MightyDragon-Dev commited on
Commit
efcd089
·
verified ·
1 Parent(s): d881e67

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +23 -3
README.md CHANGED
@@ -43,8 +43,28 @@ The "Titan DNA" protocol allows for mass-generation of 1,000+ seeds to map the l
43
  ### 3. Thermal-Aware Inference
44
  Optimized for the Surface Pro's 35W power envelope. Uses raw Torch inference with **KV-Caching** to maintain stable frame rates even under thermal pressure.
45
 
46
- ## 🎨 Visualizing the Output
47
- Recommended visualization via `matplotlib` with `magma` colormap for "Radar-style" clarity:
48
 
49
  ```python
50
- plt.imshow(tensor, cmap='magma', interpolation='lanczos')
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
43
  ### 3. Thermal-Aware Inference
44
  Optimized for the Surface Pro's 35W power envelope. Uses raw Torch inference with **KV-Caching** to maintain stable frame rates even under thermal pressure.
45
 
46
+ ## 🖼️ Sample Inference & Visualization
47
+ Use the following code to generate a high-fidelity "Radar Scan" from a specific seed. This snippet is optimized for 4K displays and technical clarity.
48
 
49
  ```python
50
+ import torch
51
+ from transformers import AutoModelForCausalLM
52
+ import matplotlib.pyplot as plt
53
+
54
+ # Load Dragon Interceptor
55
+ model = AutoModelForCausalLM.from_pretrained("MightyDragon-Dev/dragon_interceptor")
56
+
57
+ # Generate a 28x28 Blueprint (Seed 42)
58
+ seed_id = 42
59
+ input_ids = torch.tensor([[seed_id]])
60
+ output = model.generate(input_ids, max_length=784, do_sample=True, temperature=0.7)
61
+
62
+ # Reshape and Render
63
+ blueprint = output[0].view(28, 28).detach().numpy()
64
+
65
+ plt.figure(figsize=(8, 8), dpi=120)
66
+ plt.imshow(blueprint, cmap='magma', interpolation='lanczos')
67
+ plt.title(f"Dragon Interceptor: Sector Scan (Seed {seed_id})", color='white')
68
+ plt.style.use('dark_background')
69
+ plt.axis('off')
70
+ plt.show()