Update README.md
Browse files
README.md
CHANGED
|
@@ -12,7 +12,6 @@ tags:
|
|
| 12 |
|
| 13 |
|
| 14 |
|
| 15 |
-
|
| 16 |
## Model Description
|
| 17 |
|
| 18 |
- **Developed by:** [`Identrics`](https://identrics.ai/)
|
|
@@ -53,12 +52,16 @@ print(output.logits)
|
|
| 53 |
|
| 54 |
## Training Details
|
| 55 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 56 |
|
| 57 |
|
| 58 |
-
|
| 59 |
|
| 60 |
-
Achieved an f1 score of x%
|
| 61 |
-
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
| 62 |
|
| 63 |
|
| 64 |
|
|
|
|
| 12 |
|
| 13 |
|
| 14 |
|
|
|
|
| 15 |
## Model Description
|
| 16 |
|
| 17 |
- **Developed by:** [`Identrics`](https://identrics.ai/)
|
|
|
|
| 52 |
|
| 53 |
## Training Details
|
| 54 |
|
| 55 |
+
The training datasets for the model consist of a balanced set totaling 840 examples that include both propaganda and non-propaganda content. These examples are collected from a variety of traditional media and social media sources, ensuring a diverse range of content. Aditionally, the training dataset is enriched with AI-generated samples. The total distribution of the training data is shown in the table below:
|
| 56 |
+
|
| 57 |
+
|
| 58 |
+
|
| 59 |
+
|
| 60 |
+
The dataset is divided in proportions 80/10/10 for training, testing and evaluation respectively. After fine-tuning on the training dataset,
|
| 61 |
|
| 62 |
|
| 63 |
+
We achieved an f1 score of 0.807 on evaluation dataset.
|
| 64 |
|
|
|
|
|
|
|
| 65 |
|
| 66 |
|
| 67 |
|