Add BWSK model card

README.md

@@ -167,9 +167,12 @@ BWSK is a framework that classifies every neural network operation as **S-type**
 Load a specific variant:
 
 ```python
-import torch
+from transformers import AutoModelForImageClassification, AutoFeatureExtractor
+
 # Load fine-tuned conventional variant
-# Weights are in the finetune-conventional/ subdirectory
+model = AutoModelForImageClassification.from_pretrained(
+    "tzervas/bwsk-vit-base", subfolder="finetune-conventional"
+)
 ```
 
 ## Training Configuration
@@ -177,13 +180,13 @@ import torch
 | Setting | Value |
 |---------|-------|
 | **Optimizer** | AdamW |
-| **LR (fine-tune)** | 2e-5 |
-| **LR (from-scratch)** | 5e-4 |
+| **LR (fine-tune)** | 5e-05 |
+| **LR (from-scratch)** | 3e-04 |
 | **LR Schedule** | Cosine with warmup |
 | **Max Grad Norm** | 1.0 |
 | **Mixed Precision** | AMP (float16) |
 | **Early Stopping** | Patience 3 |
-| **Batch Size** | 8 |
+| **Batch Size** | 16 |
 
 ## Links