Update README.md

README.md

@@ -1,9 +1,25 @@
 ## Dutch Fine-Tuned BERT For Active/Passive Voice Classification. 
-#### Bedrijvende en Lijdende vorm classificatie voor zinnen
+### Bedrijvende en Lijdende vorm classificatie voor zinnen
 
+#### Examples
+Try the following examples in the Hosted inference API:
+1. Jan werd geslagen door zijn moeder.
+2. Wie niet weg is, is gezien
+3. Ik ben van plan om morgen te gaan werken
+4. De makelaar heeft het nieuwe huis verkocht aan de bewoners die iets verderop wonen.
+
+Label_0 = Active / Bedrijvend. Label_1 = Passive / Lijdend
+
+Answers (what they should be): 
+1. Label_1
+2. Label_1
+3. Label_0
+4. Label_0
+
+#### Basic Information
 This model is fine-tuned on BERTje (https://github.com/wietsedv/bertje) for recognizing passive and active voice in Dutch sentences. 
 
-Fine-tuning was done for 10 epochs. The calculated scores for train-validation split are listed here:
+The calculated scores for train-validation split are listed here:
 - Accuracy: 0.9791666666666666
 - F1: 0.9787234042553191
 - Loss: 0.06746523827314377
@@ -12,7 +28,7 @@ Fine-tuning was done for 10 epochs. The calculated scores for train-validation s
 
 This is the first release, have fun with it!
 
-An example on how to use it will come soon.
+An example on how to use it will come soon. Contact me at gerwindekruijf@gmail.com for further questions.
 
 
 Gerwin