[autoparallel] added new strategy constructor template

colossalai/auto_parallel/solver/op_handler/__init__.py

@@ -6,8 +6,9 @@
 from .bcast_op_handler import BcastOpHandler
 from .embedding_handler import EmbeddingHandler
 from .unary_elementwise_handler import UnaryElementwiseHandler
+from .dot_handler_v2 import LinearFunctionHandler, LinearModuleHandler
 
 __all__ = [
     'OperatorHandler', 'DotHandler', 'ConvHandler', 'BatchNormHandler', 'ReshapeHandler', 'BcastOpHandler',
-    'UnaryElementwiseHandler', 'EmbeddingHandler'
+    'UnaryElementwiseHandler', 'EmbeddingHandler', 'LinearFunctionHandler', 'LinearModuleHandler'
 ]

colossalai/auto_parallel/solver/op_handler/registry.py

@@ -14,7 +14,7 @@ def wrapper(func):
         return wrapper
 
     def get(self, source):
-        assert source in self.store
+        assert source in self.store, f'{source} not found in the {self.name} registry'
         target = self.store[source]
         return target
 

colossalai/auto_parallel/solver/sharding_strategy.py

@@ -49,9 +49,10 @@ class OperationDataType(Enum):
     """
     An operation can come from the argument list of an operator or the parameter list of a module.
     """
-    ARG = 0
-    PARAM = 1
-    OUTPUT = 2
+    INPUT = 0
+    ARG = 1
+    PARAM = 2
+    OUTPUT = 3
 
 
 @dataclass

colossalai/auto_parallel/solver/strategies_constructor.py

@@ -4,6 +4,7 @@
 from colossalai.tensor.sharding_spec import ShardingSpec
 from colossalai.device.device_mesh import DeviceMesh
 from colossalai.tensor.shape_consistency import ShapeConsistencyManager
+from colossalai.auto_parallel.solver.op_handler.registry import operator_registry
 from .options import SolverOptions
 from . import ShardingStrategy, StrategiesVector
 from .op_handler import *
@@ -16,6 +17,8 @@
 from ._utils import generate_sharding_spec, generate_resharding_costs
 import builtins
 
+__all__ = ['StrategiesConstructor', 'StrategiesConstructor_V2']
+
 
 class StrategiesConstructor:
     """
@@ -49,6 +52,10 @@ def remove_duplicated_strategy(self, strategies_vector):
                 name_checklist.append(strategy.name)
             else:
                 remove_list.append(strategy)
+
+        if remove_list:
+            print(f'remove list: {[item.name for item in remove_list]}')
+            print(f'keep list: {name_checklist}')
         for strategy in remove_list:
             strategies_vector.remove(strategy)
 
@@ -394,3 +401,91 @@ def build_strategies_and_cost(self):
             setattr(node, 'strategies_vector', strategies_vector)
             self.leaf_strategies.append(strategies_vector)
             self.strategy_map[node] = strategies_vector
+
+
+class StrategiesConstructor_V2:
+    """
+    StrategiesConstructor is used to construct the parallelization plan for the model execution.
+
+    Args:
+        graph (Graph): a Graph object used for analysis and strategy generation.
+        device_mesh (DeviceMesh): a DeviceMesh object which contains the meta information about the cluster.
+        solver_options (SolverOptions): a SolverOptions object which specifies the preferences for plan searching.
+    """
+
+    def __init__(self, graph: Graph, device_mesh: DeviceMesh, solver_options: SolverOptions):
+        self.graph = graph
+        assert graph.owning_module is not None, 'The given graph is not associated with a owning_module'
+        self.root_module = self.graph.owning_module
+        self.nodes = list(graph.nodes)
+        self.device_mesh = device_mesh
+        self.leaf_strategies = []
+        self.strategy_map = {}
+        self.solver_options = solver_options
+
+    def remove_duplicated_strategy(self, strategies_vector):
+        '''
+        In build_strategies_and_cost method, we may produce some duplicated strategies.
+        In this method, we will remove the duplicated strategies depending on the strategies name.
+        Note that this operation is in-place.
+        '''
+        name_checklist = []
+        remove_list = []
+        for strategy in strategies_vector:
+            if strategy.name not in name_checklist:
+                name_checklist.append(strategy.name)
+            else:
+                remove_list.append(strategy)
+        for strategy in remove_list:
+            strategies_vector.remove(strategy)
+
+    def build_strategies_and_cost(self):
+        """
+        This method is to build the strategy vector for each node in the computation graph.
+        """
+        for node in self.nodes:
+            strategies_vector = StrategiesVector(node)
+            input_nodes_len = 0
+            for check_node in strategies_vector.predecessor_nodes:
+                if isinstance(check_node._meta_data, torch.Tensor):
+                    input_nodes_len += 1
+            # input_nodes_len = len(strategies_vector.predecessor_nodes)
+            # placeholder node
+            if node.op == 'placeholder':
+                # TODO: implement placeholder node handler
+                pass
+
+            # get_attr node
+            elif node.op == 'get_attr':
+                # TODO: implement getattr node handler
+                pass
+
+            # call_module node
+            elif node.op == 'call_module':
+                target = node.target
+                submod = self.root_module.get_submodule(target)
+                submod_type = type(submod)
+                handler = operator_registry.get(submod_type)(node, self.device_mesh, strategies_vector)
+                handler.register_strategy()
+
+            # call_function node
+            elif node.op == 'call_function':
+                target = node.target
+                handler = operator_registry.get(target)(node, self.device_mesh, strategies_vector)
+                handler.register_strategy()
+
+            # call_method node
+            elif node.op == 'call_method':
+                method = getattr(node.args[0]._meta_data.__class__, node.target)
+                handler = operator_registry.get(method)(node, self.device_mesh, strategies_vector)
+                handler.register_strategy()
+
+            # output node
+            elif node.op == 'output':
+                # TODO: implement output node handler
+                pass
+
+            self.remove_duplicated_strategy(strategies_vector)
+            setattr(node, 'strategies_vector', strategies_vector)
+            self.leaf_strategies.append(strategies_vector)
+            self.strategy_map[node] = strategies_vector