Upload README.md with huggingface_hub

README.md

@@ -7,20 +7,34 @@ tags:
 - mixed-method
 - converted-by-gradio
 ---
-# FP8 Model with Mixed Precision Recovery
+# FP8 Model with Per-Tensor Precision Recovery
 - **Source**: `https://huggingface.co/spacepxl/Wan2.1-VAE-upscale2x`
 - **Original File**: `Wan2.1_VAE_upscale2x_imageonly_real_v1.safetensors`
 - **FP8 Format**: `E5M2`
 - **FP8 File**: `Wan2.1_VAE_upscale2x_imageonly_real_v1-fp8-e5m2.safetensors`
-- **Recovery File**: `None`
+- **Recovery File**: `Wan2.1_VAE_upscale2x_imageonly_real_v1-recovery.safetensors`
 
-## Recovery Configuration
+## Recovery Rules Used
 ```json
 [
   {
-    "element": "all",
-    "method": "lora",
-    "rank": 64
+    "key_pattern": "vae",
+    "dim": 4,
+    "method": "diff"
+  },
+  {
+    "key_pattern": "encoder",
+    "dim": 4,
+    "method": "diff"
+  },
+  {
+    "key_pattern": "decoder",
+    "dim": 4,
+    "method": "diff"
+  },
+  {
+    "key_pattern": "all",
+    "method": "none"
   }
 ]
 ```
@@ -34,7 +48,7 @@ import torch
 fp8_state = load_file("Wan2.1_VAE_upscale2x_imageonly_real_v1-fp8-e5m2.safetensors")
 
 # Load recovery weights if available
-recovery_state = load_file("") if "" and os.path.exists("") else {}
+recovery_state = load_file("Wan2.1_VAE_upscale2x_imageonly_real_v1-recovery.safetensors") if "Wan2.1_VAE_upscale2x_imageonly_real_v1-recovery.safetensors" and os.path.exists("Wan2.1_VAE_upscale2x_imageonly_real_v1-recovery.safetensors") else {}
 
 # Reconstruct high-precision weights
 reconstructed = {}
@@ -42,16 +56,19 @@ for key in fp8_state:
     fp8_weight = fp8_state[key].to(torch.float32)  # Convert to float32 for computation
     
     # Apply LoRA recovery if available
-    if f"lora_A.{key}" in recovery_state and f"lora_B.{key}" in recovery_state:
-        A = recovery_state[f"lora_A.{key}"].to(torch.float32)
-        B = recovery_state[f"lora_B.{key}"].to(torch.float32)
+    lora_a_key = f"lora_A.{key}"
+    lora_b_key = f"lora_B.{key}"
+    if lora_a_key in recovery_state and lora_b_key in recovery_state:
+        A = recovery_state[lora_a_key].to(torch.float32)
+        B = recovery_state[lora_b_key].to(torch.float32)
         # Reconstruct the low-rank approximation
         lora_weight = B @ A
         fp8_weight = fp8_weight + lora_weight
     
     # Apply difference recovery if available
-    if f"diff.{key}" in recovery_state:
-        diff = recovery_state[f"diff.{key}"].to(torch.float32)
+    diff_key = f"diff.{key}"
+    if diff_key in recovery_state:
+        diff = recovery_state[diff_key].to(torch.float32)
         fp8_weight = fp8_weight + diff
     
     reconstructed[key] = fp8_weight
@@ -65,6 +82,6 @@ model.load_state_dict(reconstructed)
 
 ## Statistics
 - **Total layers**: 194
-- **Layers with recovery**: 0
+- **Layers with recovery**: 60
   - LoRA recovery: 0
-  - Difference recovery: 0
+  - Difference recovery: 60