up font style
Browse files
README.md
CHANGED
|
@@ -61,21 +61,20 @@ Sentiment detection in Polish data (fine-tuned on tweets from political domain).
|
|
| 61 |
|
| 62 |
## Training and evaluation data
|
| 63 |
|
| 64 |
-
|
| 65 |
-
|
| 66 |
-
#### Training results: loss: 0.1358926964368792
|
| 67 |
|
| 68 |
|
| 69 |
|
| 70 |
It achieves the following results on the test set (10%):
|
| 71 |
|
| 72 |
-
|
| 73 |
|
| 74 |
-
|
| 75 |
|
| 76 |
-
|
| 77 |
|
| 78 |
-
|
| 79 |
|
| 80 |
precision recall f1-score support
|
| 81 |
|
|
|
|
| 61 |
|
| 62 |
## Training and evaluation data
|
| 63 |
|
| 64 |
+
- Trained for 3 epochs, mini-batch size of 8.
|
| 65 |
+
- Training results: loss: 0.1358926964368792
|
|
|
|
| 66 |
|
| 67 |
|
| 68 |
|
| 69 |
It achieves the following results on the test set (10%):
|
| 70 |
|
| 71 |
+
- No. examples = 100
|
| 72 |
|
| 73 |
+
- Batch size = 8
|
| 74 |
|
| 75 |
+
- accuracy = 0.950
|
| 76 |
|
| 77 |
+
- macro f1 = 0.944
|
| 78 |
|
| 79 |
precision recall f1-score support
|
| 80 |
|