fix

Author: Perseus14

src/maxdiffusion/generate_wan.py

@@ -157,7 +157,7 @@ def run(config, pipeline=None, filename_prefix=""):
     lora_config = config.lora_config
 
     if len(lora_config["lora_model_name_or_path"]) > 1:
-        max_logging.warning("Found multiple LoRAs in config, but only loading the first one for WAN 2.2.")
+        max_logging.log("Found multiple LoRAs in config, but only loading the first one for WAN 2.2.")
 
     pipeline = lora_loader.load_lora_weights(
         pipeline,

src/maxdiffusion/loaders/wan_lora_nnx_loader.py

@@ -51,7 +51,7 @@ def load_lora_weights(
         )
         lora_nnx.merge_lora(pipeline.high_noise_transformer, h_state_dict, scale)
     else:
-        max_logging.warning("high_noise_transformer not found or no weight name provided for LoRA.")
+        max_logging.log("high_noise_transformer not found or no weight name provided for LoRA.")
 
     # Handle low noise model
     if hasattr(pipeline, "low_noise_transformer") and low_noise_weight_name:
@@ -61,6 +61,6 @@ def load_lora_weights(
         )
         lora_nnx.merge_lora(pipeline.low_noise_transformer, l_state_dict, scale)
     else:
-        max_logging.warning("low_noise_transformer not found or no weight name provided for LoRA.")
+        max_logging.log("low_noise_transformer not found or no weight name provided for LoRA.")
 
     return pipeline

src/maxdiffusion/models/lora_nnx.py

@@ -217,7 +217,7 @@ def merge_lora(model: nnx.Module, state_dict: dict, scale: float):
           lora_params[module_path_str] = {}
         lora_params[module_path_str][weight_type] = jnp.array(v)
       else:
-        max_logging.warning(f"Could not parse LoRA key: {k}")
+        max_logging.log(f"Could not parse LoRA key: {k}")
 
     assigned_count = 0
     for path, module in nnx.iter_graph(model):
@@ -257,5 +257,5 @@ def merge_lora(model: nnx.Module, state_dict: dict, scale: float):
                           f"with kernel_size {module.kernel_size} > 1 is not supported."
                       )
           else:
-              max_logging.warning(f"LoRA weights for {matched_key} incomplete.")
+              max_logging.log(f"LoRA weights for {matched_key} incomplete.")
     max_logging.log(f"Merged weights into {assigned_count} layers in {type(model).__name__}.")