Update README.md
Browse files
README.md
CHANGED
@@ -164,6 +164,7 @@ if __name__ == "__main__":
|
|
164 |
## Training data
|
165 |
|
166 |
In order to train the model, it's transformers were trained with five datasets, which were:
|
|
|
167 |
- **Scientific Papers (arXiv + PubMed)**: Cohan et al. (2018) found out that there were only
|
168 |
datasets with short texts (with an average of 600 words) or datasets with longer texts with
|
169 |
extractive humam summaries. In order to fill the gap and to provide a dataset with long text
|
@@ -198,13 +199,15 @@ kind of ATS that is aimed at answering the question “What is the document abou
|
|
198 |
was obtained from BBC articles and each one of them is accompanied by a short gold-standard
|
199 |
summary often written by its very author.
|
200 |
|
|
|
|
|
201 |
## Training procedure
|
202 |
|
203 |
|
204 |
### Preprocessing
|
205 |
[PERGUNTAR ARTHUR]
|
206 |
|
207 |
-
Hey, look how easy it is to write LaTeX equations in here \\(Ax = b\\) or even $$Ax = b$$
|
208 |
## Evaluation results
|
209 |
|
210 |
|
|
|
164 |
## Training data
|
165 |
|
166 |
In order to train the model, it's transformers were trained with five datasets, which were:
|
167 |
+
|
168 |
- **Scientific Papers (arXiv + PubMed)**: Cohan et al. (2018) found out that there were only
|
169 |
datasets with short texts (with an average of 600 words) or datasets with longer texts with
|
170 |
extractive humam summaries. In order to fill the gap and to provide a dataset with long text
|
|
|
199 |
was obtained from BBC articles and each one of them is accompanied by a short gold-standard
|
200 |
summary often written by its very author.
|
201 |
|
202 |
+
Each of their documents was summarized (through every summarization method applied in the code) and evaluated
|
203 |
+
in comparison with the gold-standard summaries.
|
204 |
## Training procedure
|
205 |
|
206 |
|
207 |
### Preprocessing
|
208 |
[PERGUNTAR ARTHUR]
|
209 |
|
210 |
+
Hey, look how easy it is to write LaTeX equations in here \\(Ax = b\\) or even $$ Ax = b $$
|
211 |
## Evaluation results
|
212 |
|
213 |
|