update

游雁

2024-03-26 abf5af40e934216b397c5331e0a68dc92f0a4f4e

 funasr/train_utils/trainer.py

@@ -239,6 +239,8 @@
        Args:
            epoch (int): The current epoch number.
        """
        if self.use_ddp or self.use_fsdp:
            dist.barrier()
        logging.info(f"Train epoch: {epoch}, rank: {self.local_rank}\n")
        model.train()

@@ -249,6 +251,8 @@
        speed_stats = {}
        time5 = time.perf_counter()
        iterator_stop = torch.tensor(0).to(self.device)

        dataloader_train.batch_sampler.set_epoch(epoch)
        for batch_idx, batch in enumerate(dataloader_train):
            if self.use_ddp or self.use_fsdp:
                dist.all_reduce(iterator_stop, dist.ReduceOp.SUM)
@@ -294,13 +298,13 @@
                self.train_loss_avg = (self.train_loss_avg*batch_idx + loss.detach().cpu().item())/(batch_idx+1)
                if "acc" in stats:
                    self.train_acc_avg = (self.train_acc_avg * batch_idx + stats["acc"].detach().cpu().item()) / (batch_idx + 1)
                # if self.use_ddp or self.use_fsdp:
                #     train_loss_avg = torch.tensor(self.train_loss_avg, dtype=torch.float32).to(self.device)
                #     train_acc_avg = torch.tensor(self.train_acc_avg, dtype=torch.float32).to(self.device)
                #     dist.all_reduce(train_loss_avg, op=dist.ReduceOp.SUM)
                #     dist.all_reduce(train_acc_avg, op=dist.ReduceOp.SUM)
                #     self.train_loss_avg = train_loss_avg.detach().cpu().item() / self.world_size
                #     self.train_acc_avg = train_acc_avg.detach().cpu().item() / self.world_size
                if self.use_ddp or self.use_fsdp:
                    train_loss_avg = torch.tensor(self.train_loss_avg, dtype=torch.float32).to(self.device)
                    train_acc_avg = torch.tensor(self.train_acc_avg, dtype=torch.float32).to(self.device)
                    dist.all_reduce(train_loss_avg, op=dist.ReduceOp.SUM)
                    dist.all_reduce(train_acc_avg, op=dist.ReduceOp.SUM)
                    self.train_loss_avg = train_loss_avg.detach().cpu().item() / self.world_size
                    self.train_acc_avg = train_acc_avg.detach().cpu().item() / self.world_size
                
            
            # Perform an optimizer step only after accumulating enough gradients
@@ -336,7 +340,7 @@
    
                speed_stats["total_time"] = total_time
                lr = scheduler.get_last_lr()[0]
                batch_num_epoch = -1
                batch_num_epoch = 1
                if hasattr(dataloader_train, "__len__"):
                    batch_num_epoch = len(dataloader_train)
                self.log(epoch, batch_idx,
@@ -367,6 +371,7 @@
                
        if self.use_ddp or self.use_fsdp:
            dist.barrier()
            iterator_stop = torch.tensor(0).to(self.device)
        
        

@@ -384,6 +389,8 @@
        Args:
            epoch (int): The current epoch number.
        """
        if self.use_ddp or self.use_fsdp:
            dist.barrier()
        logging.info(f"Validate epoch: {epoch}, rank: {self.local_rank}\n")
        model.eval()
        
@@ -392,6 +399,7 @@
            speed_stats = {}
            time5 = time.perf_counter()
            iterator_stop = torch.tensor(0).to(self.device)
            dataloader_val.batch_sampler.set_epoch(epoch)
            for batch_idx, batch in enumerate(dataloader_val):
                if self.use_ddp or self.use_fsdp:
                    dist.all_reduce(iterator_stop, dist.ReduceOp.SUM)
@@ -410,7 +418,7 @@
                    # Apply weighted averaging for loss and stats
                    loss = (loss * weight.type(loss.dtype)).sum()
                    # if distributed, this method can also apply all_reduce()
                    stats, weight = recursive_average(stats, weight, distributed=True)
                    # stats, weight = recursive_average(stats, weight, distributed=True)
                    if self.use_ddp or self.use_fsdp:
                        dist.all_reduce(weight, op=dist.ReduceOp.SUM)
                    # Now weight is summation over all workers
@@ -425,15 +433,15 @@
                self.val_loss_avg = (self.val_loss_avg*batch_idx + loss.detach().cpu().item())/(batch_idx+1)
                if "acc" in stats:
                    self.val_acc_avg = (self.val_acc_avg * batch_idx + stats["acc"].detach().cpu().item()) / (batch_idx + 1)
                # if self.use_ddp or self.use_fsdp:
                #     val_loss_avg = torch.tensor(self.val_loss_avg, dtype=torch.float32).to(self.device)
                #     val_acc_avg = torch.tensor(self.val_acc_avg, dtype=torch.float32).to(self.device)
                #     dist.all_reduce(val_loss_avg, op=dist.ReduceOp.SUM)
                #     dist.all_reduce(val_acc_avg, op=dist.ReduceOp.SUM)
                #     self.val_loss_avg = val_loss_avg.detach().cpu().item() / self.world_size
                #     self.val_acc_avg = val_acc_avg.detach().cpu().item() / self.world_size
                if self.use_ddp or self.use_fsdp:
                    val_loss_avg = torch.tensor(self.val_loss_avg, dtype=torch.float32).to(self.device)
                    val_acc_avg = torch.tensor(self.val_acc_avg, dtype=torch.float32).to(self.device)
                    dist.all_reduce(val_loss_avg, op=dist.ReduceOp.SUM)
                    dist.all_reduce(val_acc_avg, op=dist.ReduceOp.SUM)
                    self.val_loss_avg = val_loss_avg.detach().cpu().item() / self.world_size
                    self.val_acc_avg = val_acc_avg.detach().cpu().item() / self.world_size
                
                batch_num_epoch = -1
                batch_num_epoch = 1
                if hasattr(dataloader_val, "__len__"):
                    batch_num_epoch = len(dataloader_val)
                self.log(epoch, batch_idx,
@@ -454,10 +462,9 @@
        self.val_acc_list.append(self.val_acc_avg)
        model.train()



        if self.use_ddp or self.use_fsdp:
            dist.barrier()
            iterator_stop = torch.tensor(0).to(self.device)
        
        
    def log(self,