Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,26 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: mit
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
language:
|
| 4 |
+
- en
|
| 5 |
+
metrics:
|
| 6 |
+
- accuracy
|
| 7 |
+
|
| 8 |
+
pipeline_tag: text-classification
|
| 9 |
+
library_name: transformers
|
| 10 |
+
tags:
|
| 11 |
+
- anti-spam
|
| 12 |
+
- spam
|
| 13 |
+
- spamclass
|
| 14 |
+
- english
|
| 15 |
+
- English
|
| 16 |
+
- EN
|
| 17 |
+
- en
|
| 18 |
+
---
|
| 19 |
+
## SpamClass 100k
|
| 20 |
+
- acc=0.8180159256801592
|
| 21 |
+
- prec=0.7736263736263737
|
| 22 |
+
- rec=0.7514944491887275
|
| 23 |
+
- f1=0.7623998267273121
|
| 24 |
+
# Trained on a custom corpus.
|
| 25 |
+
# 1x T4 was used during training.
|
| 26 |
+
It primarily is meant for english spam, its accurate 81% of the time (acceptable)
|