Update README.md

README.md

@@ -1,3 +1,51 @@
+---
+tags:
+- indonesian
+- sentiment-analysis
+- finance
+- financial-sentiment
+- indo-roberta
+- transformers
+- fine-tuned
+- huggingface
+- ihsg
+- stock-market
+- nlp
+- indonesian-roberta-base-financial-sentiment-classifier
+language:
+- id
+datasets:
+- intanm/indonesian-financial-sentiment-analysis
+metrics:
+- accuracy
+- f1
+- precision
+- recall
+model-index:
+- name: finetuned-indo-roBERTa-financial-sentiment
+  results:
+  - task:
+      type: text-classification
+      name: Text Classification (Sentiment Analysis)
+    dataset:
+      name: indonesian-financial-sentiment
+      type: intanm/indonesian-financial-sentiment-analysis
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.9749
+    - name: F1
+      type: f1
+      value: 0.9749
+    - name: Precision
+      type: precision
+      value: 0.9749
+    - name: Recall
+      type: recall
+      value: 0.9749
+license: mit
+---
+
 You're absolutely right — label flipping is **critical** here since the base model you fine-tuned uses a **non-standard label mapping** (`0 = Positive`, `2 = Negative`). Here's the **updated and corrected** model card with:
 
 * 🔁 Clear warning about label flipping
@@ -164,4 +212,4 @@ Created by
 
 For collaborations or questions, feel free to reach out via Hugging Face or GitHub.
 
----
+---