hack to fix lora loading for qwen image

Author: LostRuins

otherarch/sdcpp/name_conversion.cpp

@@ -1037,6 +1037,16 @@ std::string convert_tensor_name(std::string name, SDVersion version) {
         if (sd_version_is_unet(version) || sd_version_is_flux(version) || is_lycoris_underline) {
             name = convert_sep_to_dot(name);
         }
+
+        //kcpp hack: name surgery for qwen image, https://github.com/leejet/stable-diffusion.cpp/issues/1131
+        const std::string badprefix = "unet_transformer_blocks_";
+        const std::string goodprefix = "lora.model.diffusion_model.transformer_blocks.";
+        if(is_lora && sd_version_is_qwen_image(version) && starts_with(name,badprefix))
+        {
+            name = goodprefix + name.substr(badprefix.size());
+            name = convert_sep_to_dot(name);
+        }
+        return name;
     }
 
     std::unordered_map<std::string, std::string> prefix_map = {