Add/refresh model card and checkpoint

README.md

@@ -0,0 +1,23 @@
+# Financial Sentiment BERT (FinBERT fine-tune)
+
+**Repo ID:** `Kroalist/financial-sentiment-bert`  
+**Base model:** [`ProsusAI/finbert`](https://huggingface.co/ProsusAI/finbert)  
+**Dataset:** Financial PhraseBank (Malo et al.)  
+**Classes:** 0 = positive · 1 = negative · 2 = neutral  
+**Fine-tuning epochs:** 6  
+**Optimizer:** AdamW (lr = 2e-5)  
+**Hardware:** CPU-only training
+
+## Intended use
+
+Sentence-level sentiment tagging of English financial news, reports, or earnings calls to feed dashboards or downstream trading models.
+
+## Quick load
+
+```python
+from transformers import AutoTokenizer, AutoModelForSequenceClassification
+tok   = AutoTokenizer.from_pretrained("Kroalist/financial-sentiment-bert")
+model = AutoModelForSequenceClassification.from_pretrained("Kroalist/financial-sentiment-bert")
+```
+
+_Last updated: 2025-04-23_

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/Users/arthursavary/Desktop/Code_AI/final_project/best_model",
+  "_name_or_path": "ProsusAI/finbert",
   "architectures": [
     "BertForSequenceClassification"
   ],
@@ -10,16 +10,16 @@
   "hidden_dropout_prob": 0.1,
   "hidden_size": 768,
   "id2label": {
-    "0": "negative",
-    "1": "neutral",
-    "2": "positive"
+    "0": "positive",
+    "1": "negative",
+    "2": "neutral"
   },
   "initializer_range": 0.02,
   "intermediate_size": 3072,
   "label2id": {
-    "negative": 0,
-    "neutral": 1,
-    "positive": 2
+    "negative": 1,
+    "neutral": 2,
+    "positive": 0
   },
   "layer_norm_eps": 1e-12,
   "max_position_embeddings": 512,

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:162dcd26e997f6a83a09b4d4b7867164c174e33bcdf165d357ce46c6c2d2208a
+oid sha256:0fac89cc799e1376543aebda46dadf6b874cd592d9bf3e7cc084a14ec7c70d78
 size 438008181