modified readme and reqs

README.md

@@ -12,7 +12,7 @@ pinned: false
 
 ## Model Description
 
-This is a random baseline model for the Frugal AI Challenge 2024, specifically for the text classification task of identifying climate disinformation. The model serves as a performance floor, randomly assigning labels to text inputs without any learning.
+ModernBert model finetuned on train dataset.
 
 ### Intended Use
 
@@ -24,7 +24,7 @@ This is a random baseline model for the Frugal AI Challenge 2024, specifically f
 
 The model uses the QuotaClimat/frugalaichallenge-text-train dataset:
 - Size: ~6000 examples
-- Split: 80% train, 20% test
+- Split: 90% train, 10% test
 - 8 categories of climate disinformation claims
 
 ### Labels

requirements.txt

@@ -7,4 +7,5 @@ pydantic>=1.10.0
 python-dotenv>=1.0.0
 gradio>=4.0.0
 requests>=2.31.0
-librosa==0.10.2.post1
+librosa==0.10.2.post1
+transformers>=4.47.1