Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,20 @@
|
|
| 1 |
---
|
| 2 |
license: cc-by-sa-4.0
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: cc-by-sa-4.0
|
| 3 |
+
language:
|
| 4 |
+
- id
|
| 5 |
+
library_name: transformers
|
| 6 |
+
pipeline_tag: text-classification
|
| 7 |
---
|
| 8 |
+
# BERATBOS v0.0.1-prealpha
|
| 9 |
+
> BERT-based Automated Text Classification Based on POS
|
| 10 |
+
|
| 11 |
+
A model I made for my final year assignment, used a random script to train this on GPU,
|
| 12 |
+
as a result, the usage is kinda non-standard, as far as transformers-based models go, at least.
|
| 13 |
+
|
| 14 |
+
Model is for detecting AI-generated works via coherency comparison through POS tag differences,
|
| 15 |
+
it was only trained for 5 epochs on a dataset of 1000 human sentences and 1000 BLOOM-generated sentences.
|
| 16 |
+
|
| 17 |
+

|
| 18 |
+
|
| 19 |
+
I have plans to further this model into an actual production model,
|
| 20 |
+
but will only be done once I can properly assess on whether that's a worthy enough of an endeavour.
|