# Copyright (c) 2021 PaddlePaddle Authors. All Rights Reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and import paddle import paddle.fluid as fluid from .meta_parallel_base import MetaParallelBase from .parallel_layers.pp_layers import PipelineLayer from ..utils.hybrid_parallel_util import broadcast_mp_parameters from ..utils.hybrid_parallel_util import broadcast_dp_parameters from ..utils.hybrid_parallel_util import broadcast_sharding_parameters from ..utils.log_util import logger from ..meta_optimizers.dygraph_optimizer import HybridParallelOptimizer import paddle.fluid.framework as framework from .pp_utils import p2p_communication as p2p import paddle.fluid.core as core __all__ = [] class PipelineParallel(MetaParallelBase): def __init__(self, layers, hcg, strategy): if not isinstance(layers, PipelineLayer): raise TypeError( "The Layer should be a derived class of PipelineLayer." ) super(PipelineParallel, self).__init__(layers, hcg, strategy) self.use_data_parallel = self._hcg.get_data_parallel_world_size() > 1 self.use_model_parallel = self._hcg.get_model_parallel_world_size() > 1 self.use_sharding_parallel = ( self._hcg.get_sharding_parallel_world_size() > 1 ) self.total_loss = None self.micro_batch_size = self._strategy.pipeline_configs[ 'micro_batch_size' ] self.accumulate_steps = self._strategy.pipeline_configs[ 'accumulate_steps' ] # If sent tensor are not the same from different hosts, # they shouldn't been sent partially and then concated as a whole tensor. self._enable_partial_send_recv = self._strategy.pipeline_configs[ 'enable_partial_send_recv' ] self._using_cache = self._strategy.pipeline_configs['p2p_cache_shape'] self.num_stages = self._hcg.get_pipe_parallel_world_size() self.stage_id = self._hcg.get_stage_id() self.pp_group = self._hcg.get_pipe_parallel_group() self._virtual_pp_world_size = None self._virtual_pp_rank = None self._real_pp_world_size = self.num_stages self._real_pp_rank = self.stage_id p2p.initialize_p2p_groups( hcg, self._using_cache, self._enable_partial_send_recv ) self.global_rank = self._hcg.get_global_rank() self.micro_batch_id = 0 self._compute_loss = True logger.info( "Pipeline Info -- num_stages: {}, stage_id: {}".format( self.num_stages, self.stage_id ) ) if self.use_model_parallel: logger.info("start broadcast mp parameters") broadcast_mp_parameters(self._layers, self._hcg) if self.use_sharding_parallel: logger.info("start broadcast sharding parameters") broadcast_sharding_parameters(self._layers, self._hcg) if self.use_data_parallel: logger.info("start broadcast dp parameters") broadcast_dp_parameters(self._layers, self._hcg) def is_pipeline_first_stage(self, ignore_virtual=False): if not ignore_virtual: if self._virtual_pp_world_size is not None: assert self._virtual_pp_rank is not None if self._virtual_pp_rank != 0: return False assert self._real_pp_rank is not None return self._real_pp_rank == 0 def is_pipeline_last_stage(self, ignore_virtual=False): if not ignore_virtual: if self._virtual_pp_world_size is not None: assert self._virtual_pp_rank is not None if self._virtual_pp_rank != (self._virtual_pp_world_size - 1): return False assert self._real_pp_rank is not None assert self._real_pp_world_size is not None return self._real_pp_rank == (self._real_pp_world_size - 1) def set_virtual_pipeline_rank(self, rank): self._virtual_pp_rank = rank def forward_backward_pipeline(self, data, scaler=None): # use the 1f1b scheduling strategy. # this strategy is inspired by: # https://github.com/NVIDIA/Megatron-LM/blob/main/megatron/schedules.py self.scaler = scaler # store data for train self.data = data # store total loss of entire batch self.total_loss = None # store data id for micro_batch self.micro_batch_id = 0 startup_steps = self.num_stages - self.stage_id - 1 startup_steps = min(startup_steps, self.accumulate_steps) steady_steps = self.accumulate_steps - startup_steps input_buffers = [] output_buffers = [] for step_id in range(startup_steps): input_tensor = p2p.recv_forward(self.is_pipeline_first_stage()) output_tensor = self._forward_step(input_tensor) p2p.send_forward(output_tensor, self.is_pipeline_last_stage()) input_buffers.append(input_tensor) output_buffers.append(output_tensor) if steady_steps > 0: input_tensor = p2p.recv_forward(self.is_pipeline_first_stage()) for i in range(steady_steps): last_iter = i == (steady_steps - 1) output_tensor = self._forward_step(input_tensor) output_tensor_grad = p2p.send_forward_recv_backward( output_tensor, self.is_pipeline_last_stage() ) input_buffers.append(input_tensor) output_buffers.append(output_tensor) input_tensor, output_tensor = input_buffers.pop( 0 ), output_buffers.pop(0) input_tensor_grad = self._backward_step( input_tensor, output_tensor, output_tensor_grad ) if last_iter: input_tensor = None p2p.send_backward( input_tensor_grad, self.is_pipeline_first_stage() ) else: input_tensor = p2p.send_backward_recv_forward( input_tensor_grad, self.is_pipeline_first_stage() ) for i in range(startup_steps): input_tensor = input_buffers.pop(0) output_tensor = output_buffers.pop(0) output_tensor_grad = p2p.recv_backward( self.is_pipeline_last_stage() ) input_tensor_grad = self._backward_step( input_tensor, output_tensor, output_tensor_grad ) p2p.send_backward(input_tensor_grad, self.is_pipeline_first_stage()) self._layers.allreduce_shared_weight_gradients() with paddle.amp.auto_cast(enable=False): train_loss = self._broadcast_final_loss() return train_loss def _prepare_training(self, data, optimizer, lr_scheduler): # reset the virtual pp rank for each run self.set_virtual_pipeline_rank(0) assert isinstance( optimizer, HybridParallelOptimizer ), 'optimizer should be HybridParallelOptimizer subclass.' assert ( fluid.framework._dygraph_tracer()._has_grad ), 'Please enable the generation of gradients.' if self.is_pipeline_first_stage( ignore_virtual=True ) or self.is_pipeline_last_stage(ignore_virtual=True): assert ( data is not None ), "For the first and the last stage, the data must be set." else: data = None self.optimizer = optimizer self.lr_scheduler = lr_scheduler self._layers.train() return data def train_batch(self, data, optimizer, lr_scheduler=None, scaler=None): data = self._prepare_training(data, optimizer, lr_scheduler) # 1f1b scheduler for pipeline parallel train_loss = self.forward_backward_pipeline(data, scaler) # optimizer with paddle.amp.auto_cast(enable=False): self._optimizer_step() return train_loss def eval_batch(self, data, compute_loss=False): # reset the virtual pp rank for each run self.set_virtual_pipeline_rank(0) self._layers.eval() self._compute_loss = compute_loss # save data for eval self.data = data # store data id for micro_batch self.micro_batch_id = 0 # store total loss of entire batch self.total_loss = None startup_steps = self.num_stages - self.stage_id - 1 startup_steps = min(startup_steps, self.accumulate_steps) steady_steps = self.accumulate_steps - startup_steps input_buffers = [] output_buffers = [] for step_id in range(startup_steps): input_tensor = p2p.recv_forward(self.is_pipeline_first_stage()) output_tensor = self._forward_step(input_tensor) p2p.send_forward(output_tensor, self.is_pipeline_last_stage()) input_buffers.append(input_tensor) output_buffers.append(output_tensor) if steady_steps > 0: input_tensor = p2p.recv_forward(self.is_pipeline_first_stage()) for i in range(steady_steps): last_iter = i == (steady_steps - 1) output_tensor = self._forward_step(input_tensor) p2p.send_forward(output_tensor, self.is_pipeline_last_stage()) input_buffers.append(input_tensor) output_buffers.append(output_tensor) if not last_iter: input_tensor = p2p.recv_forward(self.is_pipeline_first_stage()) if self._compute_loss: self.train_loss = self._broadcast_final_loss() else: self.train_loss = output_buffers return self.train_loss def _forward_step(self, input_tensor, chunk_id=None): if self.is_pipeline_first_stage(): input_tensor = self._load_micro_batch(self.micro_batch_id) assert chunk_id is None or isinstance(chunk_id, int) output_tensor = self._layers.forward(input_tensor, chunk_id=chunk_id) if self.is_pipeline_last_stage(): # train calculate loss for train if self._compute_loss: assert ( self._layers._loss_fn is not None ), "loss function should exist to compute loss" labels = self._load_micro_batch(self.micro_batch_id) output_tensor = self._layers._loss_fn(output_tensor, labels) assert isinstance( output_tensor, (paddle.Tensor, core.eager.Tensor) ), "Currently, loss_fn should obtain Paddle.Tensor dtype" with paddle.amp.auto_cast(enable=False): if self.accumulate_steps > 1: output_tensor = output_tensor / self.accumulate_steps if self.total_loss is None: self.total_loss = paddle.zeros_like(output_tensor) self.total_loss += output_tensor.detach() if self.is_pipeline_first_stage() or self.is_pipeline_last_stage(): # Only increase micro batch id at virtual first/last pp stage. # The micro batch id is used to load data, therefore, only increase it when load data. self.micro_batch_id += 1 return output_tensor def _backward_step(self, input_tensor, output_tensor, output_tensor_grad): with paddle.amp.auto_cast(enable=False): if self.is_pipeline_last_stage(): assert output_tensor_grad is None if self.scaler: paddle.autograd.backward(self.scaler.scale(output_tensor)) else: paddle.autograd.backward(output_tensor) else: if isinstance(output_tensor, tuple): outputs = [t for t in output_tensor if not t.stop_gradient] assert len(outputs) == len(output_tensor_grad) paddle.autograd.backward( tensors=outputs, grad_tensors=[t for t in output_tensor_grad], ) else: paddle.autograd.backward( tensors=[output_tensor], grad_tensors=[output_tensor_grad], ) input_tensor_grad = None if input_tensor is not None: if isinstance(input_tensor, tuple): input_tensor_grad = tuple( [t.grad for t in input_tensor if not t.stop_gradient] ) else: input_tensor_grad = input_tensor.grad return input_tensor_grad def _load_micro_batch(self, cache_id): inputs = self.data begin = cache_id * self.micro_batch_size end = begin + self.micro_batch_size # The virtual first and last pipeline stage need data, all others don't need. if self.is_pipeline_first_stage(): assert len(inputs) == 2, "length of input should be 2" if isinstance(inputs[0], tuple): assert ( len(inputs[0]) > 1 ), "If you use tuple for input data, it should have at least two inputs." batch_size = inputs[0][0].shape[0] assert ( self.micro_batch_size * self.accumulate_steps == batch_size ), ( "batch_size needs to be divisible by micro_batch_size. Currently, " "batch_size = %d, micro_batch_size = %d, accumulate_steps = %d." % (batch_size, self.micro_batch_size, self.accumulate_steps) ) data = [input[begin:end, :].detach() for input in inputs[0]] return tuple(data) else: batch_size = inputs[0].shape[0] assert ( self.micro_batch_size * self.accumulate_steps == batch_size ) return inputs[0][begin:end, :].detach() elif self.is_pipeline_last_stage(): assert len(inputs) == 2, "length of input should be 2" if isinstance(inputs[1], tuple): batch_size = inputs[1][0].shape[0] assert ( self.micro_batch_size * self.accumulate_steps == batch_size ) data = [input[begin:end, :].detach() for input in inputs[1]] return tuple(data) else: batch_size = inputs[1].shape[0] assert ( self.micro_batch_size * self.accumulate_steps == batch_size ) return inputs[1][begin:end, :].detach() else: # No data input is required for other stages inputs = None def _broadcast_final_loss(self): # Since the last backward run in interleave will set the virtual rank to 0, # here we need to check last stage ignoring virtual stage. if self.is_pipeline_last_stage(ignore_virtual=True): assert ( self.total_loss is not None ), "train_batch() in last stage should obtain vaild loss" loss = self.total_loss.detach() is_fp32 = ( paddle.to_tensor(1) if loss.dtype == paddle.float32 else paddle.to_tensor(0) ) paddle.distributed.broadcast( is_fp32, src=self.global_rank, sync_op=True, group=self.pp_group ) paddle.distributed.broadcast( loss, src=self.global_rank, sync_op=True, group=self.pp_group ) else: is_fp32 = paddle.to_tensor(1) paddle.distributed.broadcast( is_fp32, src=self._hcg.get_rank_from_stage(self.num_stages - 1), sync_op=True, group=self.pp_group, ) loss = ( paddle.zeros(shape=[1], dtype="float32") if is_fp32.numpy()[0] else paddle.zeros(shape=[1], dtype="float16") ) paddle.distributed.broadcast( loss, src=self._hcg.get_rank_from_stage(self.num_stages - 1), sync_op=True, group=self.pp_group, ) return loss def _optimizer_step(self): if self.scaler: self.scaler.step(self.optimizer) self.scaler.update() else: self.optimizer.step() self.optimizer.clear_grad() if self.lr_scheduler: self.lr_scheduler.step() class PipelineParallelWithInterleave(PipelineParallel): # pipeline parallel with interleave scheduler def __init__(self, layers, hcg, strategy): super(PipelineParallelWithInterleave, self).__init__( layers=layers, hcg=hcg, strategy=strategy ) assert layers.get_num_virtual_stages() > 1 assert ( framework.in_dygraph_mode() ), "virtual pipeline stage with interleave only support eager dygraph mode" # setup for interleave scheduler self.num_model_chunks = layers.get_num_virtual_stages() self.model_chunks = layers.get_model_chunks() assert self.model_chunks is not None assert len(self.model_chunks) == self.num_model_chunks self._virtual_pp_world_size = self.num_model_chunks self._virtual_pp_rank = 0 def _get_virtual_pp_rank(self, micro_step, forward): virtual_pp_stage = micro_step % ( self.num_stages * self.num_model_chunks ) virtual_pp_stage = virtual_pp_stage // self.num_stages if not forward: virtual_pp_stage = self.num_model_chunks - virtual_pp_stage - 1 return virtual_pp_stage def _forward_step_helper(self, micro_step): virtual_pp_rank = self._get_virtual_pp_rank(micro_step, forward=True) self.set_virtual_pipeline_rank(virtual_pp_rank) # some checkers assert hasattr(self, 'input_tensors') assert hasattr(self, 'output_tensors') if not self._forward_only: assert hasattr(self, 'output_tensor_grads') if self.is_pipeline_first_stage(): if len(self.input_tensors[virtual_pp_rank]) == len( self.output_tensors[virtual_pp_rank] ): self.input_tensors[virtual_pp_rank].append(None) input_tensor = self.input_tensors[virtual_pp_rank][-1] output_tensor = self._forward_step(input_tensor, virtual_pp_rank) self.output_tensors[virtual_pp_rank].append(output_tensor) if self._forward_only: # no need to store tensor for backward self.input_tensors[virtual_pp_rank].pop() self.output_tensors[virtual_pp_rank].pop() return output_tensor def _backward_step_helper(self, micro_step): virtual_pp_rank = self._get_virtual_pp_rank(micro_step, forward=False) self.set_virtual_pipeline_rank(virtual_pp_rank) # some checkers assert hasattr(self, 'input_tensors') assert hasattr(self, 'output_tensors') assert hasattr(self, 'output_tensor_grads') if self.is_pipeline_last_stage(): if len(self.output_tensor_grads[virtual_pp_rank]) == 0: self.output_tensor_grads[virtual_pp_rank].append(None) input_tensor = self.input_tensors[virtual_pp_rank].pop(0) output_tensor = self.output_tensors[virtual_pp_rank].pop(0) output_tensor_grad = self.output_tensor_grads[virtual_pp_rank].pop(0) input_tensor_grad = self._backward_step( input_tensor, output_tensor, output_tensor_grad ) return input_tensor_grad def interleave_pipeline( self, data, scaler, forward_only=False, compute_loss=True ): # use interleave scheduling strategy. # this strategy is inspired by: # https://github.com/NVIDIA/Megatron-LM/blob/main/megatron/schedules.py if not compute_loss: assert ( not forward_only ), "compute_loss can only be set to False when forward_only is set to True" # init some attributes for this batch run self.scaler = scaler self.data = data self.total_loss = None self.micro_batch_id = 0 self._forward_only = forward_only # init some data buffers for interleave scheduler self.input_tensors = [[] for _ in range(self.num_model_chunks)] self.output_tensors = [[] for _ in range(self.num_model_chunks)] self.output_tensor_grads = [[] for _ in range(self.num_model_chunks)] num_steps = self.accumulate_steps * self.num_model_chunks all_startup_steps = False if forward_only: # If only forward, since there is no backward during running, all steps are startup steps startup_steps = num_steps else: if self.accumulate_steps == self.num_stages: startup_steps = num_steps all_startup_steps = True else: startup_steps = (self.num_stages - self.stage_id - 1) * 2 startup_steps += (self.num_model_chunks - 1) * self.num_stages startup_steps = min(startup_steps, num_steps) steady_steps = num_steps - startup_steps self.set_virtual_pipeline_rank(0) self.input_tensors[0].append( p2p.recv_forward(self.is_pipeline_first_stage(), sync_recv=False) ) # run startup steps for micro_step in range(startup_steps): output_tensor = self._forward_step_helper(micro_step) # determine whether recv forward tensor or not next_virtual_pp_rank = self._get_virtual_pp_rank( micro_step + 1, forward=True ) recv_prev = True if self.is_pipeline_first_stage(ignore_virtual=True): if next_virtual_pp_rank == 0: # next chunk is the first chunk, not need to pre recv an input tensor recv_prev = False # last micro step, no next run if micro_step == (num_steps - 1): recv_prev = False # last stage shouldn't send tensor to downstream if self.is_pipeline_last_stage(): output_tensor = None if ( micro_step == (startup_steps - 1) and not forward_only and not all_startup_steps ): input_tensor_grad = None recv_next = True if self.is_pipeline_last_stage(ignore_virtual=True): recv_next = False # the last startup step needs on four direction comm to set up for steady 1f1b ( input_tensor, output_tensor_grad, ) = p2p.send_forward_backward_recv_forward_backward( output_tensor, input_tensor_grad, recv_prev=recv_prev, recv_next=recv_next, ) self.output_tensor_grads[self.num_model_chunks - 1].append( output_tensor_grad ) else: input_tensor = p2p.send_forward_recv_forward( output_tensor, recv_prev=recv_prev ) self.input_tensors[next_virtual_pp_rank].append(input_tensor) # run 1f1b steady steps for micro_step in range(steady_steps): # forward forward_micro_step_id = micro_step + startup_steps output_tensor = self._forward_step_helper(forward_micro_step_id) # backward backward_micro_step_id = micro_step input_tensor_grad = self._backward_step_helper( backward_micro_step_id ) # four directions comm # send output tensor to downstream # send input tensor grad to upstream # recv input tensor from upstream # recv output tensor grad from downstream # last stage doesn't send rst to downstream forward_virtual_pp_rank = self._get_virtual_pp_rank( forward_micro_step_id, forward=True ) self.set_virtual_pipeline_rank(forward_virtual_pp_rank) if self.is_pipeline_last_stage(): output_tensor = None # first stage doesn't send grad to upstream backward_virtual_pp_rank = self._get_virtual_pp_rank( backward_micro_step_id, forward=False ) self.set_virtual_pipeline_rank(backward_virtual_pp_rank) if self.is_pipeline_first_stage(): input_tensor_grad = None # determine whether to recv input tensor from upstream recv_prev = True if self.is_pipeline_first_stage(ignore_virtual=True): next_forward_virtual_pp_rank = self._get_virtual_pp_rank( forward_micro_step_id - (self.num_stages - 1), forward=True ) if next_forward_virtual_pp_rank == (self.num_model_chunks - 1): # first pp stage and first virtual stage recv_prev = False next_forward_virtual_pp_rank += 1 else: next_forward_virtual_pp_rank = self._get_virtual_pp_rank( forward_micro_step_id + 1, forward=True ) # last iteration doesn't need recv from upstream if micro_step == (steady_steps - 1): recv_prev = False # determine whether to recv grad from downstream recv_next = True if self.is_pipeline_last_stage(ignore_virtual=True): next_backward_virtual_pp_rank = self._get_virtual_pp_rank( backward_micro_step_id - (self.num_stages - 1), forward=False, ) if next_backward_virtual_pp_rank == 0: # last pp stage and last virtual stage recv_next = False next_backward_virtual_pp_rank -= 1 else: next_backward_virtual_pp_rank = self._get_virtual_pp_rank( backward_micro_step_id + 1, forward=False ) ( input_tensor, output_tensor_grad, ) = p2p.send_forward_backward_recv_forward_backward( output_tensor, input_tensor_grad, recv_prev=recv_prev, recv_next=recv_next, ) if recv_prev: self.input_tensors[next_forward_virtual_pp_rank].append( input_tensor ) if recv_next: self.output_tensor_grads[next_backward_virtual_pp_rank].append( output_tensor_grad ) # remaining backward steps if not forward_only: if all_startup_steps: self.output_tensor_grads[self.num_model_chunks - 1].append( p2p.recv_backward( self.is_pipeline_last_stage(), sync_recv=False ) ) for micro_step in range(steady_steps, num_steps): # cooldown loop input_tensor_grad = self._backward_step_helper(micro_step) next_backward_virtual_pp_rank = self._get_virtual_pp_rank( micro_step + 1, forward=False ) recv_next = True if self.is_pipeline_last_stage(ignore_virtual=True): if next_backward_virtual_pp_rank == ( self.num_model_chunks - 1 ): recv_next = False if micro_step == (num_steps - 1): recv_next = False self.output_tensor_grads[next_backward_virtual_pp_rank].append( p2p.send_backward_recv_backward( input_tensor_grad, recv_next=recv_next ) ) self._layers.allreduce_shared_weight_gradients() if compute_loss: # return loss if compute loss with paddle.amp.auto_cast(enable=False): train_loss = self._broadcast_final_loss() else: # else just return all intermediate output tensor for all micro steps train_loss = self.output_tensors return train_loss def train_batch(self, data, optimizer, lr_scheduler=None, scaler=None): data = self._prepare_training(data, optimizer, lr_scheduler) # interleave scheduler for pipeline parallel train_loss = self.interleave_pipeline(data, scaler) # optimizer with paddle.amp.auto_cast(enable=False): self._optimizer_step() return train_loss def eval_batch(self, data, compute_loss=False): # reset the virtual pp rank for each run self.set_virtual_pipeline_rank(0) self._layers.eval() self._compute_loss = compute_loss return self.interleave_pipeline(data, None, forward_only=True)