembert_pretrained](https://huggingface.co/HueyNemud/das22-10-camembert_pretrained) on the None dataset.
|
15 |
It achieves the following results on the evaluation set:
|
16 |
-
- Loss: 0.
|
17 |
-
- Ebegin: {'precision': 0.
|
18 |
-
- Eend: {'precision': 0.
|
19 |
-
- Overall Precision: 0.
|
20 |
-
- Overall Recall: 0.
|
21 |
-
- Overall F1: 0.
|
22 |
-
- Overall Accuracy: 0.
|
23 |
|
24 |
## Model description
|
25 |
|
@@ -50,26 +50,14 @@ The following hyperparameters were used during training:
|
|
50 |
|
51 |
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
52 |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
53 |
-
| No log | 0.07 | 300 | 0.
|
54 |
-
| 0.
|
55 |
-
| 0.
|
56 |
-
| 0.
|
57 |
-
| 0.
|
58 |
-
| 0.
|
59 |
-
| 0.
|
60 |
-
| 0.
|
61 |
-
| 0.0069 | 0.64 | 2700 | 0.0049 | 0.9956 | 0.9829 | 0.9892 | 0.9982 |
|
62 |
-
| 0.0058 | 0.72 | 3000 | 0.0056 | 0.9810 | 0.9968 | 0.9888 | 0.9981 |
|
63 |
-
| 0.0058 | 0.79 | 3300 | 0.0043 | 0.9884 | 0.9957 | 0.9921 | 0.9987 |
|
64 |
-
| 0.0057 | 0.86 | 3600 | 0.0048 | 0.9845 | 0.9965 | 0.9905 | 0.9984 |
|
65 |
-
| 0.0057 | 0.93 | 3900 | 0.0043 | 0.9874 | 0.9964 | 0.9919 | 0.9986 |
|
66 |
-
| 0.0046 | 1.0 | 4200 | 0.0052 | 0.9844 | 0.9974 | 0.9909 | 0.9985 |
|
67 |
-
| 0.0037 | 1.07 | 4500 | 0.0036 | 0.9906 | 0.9935 | 0.9921 | 0.9987 |
|
68 |
-
| 0.0037 | 1.14 | 4800 | 0.0054 | 0.9893 | 0.9952 | 0.9922 | 0.9987 |
|
69 |
-
| 0.0036 | 1.22 | 5100 | 0.0040 | 0.9881 | 0.9956 | 0.9918 | 0.9986 |
|
70 |
-
| 0.0036 | 1.29 | 5400 | 0.0039 | 0.9885 | 0.9961 | 0.9923 | 0.9987 |
|
71 |
-
| 0.0036 | 1.36 | 5700 | 0.0037 | 0.9885 | 0.9961 | 0.9923 | 0.9987 |
|
72 |
-
| 0.003 | 1.43 | 6000 | 0.0037 | 0.9884 | 0.9962 | 0.9922 | 0.9987 |
|
73 |
|
74 |
|
75 |
### Framework versions
|
|
|
13 |
|
14 |
This model is a fine-tuned version of [HueyNemud/das22-10-camembert_pretrained](https://huggingface.co/HueyNemud/das22-10-camembert_pretrained) on the None dataset.
|
15 |
It achieves the following results on the evaluation set:
|
16 |
+
- Loss: 0.0084
|
17 |
+
- Ebegin: {'precision': 0.9916415914409896, 'recall': 0.9880079946702198, 'f1': 0.9898214583680961, 'number': 3002}
|
18 |
+
- Eend: {'precision': 0.9879919946631087, 'recall': 0.9873333333333333, 'f1': 0.9876625541847281, 'number': 3000}
|
19 |
+
- Overall Precision: 0.9898
|
20 |
+
- Overall Recall: 0.9877
|
21 |
+
- Overall F1: 0.9887
|
22 |
+
- Overall Accuracy: 0.9982
|
23 |
|
24 |
## Model description
|
25 |
|
|
|
50 |
|
51 |
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
52 |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
53 |
+
| No log | 0.07 | 300 | 0.0364 | 0.9697 | 0.9694 | 0.9695 | 0.9949 |
|
54 |
+
| 0.1737 | 0.14 | 600 | 0.0146 | 0.9849 | 0.9880 | 0.9865 | 0.9977 |
|
55 |
+
| 0.1737 | 0.21 | 900 | 0.0092 | 0.9835 | 0.9929 | 0.9881 | 0.9980 |
|
56 |
+
| 0.0158 | 0.29 | 1200 | 0.0074 | 0.9904 | 0.9912 | 0.9908 | 0.9984 |
|
57 |
+
| 0.0098 | 0.36 | 1500 | 0.0058 | 0.9866 | 0.9943 | 0.9904 | 0.9984 |
|
58 |
+
| 0.0098 | 0.43 | 1800 | 0.0075 | 0.9883 | 0.9898 | 0.9890 | 0.9982 |
|
59 |
+
| 0.0073 | 0.5 | 2100 | 0.0068 | 0.9962 | 0.9815 | 0.9888 | 0.9981 |
|
60 |
+
| 0.0073 | 0.57 | 2400 | 0.0065 | 0.9899 | 0.9900 | 0.9899 | 0.9983 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
61 |
|
62 |
|
63 |
### Framework versions
|