Retrained with adaptive-classifier v0.0.17 (k parameter fix)

README.md

@@ -17,15 +17,15 @@ You can install it with `pip install adaptive-classifier`.
 
 - Base Model: answerdotai/ModernBERT-base
 - Number of Classes: 3
-- Total Examples: 252
+- Total Examples: 15
 - Embedding Dimension: 768
 
 ## Class Distribution
 
 ```
-appropriate: 86 examples (34.1%)
-inappropriate: 82 examples (32.5%)
-spam: 84 examples (33.3%)
+appropriate: 5 examples (33.3%)
+inappropriate: 5 examples (33.3%)
+spam: 5 examples (33.3%)
 ```
 
 ## Usage

config.json

@@ -42,5 +42,10 @@
     "spam": 1
   },
   "model_name": "answerdotai/ModernBERT-base",
-  "train_steps": 8
+  "train_steps": 8,
+  "training_history": {
+    "appropriate": 100,
+    "inappropriate": 100,
+    "spam": 100
+  }
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:135f1483ace80b2fe4f8f339df5e40104e934611e1043d814fe118283691aab7
+oid sha256:fa9e4b268b424eadcdcda081c20f35eb3af5a69e6ca04ffa4c8ecb61100d13bb
 size 3558212