Downloads · 30 days
22
2% of all-time downloads
bart-base-cnn-r2-18.7-d23-hybrid
echarlaix/bart-base-cnn-r2-18.7-d23-hybrid
bart-base-cnn-r2-18.7-d23-hybrid is a summarization model from echarlaix. Use it when you need a shorter version of a longer text. It is set up for transformers. The card lists the license as apache-2.0.
This model was created using the nnpruning python library: the linear layers contains 23% of the original weights.
All-time downloads
1.2K
Public
Repo size
2.5 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin558 MB · 100%
At a glance
- Task
- Summarization
- Library
- transformers
- License
- apache-2.0
- Model type
- bart
- Access
- Public
- Created
- Mar 2, 2022
- Updated
- Aug 20, 2021
- SHA
- 89e3e2f9
Try a prompt
Datasets
Model card
From the Hugging Face model README
facebook/bart-base model fine-tuned on CNN/DailyMail
This model was created using the nn_pruning python library: the linear layers contains 23% of the original weights.
The model contains 45% of the original weights overall (the embeddings account for a significant part of the model, and they are not pruned by this method).
<div class="graph"><script src="/echarlaix/bart-base-cnn-r2-18.7-d23-hybrid/raw/main/model_card/density_info.js" id="4348cd46-05bd-4e27-b565-6693f9e0b03e"></script></div>Fine-Pruning details
This model was fine-tuned from the HuggingFace model. A side-effect of block pruning is that some of the attention heads are completely removed: 61 heads were removed on a total of 216 (28.2%).
Details of the CNN/DailyMail dataset
| Dataset | Split | # samples |
|---|---|---|
| CNN/DailyMail | train | 287K |
| CNN/DailyMail | eval | 13K |
Results
| Metric | # Value |
|---|---|
| Rouge 1 | 41.43 |
| Rouge 2 | 18.72 |
| Rouge L | 38.35 |
- Task
- Summarization
- Library
- transformers
- Type
- bart
- License
- apache-2.0
- Languages
- en
- Created
- Mar 2, 2022
- Updated
- Aug 20, 2021