Commit
•
ca46ea8
1
Parent(s):
4250676
update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,81 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- summarization
|
| 4 |
+
- ar
|
| 5 |
+
- encoder-decoder
|
| 6 |
+
- xlm-roberta
|
| 7 |
+
- Abstractive Summarization
|
| 8 |
+
- roberta
|
| 9 |
+
- generated_from_trainer
|
| 10 |
+
datasets:
|
| 11 |
+
- xlsum
|
| 12 |
+
model-index:
|
| 13 |
+
- name: xlmroberta2xlmroberta-finetune-summarization-ar
|
| 14 |
+
results: []
|
| 15 |
+
---
|
| 16 |
+
|
| 17 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 18 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 19 |
+
|
| 20 |
+
# xlmroberta2xlmroberta-finetune-summarization-ar
|
| 21 |
+
|
| 22 |
+
This model is a fine-tuned version of [](https://huggingface.co/) on the xlsum dataset.
|
| 23 |
+
It achieves the following results on the evaluation set:
|
| 24 |
+
- Loss: 4.1298
|
| 25 |
+
- Rouge-1: 21.51
|
| 26 |
+
- Rouge-2: 8.14
|
| 27 |
+
- Rouge-l: 19.33
|
| 28 |
+
- Gen Len: 19.98
|
| 29 |
+
- Bertscore: 71.02
|
| 30 |
+
|
| 31 |
+
## Model description
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Intended uses & limitations
|
| 36 |
+
|
| 37 |
+
More information needed
|
| 38 |
+
|
| 39 |
+
## Training and evaluation data
|
| 40 |
+
|
| 41 |
+
More information needed
|
| 42 |
+
|
| 43 |
+
## Training procedure
|
| 44 |
+
|
| 45 |
+
### Training hyperparameters
|
| 46 |
+
|
| 47 |
+
The following hyperparameters were used during training:
|
| 48 |
+
- learning_rate: 5e-05
|
| 49 |
+
- train_batch_size: 4
|
| 50 |
+
- eval_batch_size: 4
|
| 51 |
+
- seed: 42
|
| 52 |
+
- gradient_accumulation_steps: 8
|
| 53 |
+
- total_train_batch_size: 32
|
| 54 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 55 |
+
- lr_scheduler_type: linear
|
| 56 |
+
- lr_scheduler_warmup_steps: 250
|
| 57 |
+
- num_epochs: 10
|
| 58 |
+
- label_smoothing_factor: 0.1
|
| 59 |
+
|
| 60 |
+
### Training results
|
| 61 |
+
|
| 62 |
+
| Training Loss | Epoch | Step | Validation Loss | Rouge-1 | Rouge-2 | Rouge-l | Gen Len | Bertscore |
|
| 63 |
+
|:-------------:|:-----:|:-----:|:---------------:|:-------:|:-------:|:-------:|:-------:|:---------:|
|
| 64 |
+
| 8.0645 | 1.0 | 1172 | 7.3567 | 8.22 | 0.66 | 7.94 | 20.0 | 58.18 |
|
| 65 |
+
| 7.2042 | 2.0 | 2344 | 6.6058 | 12.12 | 2.19 | 11.4 | 20.0 | 63.24 |
|
| 66 |
+
| 6.4168 | 3.0 | 3516 | 5.8784 | 16.46 | 4.31 | 15.15 | 20.0 | 66.3 |
|
| 67 |
+
| 5.4622 | 4.0 | 4688 | 4.7931 | 17.6 | 5.87 | 15.9 | 19.99 | 69.21 |
|
| 68 |
+
| 4.7829 | 5.0 | 5860 | 4.4418 | 19.17 | 6.74 | 17.22 | 19.98 | 70.23 |
|
| 69 |
+
| 4.4829 | 6.0 | 7032 | 4.2950 | 19.8 | 7.11 | 17.74 | 19.98 | 70.38 |
|
| 70 |
+
| 4.304 | 7.0 | 8204 | 4.2155 | 20.71 | 7.59 | 18.56 | 19.98 | 70.66 |
|
| 71 |
+
| 4.1778 | 8.0 | 9376 | 4.1632 | 21.1 | 7.94 | 18.99 | 19.98 | 70.86 |
|
| 72 |
+
| 4.0886 | 9.0 | 10548 | 4.1346 | 21.44 | 8.03 | 19.28 | 19.98 | 70.93 |
|
| 73 |
+
| 4.0294 | 10.0 | 11720 | 4.1298 | 21.51 | 8.14 | 19.33 | 19.98 | 71.02 |
|
| 74 |
+
|
| 75 |
+
|
| 76 |
+
### Framework versions
|
| 77 |
+
|
| 78 |
+
- Transformers 4.19.4
|
| 79 |
+
- Pytorch 1.11.0+cu113
|
| 80 |
+
- Datasets 2.2.2
|
| 81 |
+
- Tokenizers 0.12.1
|