Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
datasets:
|
| 4 |
+
- vishnun/CodevsNL
|
| 5 |
+
language:
|
| 6 |
+
- en
|
| 7 |
+
metrics:
|
| 8 |
+
- accuracy
|
| 9 |
+
library_name: transformers
|
| 10 |
+
pipeline_tag: text-classification
|
| 11 |
+
tags:
|
| 12 |
+
- code
|
| 13 |
+
- nli
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
## PreFace
|
| 17 |
+
Code vs Natural language classification using bert-small from prajwall, below are the metrics achieved
|
| 18 |
+
|
| 19 |
+
## Training Metrics
|
| 20 |
+
|
| 21 |
+
| Epoch | Training Loss | Validation Loss | Accuracy |
|
| 22 |
+
|-------|---------------|-----------------|----------|
|
| 23 |
+
| 1 | 0.022500 | 0.012705 | 0.997203 |
|
| 24 |
+
| 2 | 0.008700 | 0.013107 | 0.996880 |
|
| 25 |
+
| 3 | 0.002700 | 0.014081 | 0.997633 |
|
| 26 |
+
| 4 | 0.001800 | 0.010666 | 0.997526 |
|
| 27 |
+
| 5 | 0.000900 | 0.010800 | 0.998063 |
|
| 28 |
+
|
| 29 |
+
## More
|
| 30 |
+
|
| 31 |
+
- Github repo for installable python package: https://github.com/Vishnunkumar
|
| 32 |
+
- Space on the extraction of code blocks from screenshots: https://huggingface.co/spaces/vishnun/SnapCode
|