[node] estimation

README.md

@@ -89,6 +89,8 @@ python app.py
 | Qwen2-VL-7B | 1024/256 | 1 | Inference | FP16 | 1 |
 | VILA-1.5-13B | 2048/512 | 2 | Inference | BF16 | 1 |
 | Qwen2-Audio-7B | 1024/256 | 1 | Inference | FP16 | 1 |
+| PhysicsNeMo-FNO-Large | 512/128 | 8 | Training | FP32 | 1 |
+| PhysicsNeMo-GraphCast-Medium | 1024/256 | 4 | Training | FP16 | 1 |
 
 ## CUDA Recommendations
 
@@ -133,6 +135,12 @@ The application provides tailored CUDA version recommendations:
 - **Qwen-Audio**: Base, Chat variants
 - **Qwen2-Audio**: 7B
 
+#### Physics-ML Models (NVIDIA PhysicsNeMo)
+- **Fourier Neural Operators (FNO)**: Small (1M), Medium (10M), Large (50M)
+- **Physics-Informed Neural Networks (PINN)**: Small (0.5M), Medium (5M), Large (20M)
+- **GraphCast**: Small (50M), Medium (200M), Large (1B) - for weather/climate modeling
+- **Spherical FNO (SFNO)**: Small (25M), Medium (100M), Large (500M) - for global simulations
+
 ### Precision Impact
 - **FP32**: Full precision (4 bytes per parameter)
 - **FP16/BF16**: Half precision (2 bytes per parameter)
@@ -145,6 +153,17 @@ The application provides tailored CUDA version recommendations:
 - **Memory Overhead**: Additional memory for vision/audio encoders and cross-modal attention
 - **Token Estimation**: Consider multimodal inputs when calculating token counts
 
+### PhysicsNeMo Considerations
+- **Grid-Based Data**: Physics models work with spatial/temporal grids rather than text tokens
+- **Batch Training**: Physics-ML models typically require larger batch sizes for stable training
+- **Memory Patterns**: Different from LLMs - less KV cache, more gradient memory for PDE constraints
+- **Precision Requirements**: Many physics simulations require FP32 for numerical stability
+- **Use Cases**: 
+  - **FNO**: Solving PDEs on regular grids (fluid dynamics, heat transfer)
+  - **PINN**: Physics-informed training with PDE constraints
+  - **GraphCast**: Weather prediction and climate modeling
+  - **SFNO**: Global atmospheric and oceanic simulations
+
 ## Limitations
 
 - Estimates are approximate and may vary based on:

app.py

@@ -272,10 +272,10 @@ def estimate_nodes_interface(
     
     # Validate inputs
     if input_tokens <= 0 or output_tokens <= 0:
-        return "Please enter valid token counts (> 0)", "", None, ""
+        return "Please enter valid token counts (> 0)", "", None, "## ⚠️ <span style='color: #E74C3C;'>**Invalid Input: Token counts must be > 0**</span>"
     
     if batch_size <= 0:
-        return "Please enter a valid batch size (> 0)", "", None, ""
+        return "Please enter a valid batch size (> 0)", "", None, "## ⚠️ <span style='color: #E74C3C;'>**Invalid Input: Batch size must be > 0**</span>"
     
     # Calculate node requirements
     nodes_needed, explanation, breakdown = estimate_h100_nodes(
@@ -288,7 +288,7 @@ def estimate_nodes_interface(
     # Create performance chart
     fig = create_performance_chart(breakdown)
     
-    return explanation, cuda_rec, fig, f"**Estimated H100 Nodes Required: {nodes_needed}**"
+    return explanation, cuda_rec, fig, f"## 🖥️ <span style='color: #4A90E2;'>**Estimated H100 Nodes Required: {nodes_needed}**</span>"
 
 # Create Gradio interface
 def create_interface():
@@ -345,7 +345,7 @@ def create_interface():
             with gr.Column(scale=2):
                 gr.Markdown("## Results")
                 
-                node_count = gr.Markdown("**Ready to estimate...**")
+                node_count = gr.Markdown("## 🖥️ <span style='color: #4A90E2;'>**Ready to estimate...**</span>")
                 
                 with gr.Tab("📊 Detailed Analysis"):
                     detailed_output = gr.Markdown()