[lora_conversion] Enhance key handling for OneTrainer components in LORA conversion utility (#11441)

Author: iamwavecut

src/diffusers/loaders/lora_conversion_utils.py

@@ -727,8 +727,25 @@ def _convert(original_key, diffusers_key, state_dict, new_state_dict):
             elif k.startswith("lora_te1_"):
                 has_te_keys = True
                 continue
+            elif k.startswith("lora_transformer_context_embedder"):
+                diffusers_key = "context_embedder"
+            elif k.startswith("lora_transformer_norm_out_linear"):
+                diffusers_key = "norm_out.linear"
+            elif k.startswith("lora_transformer_proj_out"):
+                diffusers_key = "proj_out"
+            elif k.startswith("lora_transformer_x_embedder"):
+                diffusers_key = "x_embedder"
+            elif k.startswith("lora_transformer_time_text_embed_guidance_embedder_linear_"):
+                i = int(k.split("lora_transformer_time_text_embed_guidance_embedder_linear_")[-1])
+                diffusers_key = f"time_text_embed.guidance_embedder.linear_{i}"
+            elif k.startswith("lora_transformer_time_text_embed_text_embedder_linear_"):
+                i = int(k.split("lora_transformer_time_text_embed_text_embedder_linear_")[-1])
+                diffusers_key = f"time_text_embed.text_embedder.linear_{i}"
+            elif k.startswith("lora_transformer_time_text_embed_timestep_embedder_linear_"):
+                i = int(k.split("lora_transformer_time_text_embed_timestep_embedder_linear_")[-1])
+                diffusers_key = f"time_text_embed.timestep_embedder.linear_{i}"
             else:
-                raise NotImplementedError
+                raise NotImplementedError(k)
 
             if "attn_" in k:
                 if "_to_out_0" in k: