Update README.md
Browse files
README.md
CHANGED
@@ -195,7 +195,14 @@ library_name: transformers
|
|
195 |
---
|
196 |
|
197 |
|
198 |
-
# long-t5-tglobal-base-sci-simplify
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
199 |
|
200 |
This model is a fine-tuned version of [google/long-t5-tglobal-base](https://huggingface.co/google/long-t5-tglobal-base) on the `pszemraj/scientific_lay_summarisation-elife-norm` dataset.
|
201 |
|
@@ -207,17 +214,33 @@ It achieves the following results on the evaluation set:
|
|
207 |
- Rougelsum: 35.9333
|
208 |
- Gen Len: 392.7095
|
209 |
|
210 |
-
## Model description
|
211 |
-
|
212 |
-
More information needed
|
213 |
|
214 |
## Intended uses & limitations
|
215 |
|
216 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
217 |
|
218 |
## Training and evaluation data
|
219 |
|
220 |
-
The `elife` subset of the :lay summaries dataset. Refer to `pszemraj/scientific_lay_summarisation-elife-norm
|
221 |
|
222 |
## Training procedure
|
223 |
|
|
|
195 |
---
|
196 |
|
197 |
|
198 |
+
# long-t5-tglobal-base-sci-simplify: elife subset
|
199 |
+
|
200 |
+
|
201 |
+
Exploring how well long-document models trained on "lay summaries" of scientific papers generalize.
|
202 |
+
|
203 |
+
> A lay summary is a summary of a research paper or scientific study that is written in plain language, without the use of technical jargon, and is designed to be easily understood by non-experts.
|
204 |
+
|
205 |
+
## Model description
|
206 |
|
207 |
This model is a fine-tuned version of [google/long-t5-tglobal-base](https://huggingface.co/google/long-t5-tglobal-base) on the `pszemraj/scientific_lay_summarisation-elife-norm` dataset.
|
208 |
|
|
|
214 |
- Rougelsum: 35.9333
|
215 |
- Gen Len: 392.7095
|
216 |
|
|
|
|
|
|
|
217 |
|
218 |
## Intended uses & limitations
|
219 |
|
220 |
+
- Ability to generalize outside of the dataset domain (pubmed/bioscience type papers) has to be evaluated.
|
221 |
+
|
222 |
+
|
223 |
+
## Usage
|
224 |
+
|
225 |
+
It's recommended to usage this model with [beam search decoding](https://huggingface.co/docs/transformers/generation_strategies#beamsearch-decoding). If interested, you can also use the `textsum` util repo to have most of this abstracted out for you:
|
226 |
+
|
227 |
+
|
228 |
+
```bash
|
229 |
+
pip install -U textsum
|
230 |
+
```
|
231 |
+
|
232 |
+
```python
|
233 |
+
from textsum.summarize import Summarizer
|
234 |
+
|
235 |
+
summarizer = Summarizer('pszemraj/long-t5-tglobal-base-sci-simplify')
|
236 |
+
text = "put the text you don't want to read here"
|
237 |
+
summary = summarizer.summarize_string(text)
|
238 |
+
print(summary)
|
239 |
+
```
|
240 |
|
241 |
## Training and evaluation data
|
242 |
|
243 |
+
The `elife` subset of the :lay summaries dataset. Refer to `pszemraj/scientific_lay_summarisation-elife-norm`
|
244 |
|
245 |
## Training procedure
|
246 |
|