Add evaluation metrics to README (acc 0.81 / F1 0.79)

README.md

@@ -1,27 +1,44 @@
 ---
-language:
-- en
+language: [en]
 license: apache-2.0
 library_name: transformers
-pipeline_tag: text-classification         # ✅ valid task ID
-datasets:
-- financial_phrase_bank
+pipeline_tag: text-classification
+datasets: [financial_phrase_bank]
 base_model: ProsusAI/finbert
 tags:
-- sentiment-analysis                      # free-form tag
+- sentiment-analysis
 - finance
-- finbert
 - text-classification
+model-index:
+- name: Financial Sentiment BERT (FinBERT fine-tune)
+  results:
+  - task:
+      type: text-classification
+      name: Sentiment Analysis
+    dataset:
+      name: Financial PhraseBank
+      type: financial_phrase_bank
+      split: test
+    metrics:
+    - type: accuracy
+      value: 0.81
+    - type: f1
+      name: macro F1
+      value: 0.79
 ---
 # Financial Sentiment BERT (FinBERT fine-tune)
 
 Sentence-level classifier for **English financial news**.
 
-* **Base model**: ProsusAI/finbert  
-* **Dataset**   : Financial PhraseBank  
-* **Labels**    : positive (0) · negative (1) · neutral (2)  
-* **Epochs**    : 6 (AdamW, lr = 2e-5)  
-* **Hardware**  : CPU-only training
+| Item | Value |
+|------|-------|
+| **Base model** | `ProsusAI/finbert` |
+| **Dataset** | Financial PhraseBank |
+| **Labels** | positive (0) · negative (1) · neutral (2) |
+| **Epochs** | 3 (early-stopped) |
+| **Best val acc** | 0.8356 |
+| **Test acc** | 0.81 |
+| **Hardware** | CPU-only training |
 
 ## Usage