Update README.md

README.md

@@ -1,20 +1,20 @@
 ---
 license: apache-2.0
 datasets: google-research-datasets/go_emotions
-base_model: intfloat/multilingual-e5-large
+base_model: FacebookAI/xlm-roberta-base
 language:
 - de
 metrics:
-- roc_auc value: 0.658
-- f1: 0.382
-- accuracy: 0.27
+- f1_macro: 0.45
+- accuracy: 0.41
+- kappa: 0.42
 pipeline_tag: text-classification
 tags:
 - medical
 model_description: >-
   This is basically the German translation of arpanghoshal/EmoRoBERTa. We used
   the go_emotions dataset, translated it into German and fine-tuned the
-  intfloat/multilingual-e5-large model. So this model allows the classification
+  FacebookAI/xlm-roberta-base model. So this model allows the classification
   of 28 emotions in German Transcripts ('admiration', 'amusement', 'anger',
   'annoyance', 'approval', 'caring', 'confusion', 'curiosity', 'desire',
   'disappointment', 'disapproval', 'disgust', 'embarrassment', 'excitement',
@@ -36,12 +36,12 @@ This is basically the German translation of arpanghoshal/EmoRoBERTa. We used the
 - **Finetuned from model:** intfloat/multilingual-e5-large
 - **Hyperparameters:**
     - Epochs: 10
-    - learning_rate: 1e-5
+    - learning_rate: 3e-5
     - weight_decay: 0.01
 - **Metrics:**
-    - accuracy: 0.27
-    - f1: 0.382
-    - roc_auc: 0.658
+    - accuracy: 0.41
+    - f1: 0.45
+    - kappa: 0.42
 ---