Update README.md
Browse files
README.md
CHANGED
@@ -12,7 +12,7 @@ widget:
|
|
12 |
example_title: 'Text'
|
13 |
|
14 |
model-index:
|
15 |
-
- name: IMISLab/
|
16 |
results:
|
17 |
- task:
|
18 |
type: summarization
|
@@ -41,10 +41,10 @@ model-index:
|
|
41 |
verified: true
|
42 |
---
|
43 |
|
44 |
-
#
|
45 |
|
46 |
A Greek encyclopedic article summarization model trained and evaluated on [GreekWikipedia](https://huggingface.co/datasets/IMISLab/GreekWikipedia/).
|
47 |
-
This model was trained as part of our research paper:
|
48 |
[Giarelis, N., Mastrokostas, C., & Karacapilidis, N. (2024) Greek Wikipedia: A Study on Abstractive Summarization]()
|
49 |
For more information see the evaluation section below.
|
50 |
|
@@ -52,7 +52,7 @@ For more information see the evaluation section below.
|
|
52 |
|
53 |
## Training dataset
|
54 |
|
55 |
-
The training dataset of `
|
56 |
This dataset contains 93,433 articles collected from the Greek part of [Wikipedia](https://el.wikipedia.org/).
|
57 |
|
58 |
## Training configuration
|
@@ -83,7 +83,7 @@ LEAD|18.51|3.18|11.48|65.77
|
|
83 |
```python
|
84 |
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer, pipeline
|
85 |
|
86 |
-
model_name = 'IMISLab/
|
87 |
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
|
88 |
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
89 |
|
|
|
12 |
example_title: 'Text'
|
13 |
|
14 |
model-index:
|
15 |
+
- name: IMISLab/GreekWiki-umt5-base
|
16 |
results:
|
17 |
- task:
|
18 |
type: summarization
|
|
|
41 |
verified: true
|
42 |
---
|
43 |
|
44 |
+
# GreekWiki (umt5-base)
|
45 |
|
46 |
A Greek encyclopedic article summarization model trained and evaluated on [GreekWikipedia](https://huggingface.co/datasets/IMISLab/GreekWikipedia/).
|
47 |
+
This model was trained as part of our upcoming research paper:
|
48 |
[Giarelis, N., Mastrokostas, C., & Karacapilidis, N. (2024) Greek Wikipedia: A Study on Abstractive Summarization]()
|
49 |
For more information see the evaluation section below.
|
50 |
|
|
|
52 |
|
53 |
## Training dataset
|
54 |
|
55 |
+
The training dataset of `GreekWiki (umt5-base)` is [GreekWikipedia](https://huggingface.co/datasets/IMISLab/GreekWikipedia/), which is the first encyclopedic summarization dataset for the Greek Language.
|
56 |
This dataset contains 93,433 articles collected from the Greek part of [Wikipedia](https://el.wikipedia.org/).
|
57 |
|
58 |
## Training configuration
|
|
|
83 |
```python
|
84 |
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer, pipeline
|
85 |
|
86 |
+
model_name = 'IMISLab/GreekWiki-umt5-base'
|
87 |
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
|
88 |
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
89 |
|