Update README.md
Browse files
README.md
CHANGED
|
@@ -24,11 +24,13 @@ model-index:
|
|
| 24 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 25 |
should probably proofread and complete it, then remove this comment. -->
|
| 26 |
|
| 27 |
-
#
|
| 28 |
|
| 29 |
-
|
| 30 |
|
| 31 |
-
|
|
|
|
|
|
|
| 32 |
|
| 33 |
It achieves the following results on the evaluation set:
|
| 34 |
|
|
@@ -37,17 +39,7 @@ It achieves the following results on the evaluation set:
|
|
| 37 |
- Rougel: 38.1981
|
| 38 |
- Rougelsum: 39.9453
|
| 39 |
|
| 40 |
-
|
| 41 |
-
|
| 42 |
-
More information needed
|
| 43 |
-
|
| 44 |
-
## Intended uses & limitations
|
| 45 |
-
|
| 46 |
-
More information needed
|
| 47 |
-
|
| 48 |
-
## Training and evaluation data
|
| 49 |
-
|
| 50 |
-
More information needed
|
| 51 |
|
| 52 |
## Training procedure
|
| 53 |
|
|
@@ -69,26 +61,6 @@ The following hyperparameters were used during training:
|
|
| 69 |
- mixed_precision_training: Native AMP
|
| 70 |
- label_smoothing_factor: 0.1
|
| 71 |
|
| 72 |
-
### Training results
|
| 73 |
-
|
| 74 |
-
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|
| 75 |
-
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:---------:|:-------:|
|
| 76 |
-
| 4.4304 | 1.0 | 3895 | 4.3749 | 33.2844 | 22.8262 | 29.9423 | 30.7953 | 19.7732 |
|
| 77 |
-
| 3.65 | 2.0 | 7790 | 3.7414 | 33.8392 | 23.517 | 30.4871 | 31.3309 | 19.9031 |
|
| 78 |
-
| 3.397 | 3.0 | 11685 | 3.5651 | 34.2335 | 23.9113 | 30.9237 | 31.7434 | 19.894 |
|
| 79 |
-
| 3.2202 | 4.0 | 15580 | 3.5054 | 34.2535 | 23.9595 | 30.9811 | 31.7961 | 19.9212 |
|
| 80 |
-
| 3.0827 | 5.0 | 19475 | 3.4547 | 34.5545 | 24.1991 | 31.2609 | 32.085 | 19.9195 |
|
| 81 |
-
| 2.9801 | 6.0 | 23370 | 3.4328 | 34.6721 | 24.2537 | 31.372 | 32.1777 | 19.9331 |
|
| 82 |
-
| 2.8689 | 7.0 | 27265 | 3.4377 | 34.6764 | 24.3314 | 31.4376 | 32.1981 | 19.9278 |
|
| 83 |
-
| 2.7813 | 8.0 | 31160 | 3.4407 | 34.746 | 24.345 | 31.4511 | 32.2708 | 19.9468 |
|
| 84 |
-
| 2.6848 | 9.0 | 35055 | 3.4539 | 34.7376 | 24.3224 | 31.4784 | 32.2817 | 19.9096 |
|
| 85 |
-
| 2.5974 | 10.0 | 38950 | 3.4683 | 34.9174 | 24.4716 | 31.5641 | 32.4039 | 19.9384 |
|
| 86 |
-
| 2.5228 | 11.0 | 42845 | 3.4903 | 34.9845 | 24.4972 | 31.6585 | 32.4753 | 19.93 |
|
| 87 |
-
| 2.4633 | 12.0 | 46740 | 3.5105 | 34.8496 | 24.3559 | 31.5256 | 32.3635 | 19.9275 |
|
| 88 |
-
| 2.4022 | 13.0 | 50635 | 3.5234 | 34.9109 | 24.4008 | 31.5449 | 32.4021 | 19.9374 |
|
| 89 |
-
| 2.3605 | 14.0 | 54530 | 3.5306 | 34.9545 | 24.4365 | 31.6208 | 32.4711 | 19.9366 |
|
| 90 |
-
| 2.3216 | 15.0 | 58425 | 3.5379 | 34.9079 | 24.4077 | 31.5734 | 32.4287 | 19.9365 |
|
| 91 |
-
|
| 92 |
### Framework versions
|
| 93 |
|
| 94 |
- Transformers 4.11.3
|
|
|
|
| 24 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 25 |
should probably proofread and complete it, then remove this comment. -->
|
| 26 |
|
| 27 |
+
# [Mukayese: Turkish NLP Strikes Back](https://arxiv.org/abs/2203.01215)
|
| 28 |
|
| 29 |
+
## Turkish News Summarization
|
| 30 |
|
| 31 |
+
## mukayese/bart-turkish-mlsum
|
| 32 |
+
|
| 33 |
+
_This model is uncased_, it was initialized from scratch and trained only the mlsum/tu dataset with no pre-training.
|
| 34 |
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
|
|
|
|
| 39 |
- Rougel: 38.1981
|
| 40 |
- Rougelsum: 39.9453
|
| 41 |
|
| 42 |
+
Check this paper for more details on the model and the dataset.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 43 |
|
| 44 |
## Training procedure
|
| 45 |
|
|
|
|
| 61 |
- mixed_precision_training: Native AMP
|
| 62 |
- label_smoothing_factor: 0.1
|
| 63 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 64 |
### Framework versions
|
| 65 |
|
| 66 |
- Transformers 4.11.3
|