Update README.md
Browse files
README.md
CHANGED
@@ -22,47 +22,33 @@ model-index:
|
|
22 |
metrics:
|
23 |
- name: Validation ROGUE-1
|
24 |
type: rouge-1
|
25 |
-
value:
|
26 |
verified: true
|
27 |
- name: Validation ROGUE-2
|
28 |
type: rouge-2
|
29 |
-
value: 10.
|
30 |
verified: true
|
31 |
- name: Validation ROGUE-L
|
32 |
type: rouge-l
|
33 |
-
value:
|
|
|
|
|
|
|
|
|
34 |
verified: true
|
35 |
---
|
36 |
|
37 |
# bart-base-cnn-swe
|
|
|
38 |
|
39 |
-
|
40 |
-
- 1. Further fine-tune on checkpoint with cnn-daily-swe.
|
41 |
-
- 2. Further fine-tune on xsum-swe.
|
42 |
-
- 3. Lastly fine-tune on smaller domain dataset.
|
43 |
|
44 |
This model is a fine-tuned version of [KBLab/bart-base-swedish-cased](https://huggingface.co/KBLab/bart-base-swedish-cased) on the [Gabriel/bart-base-cnn-swe](https://huggingface.co/datasets/Gabriel/cnn_daily_swe) dataset.
|
45 |
|
46 |
|
47 |
-
|
48 |
-
# bart-base-cnn-swe
|
49 |
-
|
50 |
-
This model is a fine-tuned version of [Gabriel/bart-base-cnn-swe](https://huggingface.co/Gabriel/bart-base-cnn-swe) on the None dataset.
|
51 |
-
It achieves the following results on the evaluation set:
|
52 |
-
- Loss: 1.9759
|
53 |
-
- Rouge1: 22.2046
|
54 |
-
- Rouge2: 10.4332
|
55 |
-
- Rougel: 18.1753
|
56 |
-
- Rougelsum: 20.846
|
57 |
-
- Gen Len: 19.9971
|
58 |
-
|
59 |
-
## Model description
|
60 |
-
|
61 |
-
More information needed
|
62 |
-
|
63 |
## Intended uses & limitations
|
64 |
|
65 |
-
|
66 |
|
67 |
## Training and evaluation data
|
68 |
|
@@ -88,8 +74,10 @@ The following hyperparameters were used during training:
|
|
88 |
|
89 |
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|
90 |
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
|
91 |
-
|
|
92 |
-
|
|
|
|
|
|
93 |
|
94 |
|
95 |
### Framework versions
|
|
|
22 |
metrics:
|
23 |
- name: Validation ROGUE-1
|
24 |
type: rouge-1
|
25 |
+
value: 22.2046
|
26 |
verified: true
|
27 |
- name: Validation ROGUE-2
|
28 |
type: rouge-2
|
29 |
+
value: 10.4332
|
30 |
verified: true
|
31 |
- name: Validation ROGUE-L
|
32 |
type: rouge-l
|
33 |
+
value: 18.1753
|
34 |
+
verified: true
|
35 |
+
- name: Validation ROGUE-L-SUM
|
36 |
+
type: rouge-l-sum
|
37 |
+
value: 20.846
|
38 |
verified: true
|
39 |
---
|
40 |
|
41 |
# bart-base-cnn-swe
|
42 |
+
This model is a W.I.P
|
43 |
|
44 |
+
## Model description
|
|
|
|
|
|
|
45 |
|
46 |
This model is a fine-tuned version of [KBLab/bart-base-swedish-cased](https://huggingface.co/KBLab/bart-base-swedish-cased) on the [Gabriel/bart-base-cnn-swe](https://huggingface.co/datasets/Gabriel/cnn_daily_swe) dataset.
|
47 |
|
48 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
49 |
## Intended uses & limitations
|
50 |
|
51 |
+
This model should only be used to fine-tune further on.
|
52 |
|
53 |
## Training and evaluation data
|
54 |
|
|
|
74 |
|
75 |
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|
76 |
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
|
77 |
+
| 2.2349 | 1.0 | 17944 | 2.0643 | 21.9564 | 10.2133 | 17.9958 | 20.6502 | 19.9992 |
|
78 |
+
| 2.0726 | 2.0 | 35888 | 2.0253 | 22.0568 | 10.3302 | 18.0648 | 20.7482 | 19.9996 |
|
79 |
+
| 1.8658 | 3.0 | 53832 | 2.0333 | 22.0871 | 10.2902 | 18.0577 | 20.7082 | 19.998 |
|
80 |
+
| 1.8121 | 4.0 | 71776 | 1.9759 | 22.2046 | 10.4332 | 18.1753 | 20.846 | 19.9971 |
|
81 |
|
82 |
|
83 |
### Framework versions
|