[Chore] switch to logger.warning

examples/advanced_diffusion_training/train_dreambooth_lora_sd15_advanced.py

@@ -1215,7 +1215,7 @@ def main(args):
 
             xformers_version = version.parse(xformers.__version__)
             if xformers_version == version.parse("0.0.16"):
-                logger.warn(
+                logger.warning(
                     "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, "
                     "please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                 )
@@ -1366,14 +1366,14 @@ def load_model_hook(models, input_dir):
 
     # Optimizer creation
     if not (args.optimizer.lower() == "prodigy" or args.optimizer.lower() == "adamw"):
-        logger.warn(
+        logger.warning(
             f"Unsupported choice of optimizer: {args.optimizer}.Supported optimizers include [adamW, prodigy]."
             "Defaulting to adamW"
         )
         args.optimizer = "adamw"
 
     if args.use_8bit_adam and not args.optimizer.lower() == "adamw":
-        logger.warn(
+        logger.warning(
             f"use_8bit_adam is ignored when optimizer is not set to 'AdamW'. Optimizer was "
             f"set to {args.optimizer.lower()}"
         )
@@ -1407,11 +1407,11 @@ def load_model_hook(models, input_dir):
         optimizer_class = prodigyopt.Prodigy
 
         if args.learning_rate <= 0.1:
-            logger.warn(
+            logger.warning(
                 "Learning rate is too low. When using prodigy, it's generally better to set learning rate around 1.0"
             )
         if args.train_text_encoder and args.text_encoder_lr:
-            logger.warn(
+            logger.warning(
                 f"Learning rates were provided both for the unet and the text encoder- e.g. text_encoder_lr:"
                 f" {args.text_encoder_lr} and learning_rate: {args.learning_rate}. "
                 f"When using prodigy only learning_rate is used as the initial learning rate."

examples/advanced_diffusion_training/train_dreambooth_lora_sdxl_advanced.py

@@ -1317,7 +1317,7 @@ def main(args):
 
             xformers_version = version.parse(xformers.__version__)
             if xformers_version == version.parse("0.0.16"):
-                logger.warn(
+                logger.warning(
                     "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, "
                     "please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                 )
@@ -1522,14 +1522,14 @@ def load_model_hook(models, input_dir):
 
     # Optimizer creation
     if not (args.optimizer.lower() == "prodigy" or args.optimizer.lower() == "adamw"):
-        logger.warn(
+        logger.warning(
             f"Unsupported choice of optimizer: {args.optimizer}.Supported optimizers include [adamW, prodigy]."
             "Defaulting to adamW"
         )
         args.optimizer = "adamw"
 
     if args.use_8bit_adam and not args.optimizer.lower() == "adamw":
-        logger.warn(
+        logger.warning(
             f"use_8bit_adam is ignored when optimizer is not set to 'AdamW'. Optimizer was "
             f"set to {args.optimizer.lower()}"
         )
@@ -1563,11 +1563,11 @@ def load_model_hook(models, input_dir):
         optimizer_class = prodigyopt.Prodigy
 
         if args.learning_rate <= 0.1:
-            logger.warn(
+            logger.warning(
                 "Learning rate is too low. When using prodigy, it's generally better to set learning rate around 1.0"
             )
         if args.train_text_encoder and args.text_encoder_lr:
-            logger.warn(
+            logger.warning(
                 f"Learning rates were provided both for the unet and the text encoder- e.g. text_encoder_lr:"
                 f" {args.text_encoder_lr} and learning_rate: {args.learning_rate}. "
                 f"When using prodigy only learning_rate is used as the initial learning rate."

examples/community/pipeline_stable_diffusion_xl_instantid.py

@@ -452,7 +452,7 @@ def cuda(self, dtype=torch.float16, use_xformers=False):
 
                 xformers_version = version.parse(xformers.__version__)
                 if xformers_version == version.parse("0.0.16"):
-                    logger.warn(
+                    logger.warning(
                         "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                     )
                 self.enable_xformers_memory_efficient_attention()

examples/consistency_distillation/train_lcm_distill_lora_sd_wds.py

@@ -308,7 +308,7 @@ def log_validation(vae, unet, args, accelerator, weight_dtype, step):
 
             tracker.log({"validation": formatted_images})
         else:
-            logger.warn(f"image logging not implemented for {tracker.name}")
+            logger.warning(f"image logging not implemented for {tracker.name}")
 
         del pipeline
         gc.collect()
@@ -1068,7 +1068,7 @@ def load_model_hook(models, input_dir):
 
             xformers_version = version.parse(xformers.__version__)
             if xformers_version == version.parse("0.0.16"):
-                logger.warn(
+                logger.warning(
                     "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                 )
             unet.enable_xformers_memory_efficient_attention()

examples/consistency_distillation/train_lcm_distill_lora_sdxl.py

@@ -180,7 +180,7 @@ def log_validation(vae, args, accelerator, weight_dtype, step, unet=None, is_fin
             logger_name = "test" if is_final_validation else "validation"
             tracker.log({logger_name: formatted_images})
         else:
-            logger.warn(f"image logging not implemented for {tracker.name}")
+            logger.warning(f"image logging not implemented for {tracker.name}")
 
         del pipeline
         gc.collect()
@@ -928,7 +928,7 @@ def load_model_hook(models, input_dir):
 
             xformers_version = version.parse(xformers.__version__)
             if xformers_version == version.parse("0.0.16"):
-                logger.warn(
+                logger.warning(
                     "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                 )
             unet.enable_xformers_memory_efficient_attention()

examples/consistency_distillation/train_lcm_distill_lora_sdxl_wds.py

@@ -325,7 +325,7 @@ def log_validation(vae, unet, args, accelerator, weight_dtype, step):
 
             tracker.log({"validation": formatted_images})
         else:
-            logger.warn(f"image logging not implemented for {tracker.name}")
+            logger.warning(f"image logging not implemented for {tracker.name}")
 
         del pipeline
         gc.collect()
@@ -1083,7 +1083,7 @@ def load_model_hook(models, input_dir):
 
             xformers_version = version.parse(xformers.__version__)
             if xformers_version == version.parse("0.0.16"):
-                logger.warn(
+                logger.warning(
                     "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                 )
             unet.enable_xformers_memory_efficient_attention()

examples/consistency_distillation/train_lcm_distill_sd_wds.py

@@ -285,7 +285,7 @@ def log_validation(vae, unet, args, accelerator, weight_dtype, step, name="targe
 
             tracker.log({f"validation/{name}": formatted_images})
         else:
-            logger.warn(f"image logging not implemented for {tracker.name}")
+            logger.warning(f"image logging not implemented for {tracker.name}")
 
         del pipeline
         gc.collect()
@@ -1023,7 +1023,7 @@ def load_model_hook(models, input_dir):
 
             xformers_version = version.parse(xformers.__version__)
             if xformers_version == version.parse("0.0.16"):
-                logger.warn(
+                logger.warning(
                     "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                 )
             unet.enable_xformers_memory_efficient_attention()

examples/consistency_distillation/train_lcm_distill_sdxl_wds.py

@@ -303,7 +303,7 @@ def log_validation(vae, unet, args, accelerator, weight_dtype, step, name="targe
 
             tracker.log({f"validation/{name}": formatted_images})
         else:
-            logger.warn(f"image logging not implemented for {tracker.name}")
+            logger.warning(f"image logging not implemented for {tracker.name}")
 
         del pipeline
         gc.collect()
@@ -1083,7 +1083,7 @@ def load_model_hook(models, input_dir):
 
             xformers_version = version.parse(xformers.__version__)
             if xformers_version == version.parse("0.0.16"):
-                logger.warn(
+                logger.warning(
                     "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                 )
             unet.enable_xformers_memory_efficient_attention()

examples/controlnet/train_controlnet.py

@@ -178,7 +178,7 @@ def log_validation(
 
             tracker.log({tracker_key: formatted_images})
         else:
-            logger.warn(f"image logging not implemented for {tracker.name}")
+            logger.warning(f"image logging not implemented for {tracker.name}")
 
         del pipeline
         gc.collect()
@@ -861,7 +861,7 @@ def load_model_hook(models, input_dir):
 
             xformers_version = version.parse(xformers.__version__)
             if xformers_version == version.parse("0.0.16"):
-                logger.warn(
+                logger.warning(
                     "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                 )
             unet.enable_xformers_memory_efficient_attention()

examples/controlnet/train_controlnet_flax.py

@@ -128,7 +128,7 @@ def log_validation(pipeline, pipeline_params, controlnet_params, tokenizer, args
 
         wandb.log({"validation": formatted_images})
     else:
-        logger.warn(f"image logging not implemented for {args.report_to}")
+        logger.warning(f"image logging not implemented for {args.report_to}")
 
     return image_logs
 

examples/controlnet/train_controlnet_sdxl.py

@@ -178,7 +178,7 @@ def log_validation(vae, unet, controlnet, args, accelerator, weight_dtype, step,
 
             tracker.log({tracker_key: formatted_images})
         else:
-            logger.warn(f"image logging not implemented for {tracker.name}")
+            logger.warning(f"image logging not implemented for {tracker.name}")
 
         del pipeline
         gc.collect()
@@ -929,7 +929,7 @@ def load_model_hook(models, input_dir):
 
             xformers_version = version.parse(xformers.__version__)
             if xformers_version == version.parse("0.0.16"):
-                logger.warn(
+                logger.warning(
                     "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                 )
             unet.enable_xformers_memory_efficient_attention()

examples/custom_diffusion/train_custom_diffusion.py

@@ -904,7 +904,7 @@ def main(args):
 
             xformers_version = version.parse(xformers.__version__)
             if xformers_version == version.parse("0.0.16"):
-                logger.warn(
+                logger.warning(
                     "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                 )
             attention_class = CustomDiffusionXFormersAttnProcessor

examples/dreambooth/train_dreambooth.py

@@ -987,7 +987,7 @@ def load_model_hook(models, input_dir):
 
             xformers_version = version.parse(xformers.__version__)
             if xformers_version == version.parse("0.0.16"):
-                logger.warn(
+                logger.warning(
                     "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                 )
             unet.enable_xformers_memory_efficient_attention()

examples/dreambooth/train_dreambooth_lora.py

@@ -895,7 +895,7 @@ def main(args):
 
             xformers_version = version.parse(xformers.__version__)
             if xformers_version == version.parse("0.0.16"):
-                logger.warn(
+                logger.warning(
                     "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                 )
             unet.enable_xformers_memory_efficient_attention()

examples/dreambooth/train_dreambooth_lora_sdxl.py

@@ -1141,7 +1141,7 @@ def main(args):
 
             xformers_version = version.parse(xformers.__version__)
             if xformers_version == version.parse("0.0.16"):
-                logger.warn(
+                logger.warning(
                     "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, "
                     "please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                 )
@@ -1317,14 +1317,14 @@ def load_model_hook(models, input_dir):
 
     # Optimizer creation
     if not (args.optimizer.lower() == "prodigy" or args.optimizer.lower() == "adamw"):
-        logger.warn(
+        logger.warning(
             f"Unsupported choice of optimizer: {args.optimizer}.Supported optimizers include [adamW, prodigy]."
             "Defaulting to adamW"
         )
         args.optimizer = "adamw"
 
     if args.use_8bit_adam and not args.optimizer.lower() == "adamw":
-        logger.warn(
+        logger.warning(
             f"use_8bit_adam is ignored when optimizer is not set to 'AdamW'. Optimizer was "
             f"set to {args.optimizer.lower()}"
         )
@@ -1358,11 +1358,11 @@ def load_model_hook(models, input_dir):
         optimizer_class = prodigyopt.Prodigy
 
         if args.learning_rate <= 0.1:
-            logger.warn(
+            logger.warning(
                 "Learning rate is too low. When using prodigy, it's generally better to set learning rate around 1.0"
             )
         if args.train_text_encoder and args.text_encoder_lr:
-            logger.warn(
+            logger.warning(
                 f"Learning rates were provided both for the unet and the text encoder- e.g. text_encoder_lr:"
                 f" {args.text_encoder_lr} and learning_rate: {args.learning_rate}. "
                 f"When using prodigy only learning_rate is used as the initial learning rate."

examples/instruct_pix2pix/train_instruct_pix2pix.py

@@ -488,7 +488,7 @@ def main():
 
             xformers_version = version.parse(xformers.__version__)
             if xformers_version == version.parse("0.0.16"):
-                logger.warn(
+                logger.warning(
                     "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                 )
             unet.enable_xformers_memory_efficient_attention()

examples/instruct_pix2pix/train_instruct_pix2pix_sdxl.py

@@ -580,7 +580,7 @@ def main():
 
             xformers_version = version.parse(xformers.__version__)
             if xformers_version == version.parse("0.0.16"):
-                logger.warn(
+                logger.warning(
                     "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                 )
             unet.enable_xformers_memory_efficient_attention()

examples/kandinsky2_2/text_to_image/train_text_to_image_decoder.py

@@ -177,7 +177,7 @@ def log_validation(vae, image_encoder, image_processor, unet, args, accelerator,
                 }
             )
         else:
-            logger.warn(f"image logging not implemented for {tracker.name}")
+            logger.warning(f"image logging not implemented for {tracker.name}")
 
     del pipeline
     torch.cuda.empty_cache()
@@ -534,7 +534,7 @@ def deepspeed_zero_init_disabled_context_manager():
 
             xformers_version = version.parse(xformers.__version__)
             if xformers_version == version.parse("0.0.16"):
-                logger.warn(
+                logger.warning(
                     "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                 )
             unet.enable_xformers_memory_efficient_attention()

examples/kandinsky2_2/text_to_image/train_text_to_image_prior.py

@@ -180,7 +180,7 @@ def log_validation(
                 }
             )
         else:
-            logger.warn(f"image logging not implemented for {tracker.name}")
+            logger.warning(f"image logging not implemented for {tracker.name}")
 
     del pipeline
     torch.cuda.empty_cache()

examples/research_projects/consistency_training/train_cm_ct_unconditional.py

@@ -219,7 +219,7 @@ def log_validation(unet, scheduler, args, accelerator, weight_dtype, step, name=
         if args.num_classes is not None:
             class_labels = list(range(args.num_classes))
         else:
-            logger.warn(
+            logger.warning(
                 "The model is class-conditional but the number of classes is not set. The generated images will be"
                 " unconditional rather than class-conditional."
             )
@@ -266,7 +266,7 @@ def log_validation(unet, scheduler, args, accelerator, weight_dtype, step, name=
 
             tracker.log({f"validation/{name}": formatted_images})
         else:
-            logger.warn(f"image logging not implemented for {tracker.name}")
+            logger.warning(f"image logging not implemented for {tracker.name}")
 
     del pipeline
     gc.collect()
@@ -863,14 +863,14 @@ def main(args):
     elif args.model_config_name_or_path is None:
         # TODO: use default architectures from iCT paper
         if not args.class_conditional and (args.num_classes is not None or args.class_embed_type is not None):
-            logger.warn(
+            logger.warning(
                 f"`--class_conditional` is set to `False` but `--num_classes` is set to {args.num_classes} and"
                 f" `--class_embed_type` is set to {args.class_embed_type}. These values will be overridden to `None`."
             )
             args.num_classes = None
             args.class_embed_type = None
         elif args.class_conditional and args.num_classes is None and args.class_embed_type is None:
-            logger.warn(
+            logger.warning(
                 "`--class_conditional` is set to `True` but neither `--num_classes` nor `--class_embed_type` is set."
                 "`class_conditional` will be overridden to `False`."
             )
@@ -996,7 +996,7 @@ def load_model_hook(models, input_dir):
 
             xformers_version = version.parse(xformers.__version__)
             if xformers_version == version.parse("0.0.16"):
-                logger.warn(
+                logger.warning(
                     "xFormers 0.0.16 cannot be used for training in some GPUs. If you observe problems during training, please update xFormers to at least 0.0.17. See https://huggingface.co/docs/diffusers/main/en/optimization/xformers for more details."
                 )
             unet.enable_xformers_memory_efficient_attention()