nicolauduran45
commited on
Update README.md
Browse files
README.md
CHANGED
@@ -150,9 +150,41 @@ result = citation_classifier(citation_text)
|
|
150 |
print(result)
|
151 |
```
|
152 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
153 |
|
154 |
## Additional information
|
155 |
|
156 |
### Authors
|
157 |
|
158 |
- SIRIS Lab, Research Division of SIRIS Academic.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
150 |
print(result)
|
151 |
```
|
152 |
|
153 |
+
## Training
|
154 |
+
|
155 |
+
The model was trained using the **Citation Pre-Screening Dataset** consisting of:
|
156 |
+
|
157 |
+
- **Training data**: 3599 samples
|
158 |
+
- **Test data**: 400 samples
|
159 |
+
|
160 |
+
The following hyperparameters were used for training:
|
161 |
+
|
162 |
+
- **Model Path**: `distilbert/distilbert-base-multilingual-cased`
|
163 |
+
- **Batch Size**: 32
|
164 |
+
- **Number of Epochs**: 4
|
165 |
+
- **Learning Rate**: 2e-5
|
166 |
+
- **Max Sequence Length**: 512
|
167 |
+
|
168 |
+
## Evaluation Metrics
|
169 |
+
|
170 |
+
The model's performance was evaluated on the test set, and the following results were obtained:
|
171 |
+
|
172 |
+
| Metric | Value |
|
173 |
+
|----------------------|--------|
|
174 |
+
| **Accuracy** | 0.95 |
|
175 |
+
| **Macro avg F1** | 0.94 |
|
176 |
+
| **Weighted avg F1** | 0.95 |
|
177 |
|
178 |
## Additional information
|
179 |
|
180 |
### Authors
|
181 |
|
182 |
- SIRIS Lab, Research Division of SIRIS Academic.
|
183 |
+
|
184 |
+
### License
|
185 |
+
|
186 |
+
This work is distributed under a [Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0).
|
187 |
+
|
188 |
+
### Contact
|
189 |
+
For further information, send an email to either [[email protected]](mailto:[email protected]) or [[email protected]](mailto:[email protected]).
|
190 |
+
|