Update README.md

README.md

@@ -47,8 +47,9 @@ against traditional/manual augmentation techniques for CNN classification.
 
 #### Training Hyperparameters
 
-==========
-               PIXEL DDPM (Baseline Model) ===========================================================
+===================
+               PIXEL DDPM (Baseline Model) ==================
+               
 • Optimization Strategy:
   - Optimizer:      AdamW
   - Learning Rate:  2e-4 (with Cosine Annealing)
@@ -63,9 +64,10 @@ against traditional/manual augmentation techniques for CNN classification.
   - EMA:            Disabled
   - Diffusion:      1000 Timesteps (Linear Schedule)
 
-==========
-               LATENT DDPM (LDM) ===========================================================
-• Optimization Strategy:
+===================
+               LATENT DDPM (LDM) ============================
+               
+• Optimization Strategy:               
   - Optimizer:      AdamW
   - Learning Rate:  1e-4 (with Cosine Annealing)
   - Weight Decay:   1e-3
@@ -78,9 +80,6 @@ against traditional/manual augmentation techniques for CNN classification.
   - Channel Mults:  (2, 2)
   - EMA:            Enabled (Decay: 0.99)
 
-• Conditioning (CFG):
-  - Label Dropout:  15% (during training)
-  - Guidance Scale: 5.0 (during inference)
 ===========================================================