Update README.md
Browse files
README.md
CHANGED
|
@@ -11,7 +11,7 @@ pipeline_tag: token-classification
|
|
| 11 |
|
| 12 |
The model is intended to be used on segments of **250** characters. Anything else will compromise performance.
|
| 13 |
|
| 14 |
-
In initial testing, using **0.25** as a probability threshold
|
| 15 |
|
| 16 |
|
| 17 |
|
|
|
|
| 11 |
|
| 12 |
The model is intended to be used on segments of **250** characters. Anything else will compromise performance.
|
| 13 |
|
| 14 |
+
In initial testing, using **0.25** as a probability threshold typically produced the best F1 score across all degrees of corruption.
|
| 15 |
|
| 16 |
|
| 17 |
|