[autoparallel] add following node generator (#1673)

* [autoparallel] add following node generator

* polish code

* polish code

* update name of arguments

colossalai/auto_parallel/solver/op_handler/getitem_handler.py

@@ -0,0 +1,39 @@
+import torch
+from .node_handler import NodeHandler
+from ..sharding_strategy import ShardingStrategy_V2, OperationDataType, OperationData, StrategiesVector
+from ..strategy import TensorStrategyGenerator, TensorTupleStrategyGenerator, StrategyGenerator_V2
+from typing import List, Dict
+from .registry import operator_registry
+import operator
+
+__all__ = ['GetItemHandler']
+
+
+@operator_registry.register(operator.getitem)
+class GetItemHandler(NodeHandler):
+    """
+    A GetItemHandler which deals with the sharding strategies for operator.getitem.
+    """
+
+    def get_strategy_generator(self) -> List[StrategyGenerator_V2]:
+        op_data_mapping = self.get_operation_data_mapping()
+        generators = []
+        if isinstance(op_data_mapping["input"].data, torch.Tensor):
+            generators.append(TensorStrategyGenerator(op_data_mapping, self.device_mesh, self.node.args[0]))
+        else:
+            generators.append(TensorTupleStrategyGenerator(op_data_mapping, self.device_mesh, self.node.args[0]))
+
+        return generators
+
+    def get_operation_data_mapping(self) -> Dict[str, OperationData]:
+        # use transposed shape for strategies
+        # the strategies will be transformed back to its original shape in self.post_process
+        physical_input_operand = OperationData(name=str(self.node.args[0]),
+                                               type=OperationDataType.ARG,
+                                               data=self.node.args[0]._meta_data)
+        physical_other_operand = OperationData(name="index", type=OperationDataType.ARG, data=self.node.args[1])
+        physical_output = OperationData(name=str(self.node), type=OperationDataType.OUTPUT, data=self.node._meta_data)
+
+        mapping = {"input": physical_input_operand, "index": physical_other_operand, "output": physical_output}
+
+        return mapping

colossalai/auto_parallel/solver/sharding_strategy.py

@@ -63,24 +63,27 @@ class OperationData:
     Args:
         name (str): the name of the operation-related data
         type (OperationDataType): the type of the operation data
-        data (torch.Tensor): the value for this data, usually it is a meta tensor.
+        data (Any): the value for this data, usually it is a meta tensor.
         logical_shape (Tuple[int]): the logical shape of the data, it can be different from the its actual shape in memory.
     """
     name: str
     type: OperationDataType
-    data: torch.Tensor
+    data: Any
     logical_shape: Tuple[int] = None
 
     def __post_init__(self):
         # if no logical shape is specified, use the data shape as the logical shape
-        if self.logical_shape is None:
+        if self.logical_shape is None and isinstance(self.data, torch.Tensor):
             self.logical_shape = self.data.shape
 
     def __repr__(self) -> str:
         return f'OperationData(name={self.name}, type={self.type})'
 
+    def __eq__(self, other) -> bool:
+        return other.name == self.name
+
     def __hash__(self) -> int:
-        return hash(f'{self.name}-{self.type}')
+        return hash(f'{self.name}')
 
 
 @dataclass
@@ -123,7 +126,7 @@ class ShardingStrategy_V2:
                                                   strategy.(default to None)
     """
     name: str
-    sharding_specs: Dict[OperationData, ShardingSpec] = None
+    sharding_specs: Dict[OperationData, Union[ShardingSpec, Tuple[ShardingSpec]]] = None
     compute_cost: TrainCycleItem = None
     communication_cost: TrainCycleItem = None
     memory_cost: TrainCycleItem = None

colossalai/auto_parallel/solver/strategy/__init__.py

@@ -2,10 +2,12 @@ from .strategy_generator import StrategyGenerator_V2
 from .matmul_strategy_generator import DotProductStrategyGenerator, MatVecStrategyGenerator, LinearProjectionStrategyGenerator, BatchedMatMulStrategyGenerator
 from .conv_strategy_generator import ConvStrategyGenerator
 from .batch_norm_generator import BatchNormStrategyGenerator
+from .getitem_generator import GetItemStrategyGenerator, TensorStrategyGenerator, TensorTupleStrategyGenerator
 from .layer_norm_generator import LayerNormGenerator
 
 __all__ = [
     'StrategyGenerator_V2', 'DotProductStrategyGenerator', 'MatVecStrategyGenerator',
     'LinearProjectionStrategyGenerator', 'BatchedMatMulStrategyGenerator', 'ConvStrategyGenerator',
-    'BatchNormStrategyGenerator', 'LayerNormGenerator'
+    'BatchNormStrategyGenerator', 'GetItemStrategyGenerator', 'TensorStrategyGenerator', 'TensorTupleStrategyGenerator',
+    'LayerNormGenerator'
 ]

colossalai/auto_parallel/solver/strategy/batch_norm_generator.py

@@ -86,12 +86,12 @@ class BatchNormStrategyGenerator(StrategyGenerator_V2):
         # compute bwd cost incurred
         # bwd_cost = input_grad + other_grad + bias_grad
         bwd_activation_cost = sum([v for k, v in backward_size_mapping.items() if not self.is_param(k)])
-        bwd_activation_cost = sum([v for k, v in backward_size_mapping.items() if self.is_param(k)])
+        bwd_parameter_cost = sum([v for k, v in backward_size_mapping.items() if self.is_param(k)])
-        bwd_mem_cost = MemoryCost(activation=bwd_activation_cost, parameter=bwd_activation_cost)
+        bwd_mem_cost = MemoryCost(activation=bwd_activation_cost, parameter=bwd_parameter_cost)
 
         # compute total cost
         total_mem_cost = MemoryCost(activation=fwd_activation_cost + bwd_activation_cost,
-                                    parameter=fwd_parameter_cost + bwd_activation_cost)
+                                    parameter=fwd_parameter_cost + bwd_parameter_cost)
         memory_cost = TrainCycleItem(fwd=fwd_mem_cost, bwd=bwd_mem_cost, total=total_mem_cost)
         strategy.memory_cost = memory_cost
 
@@ -288,4 +288,9 @@ class BatchNormStrategyGenerator(StrategyGenerator_V2):
         # S01R = S01R x R WITH SYNC_BN
         strategy_list.append(self.split_input_batch_1d(0, 1))
 
+        for strategy in strategy_list:
+            self.update_communication_cost(strategy)
+            self.update_compute_cost(strategy)
+            self.update_memory_cost(strategy)
+
         return strategy_list

colossalai/auto_parallel/solver/strategy/conv_strategy_generator.py

@@ -91,12 +91,12 @@ class ConvStrategyGenerator(StrategyGenerator_V2):
         # compute bwd cost incurred
         # bwd_cost = input_grad + other_grad + bias_grad
         bwd_activation_cost = sum([v for k, v in backward_size_mapping.items() if not self.is_param(k)])
-        bwd_activation_cost = sum([v for k, v in backward_size_mapping.items() if self.is_param(k)])
+        bwd_parameter_cost = sum([v for k, v in backward_size_mapping.items() if self.is_param(k)])
-        bwd_mem_cost = MemoryCost(activation=bwd_activation_cost, parameter=bwd_activation_cost)
+        bwd_mem_cost = MemoryCost(activation=bwd_activation_cost, parameter=bwd_parameter_cost)
 
         # compute total cost
         total_mem_cost = MemoryCost(activation=fwd_activation_cost + bwd_activation_cost,
-                                    parameter=fwd_parameter_cost + bwd_activation_cost)
+                                    parameter=fwd_parameter_cost + bwd_parameter_cost)
         memory_cost = TrainCycleItem(fwd=fwd_mem_cost, bwd=bwd_mem_cost, total=total_mem_cost)
         strategy.memory_cost = memory_cost
 

colossalai/auto_parallel/solver/strategy/getitem_generator.py

@@ -0,0 +1,147 @@
+import operator
+from functools import reduce
+from ..sharding_strategy import ShardingStrategy_V2, TrainCycleItem, MemoryCost
+from colossalai.tensor.shape_consistency import CollectiveCommPattern
+from .strategy_generator import FollowingStrategyGenerator
+from typing import List
+from .._utils import exception_handler
+import copy
+
+__all__ = ['GetItemStrategyGenerator', 'TensorStrategyGenerator', 'TensorTupleStrategyGenerator']
+
+
+class GetItemStrategyGenerator(FollowingStrategyGenerator):
+    """
+    GetItemStrategyGenerator is a generic class to generate strategies for operator.getitem.
+    The operation data is defined as `output = input[other]`.
+
+    There are mainly three use cases:
+        1. args_0._meta_data: torch.Tensor, args_1._meta_data: int
+        2. args_0._meta_data: torch.Tensor, args_1._meta_data: slice
+        3. args_0._meta_data: Tuple[torch.Tensor], args_1._meta_data: int
+    """
+
+    @property
+    def has_bias(self):
+        return 'bias' in self.op_data
+
+    def validate(self) -> bool:
+        return super().validate()
+
+    def update_compute_cost(self, strategy: ShardingStrategy_V2) -> TrainCycleItem:
+        return TrainCycleItem(fwd=10, bwd=10, total=20)
+
+    def update_memory_cost(self, strategy: ShardingStrategy_V2) -> TrainCycleItem:
+        '''
+        Compute the memory cost per device with this specific strategy.
+        '''
+        forward_size_mapping = {
+            'input': self._compute_size_in_bytes(strategy, "input"),
+            'output': self._compute_size_in_bytes(strategy, "output")
+        }
+
+        backward_size_mapping = copy.deepcopy(forward_size_mapping)
+        backward_size_mapping.pop("output")
+        # compute fwd cost incurred
+        # fwd_cost = input + output
+        fwd_activation_cost = sum([v for k, v in forward_size_mapping.items() if not self.is_param(k)])
+        fwd_parameter_cost = sum([v for k, v in forward_size_mapping.items() if self.is_param(k)])
+        fwd_mem_cost = MemoryCost(activation=fwd_activation_cost, parameter=fwd_parameter_cost)
+
+        # compute bwd cost incurred
+        # bwd_cost = input_grad
+        bwd_activation_cost = sum([v for k, v in backward_size_mapping.items() if not self.is_param(k)])
+        bwd_parameter_cost = sum([v for k, v in backward_size_mapping.items() if self.is_param(k)])
+        bwd_mem_cost = MemoryCost(activation=bwd_activation_cost, parameter=bwd_parameter_cost)
+
+        # compute total cost
+        total_mem_cost = MemoryCost(activation=fwd_activation_cost + bwd_activation_cost,
+                                    parameter=fwd_parameter_cost + bwd_parameter_cost)
+        memory_cost = TrainCycleItem(fwd=fwd_mem_cost, bwd=bwd_mem_cost, total=total_mem_cost)
+        strategy.memory_cost = memory_cost
+        return super().update_memory_cost(strategy)
+
+
+class TensorStrategyGenerator(GetItemStrategyGenerator):
+    '''
+    Deal with case 1 and 2.
+    '''
+
+    def generate(self):
+        strategy_list = []
+        for strategy in self.predecessor_node.strategies_vector:
+            dim_partition_dict_mapping = {}
+            communication_action_mapping = {}
+            dim_partition_dict_for_input = strategy.output_sharding_specs[self.op_data["input"]].dim_partition_dict
+            dim_partition_dict_for_output = copy.deepcopy(dim_partition_dict_for_input)
+            gather_input = 0 in dim_partition_dict_for_input
+            if gather_input:
+                logical_process_axis = dim_partition_dict_for_output.pop(0)
+
+            shift_dim_partition_dict_for_output = {}
+            for dim, mesh_dim_list in dim_partition_dict_for_output.items():
+                shift_dim_partition_dict_for_output[dim - 1] = mesh_dim_list
+            dim_partition_dict_for_output = shift_dim_partition_dict_for_output
+            dim_partition_dict_mapping = {
+                "input": dim_partition_dict_for_input,
+                "output": dim_partition_dict_for_output,
+            }
+            sharding_spec_mapping = self.to_sharding_spec_mapping(dim_partition_dict_mapping)
+            if gather_input:
+                input_communication_spec = self.get_communication_spec(
+                    sharding_spec_mapping["input"],
+                    communication_pattern=CollectiveCommPattern.GATHER_FWD_SPLIT_BWD,
+                    logical_process_axis=logical_process_axis)
+                communication_action_mapping["input"] = input_communication_spec
+
+            name = f'{sharding_spec_mapping["output"].sharding_sequence} = {sharding_spec_mapping["input"].sharding_sequence}'
+
+            strategy = self.get_sharding_strategy(name=name,
+                                                  sharding_spec_mapping=sharding_spec_mapping,
+                                                  communication_action_mapping=communication_action_mapping)
+
+            strategy_list.append(strategy)
+
+        for strategy in strategy_list:
+            self.update_communication_cost(strategy)
+            self.update_compute_cost(strategy)
+            self.update_memory_cost(strategy)
+
+        return strategy_list
+
+
+class TensorTupleStrategyGenerator(GetItemStrategyGenerator):
+    '''
+    Deal with case 3.
+    '''
+
+    def generate(self):
+        strategy_list = []
+        index = self.op_data["index"].data
+
+        for strategy in self.predecessor_node.strategies_vector:
+            # the sharding spec for input in this case is a tuple of ShardingSpec.
+            sharding_spec_for_input = strategy.output_sharding_specs[self.op_data["input"]]
+            dim_partition_dict_for_output = sharding_spec_for_input[index].dim_partition_dict
+            dim_partition_dict_mapping = {}
+            communication_action_mapping = {}
+            dim_partition_dict_mapping = {
+                "output": dim_partition_dict_for_output,
+            }
+            sharding_spec_mapping = self.to_sharding_spec_mapping(dim_partition_dict_mapping)
+            sharding_spec_mapping["input"] = sharding_spec_for_input
+
+            name = f'{sharding_spec_mapping["output"].sharding_sequence} = {sharding_spec_mapping["input"].sharding_sequence}'
+
+            strategy = self.get_sharding_strategy(name=name,
+                                                  sharding_spec_mapping=sharding_spec_mapping,
+                                                  communication_action_mapping=communication_action_mapping)
+
+            strategy_list.append(strategy)
+
+        for strategy in strategy_list:
+            self.update_communication_cost(strategy)
+            self.update_compute_cost(strategy)
+            self.update_memory_cost(strategy)
+
+        return strategy_list

colossalai/auto_parallel/solver/strategy/strategy_generator.py

@@ -8,6 +8,8 @@ from colossalai.tensor.sharding_spec import ShardingSpec
 from colossalai.device.device_mesh import DeviceMesh
 from typing import Dict, List, Union, Any
 from ..sharding_strategy import OperationData, ShardingStrategy_V2, TrainCycleItem, OperationDataType
+from torch.fx import Node
+import copy
 
 
 class StrategyGenerator_V2(ABC):
@@ -72,10 +74,6 @@ class StrategyGenerator_V2(ABC):
         """
         A factory method to produce a CommSpec object.
         """
-        # use flatten device mesh the same action is applied to two axes
-        if isinstance(logical_process_axis, list) and len(logical_process_axis) == 2:
-            sharding_spec.device_mesh = sharding_spec.device_mesh.flatten()
-            logical_process_axis = 0
         return CommSpec(comm_pattern=communication_pattern,
                         sharding_spec=sharding_spec,
                         logical_process_axis=logical_process_axis)
@@ -150,3 +148,17 @@ class StrategyGenerator_V2(ABC):
         If True, means this generator can be used for the current operation.
         """
         pass
+
+
+class FollowingStrategyGenerator(StrategyGenerator_V2):
+    """
+    FollowingStrategyGenerator is used to generate the sharding strategies which depends on its predecessor node. 
+
+    TODO: remove the original strategy_generator.py after refactoring
+    """
+
+    def __init__(self, operation_data_mapping: Dict[str, OperationData], device_mesh: DeviceMesh,
+                 predecessor_node: Node):
+        self.op_data = operation_data_mapping
+        self.device_mesh = device_mesh
+        self.predecessor_node = predecessor_node

tests/test_auto_parallel/test_node_handler/test_conv_handler_v2.py

@@ -165,7 +165,7 @@ def test_conv_function_handler():
     assert mapping['output'].data.shape == torch.Size([4, 16, 64, 64])
     assert mapping['output'].type == OperationDataType.OUTPUT
 
-    strategies_vector = handler.register_strategy()
+    handler.register_strategy()
     strategy_name_list = [val.name for val in strategies_vector]
 
     # SS = SR x RS

tests/test_auto_parallel/test_node_handler/test_getitem_handler.py

@@ -0,0 +1,85 @@
+from colossalai.fx.tracer.meta_patch.patched_module import linear
+import torch
+import torch.nn as nn
+from colossalai.fx import ColoTracer, ColoGraphModule
+from colossalai.auto_parallel.solver.op_handler.getitem_handler import GetItemHandler
+from colossalai.auto_parallel.solver.op_handler.conv_handler_v2 import ConvFunctionHandler
+from colossalai.auto_parallel.solver.sharding_strategy import OperationData, OperationDataType, StrategiesVector
+from colossalai.device.device_mesh import DeviceMesh
+
+
+class GetItemModel(nn.Module):
+
+    def __init__(self):
+        super().__init__()
+
+    def forward(self, input, other):
+        conv_node = nn.functional.conv2d(input, other)
+        x = conv_node[1]
+        return x
+
+
+def test_getitem_function_handler():
+    model = GetItemModel()
+    tracer = ColoTracer()
+    # graph():
+    #     %input_1 : torch.Tensor [#users=1] = placeholder[target=input]
+    #     %other : torch.Tensor [#users=1] = placeholder[target=other]
+    #     %conv2d : [#users=1] = call_function[target=torch.conv2d](args = (%input_1, %other), kwargs = {})
+    #     %getitem : [#users=1] = call_function[target=operator.getitem](args = (%conv2d, 1), kwargs = {})
+    #     return getitem
+    graph = tracer.trace(model,
+                         meta_args={
+                             "input": torch.rand(4, 4, 64, 64).to('meta'),
+                             "other": torch.rand(4, 16, 3, 3).to('meta'),
+                         })
+    gm = ColoGraphModule(model, graph)
+    physical_mesh_id = torch.arange(0, 4)
+
+    mesh_shape = (2, 2)
+    device_mesh = DeviceMesh(physical_mesh_id, mesh_shape)
+    conv_mod_node = list(graph.nodes)[2]
+    getitem_mod_node = list(graph.nodes)[3]
+    getitem_strategies_vector = StrategiesVector(getitem_mod_node)
+    conv_strategies_vector = StrategiesVector(conv_mod_node)
+
+    # build handler
+    conv_handler = ConvFunctionHandler(node=conv_mod_node,
+                                       device_mesh=device_mesh,
+                                       strategies_vector=conv_strategies_vector)
+    conv_handler.register_strategy()
+    setattr(conv_mod_node, 'strategies_vector', conv_strategies_vector)
+    getitem_handler = GetItemHandler(node=getitem_mod_node,
+                                     device_mesh=device_mesh,
+                                     strategies_vector=getitem_strategies_vector)
+
+    getitem_handler.register_strategy()
+    # check operation data mapping
+    mapping = getitem_handler.get_operation_data_mapping()
+
+    for name, op_data in mapping.items():
+        op_data: OperationData
+        # make sure they have valid values
+        assert op_data.data is not None
+
+    assert mapping['input'].name == "conv2d"
+    assert mapping['input'].data.is_meta
+    assert mapping['input'].data.shape == torch.Size([4, 4, 62, 62])
+    assert mapping['input'].type == OperationDataType.ARG
+    assert mapping['input'].logical_shape == torch.Size([4, 4, 62, 62])
+
+    assert mapping['index'].name == "index"
+    assert isinstance(mapping['index'].data, int)
+    assert mapping['index'].type == OperationDataType.ARG
+
+    assert mapping['output'].name == "getitem"
+    assert mapping['output'].data.is_meta
+    assert mapping['output'].data.shape == torch.Size([4, 62, 62])
+    assert mapping['output'].type == OperationDataType.OUTPUT
+
+    # getitem is a following strategy handler, so the number of strategies is equal to the predecessor node.
+    assert len(getitem_strategies_vector) == len(conv_strategies_vector)
+
+
+if __name__ == '__main__':
+    test_getitem_function_handler()