fix problem of 'accelerator.is_main_process' to run in mutiple GPUs (h…

…uggingface#5340)

fix problem of 'accelerator.is_main_process' to run in mutiple GPUs or NPUs

Co-authored-by: jiaqiw <wangjiaqi50@huawei.com>

examples/unconditional_image_generation/train_unconditional.py

@@ -607,28 +607,28 @@ def transform_images(examples):
                 progress_bar.update(1)
                 global_step += 1
 
-                if global_step % args.checkpointing_steps == 0:
-                    # _before_ saving state, check if this save would set us over the `checkpoints_total_limit`
-                    if args.checkpoints_total_limit is not None:
-                        checkpoints = os.listdir(args.output_dir)
-                        checkpoints = [d for d in checkpoints if d.startswith("checkpoint")]
-                        checkpoints = sorted(checkpoints, key=lambda x: int(x.split("-")[1]))
-
-                        # before we save the new checkpoint, we need to have at _most_ `checkpoints_total_limit - 1` checkpoints
-                        if len(checkpoints) >= args.checkpoints_total_limit:
-                            num_to_remove = len(checkpoints) - args.checkpoints_total_limit + 1
-                            removing_checkpoints = checkpoints[0:num_to_remove]
-
-                            logger.info(
-                                f"{len(checkpoints)} checkpoints already exist, removing {len(removing_checkpoints)} checkpoints"
-                            )
-                            logger.info(f"removing checkpoints: {', '.join(removing_checkpoints)}")
-
-                            for removing_checkpoint in removing_checkpoints:
-                                removing_checkpoint = os.path.join(args.output_dir, removing_checkpoint)
-                                shutil.rmtree(removing_checkpoint)
-
-                    if accelerator.is_main_process:
+                if accelerator.is_main_process:
+                    if global_step % args.checkpointing_steps == 0:
+                        # _before_ saving state, check if this save would set us over the `checkpoints_total_limit`
+                        if args.checkpoints_total_limit is not None:
+                            checkpoints = os.listdir(args.output_dir)
+                            checkpoints = [d for d in checkpoints if d.startswith("checkpoint")]
+                            checkpoints = sorted(checkpoints, key=lambda x: int(x.split("-")[1]))
+
+                            # before we save the new checkpoint, we need to have at _most_ `checkpoints_total_limit - 1` checkpoints
+                            if len(checkpoints) >= args.checkpoints_total_limit:
+                                num_to_remove = len(checkpoints) - args.checkpoints_total_limit + 1
+                                removing_checkpoints = checkpoints[0:num_to_remove]
+
+                                logger.info(
+                                    f"{len(checkpoints)} checkpoints already exist, removing {len(removing_checkpoints)} checkpoints"
+                                )
+                                logger.info(f"removing checkpoints: {', '.join(removing_checkpoints)}")
+
+                                for removing_checkpoint in removing_checkpoints:
+                                    removing_checkpoint = os.path.join(args.output_dir, removing_checkpoint)
+                                    shutil.rmtree(removing_checkpoint)
+
                         save_path = os.path.join(args.output_dir, f"checkpoint-{global_step}")
                         accelerator.save_state(save_path)
                         logger.info(f"Saved state to {save_path}")