feat(mge): add python graph for mgb graph editing

GitOrigin-RevId: 6a9d5beba2

imperative/python/megengine/core/tensor/utils.py

@@ -11,6 +11,7 @@ from typing import Iterable, Union
 
 import numpy as np
 
+from .._imperative_rt import VarNode
 from .._imperative_rt.core2 import Tensor, apply, dtype_promotion, get_device
 from ..ops import builtin
 from ..ops.special import Const
@@ -59,7 +60,7 @@ def astype(x, dtype):
 
 
 def convert_single_value(v, *, dtype=None, device=None):
-    if isinstance(v, Tensor):
+    if isinstance(v, (Tensor, VarNode)):
         if not is_quantize(v.dtype):
             v = astype(v, dtype)
     else:

imperative/python/megengine/functional/elemwise.py

@@ -12,11 +12,12 @@ import functools
 import numpy as np
 
 from ..core._imperative_rt.core2 import apply
+from ..core._imperative_rt.graph import VarNode
 from ..core.ops import builtin
 from ..core.ops.builtin import Elemwise
 from ..core.tensor import utils
 from ..core.tensor.array_method import _elwise_apply
-from ..core.tensor.utils import isscalar, setscalar
+from ..core.tensor.utils import astype, isscalar, setscalar
 from ..device import get_default_device
 from ..jit.tracing import is_tracing
 from ..tensor import Tensor
@@ -77,7 +78,7 @@ __all__ = [
 
 
 def _elwise(*args, mode):
-    tensor_args = list(filter(lambda x: isinstance(x, Tensor), args))
+    tensor_args = list(filter(lambda x: isinstance(x, (Tensor, VarNode)), args))
     if len(tensor_args) == 0:
         dtype = utils.dtype_promotion(args)
         first_arg = Tensor(args[0], dtype=dtype, device=get_default_device())
@@ -109,7 +110,7 @@ def _elwise(*args, mode):
             Elemwise.Mode.ROUND,
         ) and np.issubdtype(args[0].dtype, np.integer):
             return args[0]
-        args = tuple(map(lambda x: x.astype("float32"), args))
+        args = tuple(map(lambda x: astype(x, "float32"), args))
     return _elwise_apply(args, mode)
 
 

imperative/python/megengine/utils/comp_graph_tools.py

@@ -65,7 +65,6 @@ def get_owner_opr_inputs(var: VarNode) -> List[VarNode]:
     """
     Gets the inputs of owner opr of a variable.
     """
-    assert isinstance(var, VarNode)
     return var.owner.inputs
 
 
@@ -74,7 +73,6 @@ def get_owner_opr_type(var: VarNode) -> str:
     Gets the type of owner opr of a variable.
 
     """
-    assert isinstance(var, VarNode)
     return var.owner.type
 
 
@@ -109,7 +107,7 @@ def graph_traversal(outputs: VarNode):
     var2oprs = collections.defaultdict(list)
     opr2receivers = collections.defaultdict(list)
 
-    queue = list(map(lambda x: x.owner, outputs))
+    queue = list(set(map(lambda x: x.owner, outputs)))
     visited = set(map(lambda x: x.id, queue))
 
     # iterate through whole comp_graph, fill in meta information
@@ -143,12 +141,15 @@ def graph_traversal(outputs: VarNode):
     return map_oprs, map_vars, var2oprs, opr2receivers, indegree2opr, opr2indegree
 
 
-def get_oprs_seq(outputs: List[VarNode], prune_reshape=False) -> List[OperatorNode]:
+def get_oprs_seq(
+    outputs: List[VarNode], prune_reshape=False, prune_immtensor=True
+) -> List[OperatorNode]:
     """
     Gets oprs in some topological order for a dumped model.
 
     :param outputs: model outputs.
-    :param prune_reshape: whether to prune the useless operators during inference.
+    :param prune_reshape: whether to prune the useless operators used by Reshape opr during inference.
+    :param prune_immtensor: whether to prune the ImmutableTensor opr.
     :return: opr list with some correct execution order.
     """
 
@@ -160,9 +161,7 @@ def get_oprs_seq(outputs: List[VarNode], prune_reshape=False) -> List[OperatorNo
             opr_id = indegree2opr[0].pop()
             opr = map_oprs[opr_id]
             nr_remain -= 1
-
-            # skip const value generation operator
-            if get_opr_type(opr) != "ImmutableTensor":
+            if opr.type != "ImmutableTensor" or not prune_immtensor:
                 oprs_seq.append(opr)
 
             for post_id in opr2receivers[opr_id]:

imperative/python/megengine/utils/network.py

@@ -0,0 +1,682 @@
+# -*- coding: utf-8 -*-
+# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
+#
+# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+import collections
+import fnmatch
+import itertools
+import re
+from collections import OrderedDict
+from typing import Dict, List
+
+import numpy as np
+
+from ..core._imperative_rt import ComputingGraph
+from ..core.tensor import megbrain_graph as G
+from .comp_graph_tools import get_dep_vars, get_opr_type, get_oprs_seq
+from .network_node import (
+    NetworkNode,
+    Host2DeviceCopy,
+    ImmutableTensor,
+    OpNode,
+    VarNode,
+    str_to_mge_class,
+)
+
+
+class Network:
+    def __init__(self):
+        self.input_vars = []  # input var of graph
+        self._orig_inputs = []
+        self.output_vars = []  # output var of graph
+        self._orig_outputs = []
+        self.all_oprs_map = OrderedDict()
+        self.all_vars_map = OrderedDict()
+        self.graph = ComputingGraph()
+
+    @classmethod
+    def load(cls, model_path: str, outspec: List[str] = None):
+        """
+        Loads a computing graph as a Network object.
+        :param model_path: file path of mge model.
+        :param outspec: only load the subgraph with outspec as its endpoints.
+        """
+        self = cls()
+        _, _, outputs = G.load_graph(model_path)
+        if outspec is not None:
+            output_spec = outspec.copy()
+            all_vars = get_dep_vars(outputs) + outputs
+            new_outputs = {}
+            for i in all_vars:
+                if i.name in output_spec:
+                    new_outputs[i.name] = i
+                    output_spec.remove(i.name)
+            assert len(output_spec) == 0, "Can not find {} in this model".format(
+                output_spec
+            )
+            outputs = [new_outputs[i] for i in outspec]
+        self._orig_outputs = outputs
+        self.add_dep_oprs(*outputs)
+        for x in self._orig_inputs:
+            self.input_vars.append(self._get_var(x))
+
+        for x in self._orig_outputs:
+            self.output_vars.append(self._get_var(x))
+        self.graph = self._orig_outputs[0].graph
+        return self
+
+    def _compile(self):
+        self.all_oprs_map = {}
+        self.all_vars_map = {}
+        for opr in self.all_oprs:
+            if isinstance(opr, (ImmutableTensor, Host2DeviceCopy)):
+                opr.compile(self.graph)
+            else:
+                opr.compile()
+            if opr.name is not None:
+                opr._opr.name = opr.name
+            self.all_oprs_map[opr._opr.id] = opr
+            for o in opr.outputs:
+                self.all_vars_map[o.var.id] = o
+
+    def dump(
+        self,
+        file,
+        *,
+        keep_var_name: int = 1,
+        keep_opr_name: bool = False,
+        keep_param_name: bool = False,
+        keep_opr_priority: bool = False,
+        strip_info_file=None,
+        append_json=False,
+        optimize_for_inference=True,
+        append=False,
+        **kwargs
+    ):
+        """
+        Serializes graph to file.
+
+        :param file: output file, could be file object or filename.
+        :param append: whether output is appended to ``file``.
+            Only works when ``file`` is str.
+        :param keep_var_name: level for keeping variable names:
+
+            * 0: none of the names are kept
+            * 1: (default)keep names of output vars
+            * 2: keep names of all (output and internal) vars
+        :param keep_opr_name: whether to keep operator names.
+        :param keep_param_name: whether to keep param names, so param values can be
+            easily manipulated after loading model
+        :param keep_opr_priority: whether to keep priority setting for operators
+        :param strip_info_file: a string for path or a file handler. if is not None,
+            then the dump information for code strip would be written to ``strip_info_file``
+        :param append_json: will be check when `strip_info_file` is not None. if set
+            true, the information for code strip will be append to strip_info_file.
+            if set false, will rewrite strip_info_file
+        :param optimize_for_inference: enbale optmizations,
+            will skip all optimize options if this is False. Default: True
+
+        :Keyword Arguments:
+
+            * enable_io16xc32 --
+                whether to use float16 for I/O between oprs and use
+                float32 as internal computation precision. Note the output var would be
+                changed to float16.
+            * enable_ioc16 --
+                whether to use float16 for both I/O and computation
+                precision.
+
+            * enable_hwcd4 --
+                whether to use NHWCD4 data layout. This is faster on some
+                OpenCL backend.
+            * enable_nchw88 --
+                whether to use NCHW88 data layout, currently
+                used in X86 AVX backend.
+            * enable_nchw44 --
+                whether to use NCHW44 data layout, currently
+                used in arm backend.
+            * enable_nchw44_dot --
+                whether to use NCHW44_dot data layout, currently
+                used in armv8.2+dotprod backend.
+            * enable_nchw4 --
+                whether to use NCHW4 data layout, currently
+                used in nvidia backend(based on cudnn).
+            * enable_nchw32 --
+                whether to use NCHW32 data layout, currently
+                used in nvidia backend with tensorcore(based on cudnn).
+            * enable_chwn4 --
+                whether to use CHWN4 data layout, currently
+                used in nvidia backend with tensorcore.
+
+            * enable_fuse_conv_bias_nonlinearity: whether to fuse conv+bias+nonlinearty
+                into one opr.
+            * enable_fuse_conv_bias_with_z: whether to fuse conv_bias with z
+                input for inference on nvidia backend(this optimization pass will
+                result in mismatch of the precision of output of training and
+                inference)
+        """
+
+        self._compile()
+        out = [G.VarNode(var.var) for var in self.output_vars]
+
+        if optimize_for_inference:
+            out = G.optimize_for_inference(out, **kwargs)
+
+        dump_content, _ = G.dump_graph(
+            out,
+            keep_var_name=keep_var_name,
+            keep_opr_name=keep_opr_name,
+            keep_param_name=keep_param_name,
+            keep_opr_priority=keep_opr_priority,
+            strip_info_file=strip_info_file,
+            append_json=append_json,
+        )
+        if isinstance(file, str):
+            permission = "wb" if append == False else "ab"
+            file = open(file, permission)
+        file.write(dump_content)
+
+    def make_const(self, data, name=None, device=None):
+        """Makes an ImmutableTensor OpNode to provide a parameter for the network.
+        """
+        node = ImmutableTensor(data, name, device, self.graph)
+        node.compile(self.graph)
+        return node.outputs[0]
+
+    def make_input_node(self, shape, dtype, name=None, device=None):
+        """Makes a Host2DeviceCopy OpNode to provide an input varnode for the network.
+        """
+        node = Host2DeviceCopy(shape, dtype, name, device)
+        node.compile(self.graph)
+        return node.outputs[0]
+
+    def add_output(self, *vars: VarNode):
+        """Adds vars into the network output node list
+        """
+        for var in vars:
+            if var not in self.output_vars:
+                self.output_vars.append(var)
+
+    def remove_output(self, *vars: VarNode):
+        """Removes vars from the network output node list.
+        """
+        for var in vars:
+            if var in self.output_vars:
+                self.output_vars.remove(var)
+
+    def add_dep_oprs(self, *vars):
+        """Adds dependent opnodes and varnodes of vars into network
+        """
+        oprs = get_oprs_seq(vars, False, False)
+        for mge_opr in oprs:
+            if get_opr_type(mge_opr) == "Host2DeviceCopy":
+                self._orig_inputs.extend(mge_opr.outputs)
+            opr = self._add_opr(mge_opr)
+            if opr is not None:
+                for x in mge_opr.inputs:
+                    opr.add_inp_var(self._get_var(x))
+                # set out var
+                for x in mge_opr.outputs:
+                    opr.add_out_var(self._get_var(x))
+
+        return [self.all_vars_map[var.id] for var in vars]
+
+    def modify_opr_names(self, modifier):
+        """Modifies names of operators **inplace**; useful for merging loaded
+        network into another network
+
+        :param modifier: a string to be prepended to the name, or a function
+            that maps from name to name
+        :type modifier: str or callable
+        """
+        if isinstance(modifier, str):
+            om = modifier
+            modifier = lambda v: "{}.{}".format(om, v)
+        assert isinstance(modifier, collections.Callable)
+        for i in self.all_oprs:
+            v0 = i.name
+            v1 = modifier(v0)
+            assert isinstance(v1, str)
+            i.name = v1
+
+    def reset_batch_size(self, batchsize, *, blacklist=()):
+        """Helper for reset batch size; first dimension of all data providers
+        not in blacklist are assumed to be the batch size
+
+        :param blacklist: data provider names whose first dimension is not
+            batchbatch size
+        """
+        blacklist = set(blacklist)
+        prev_batchsize = None
+        for i in self.data_providers_filter:
+            if i.name in blacklist:
+                blacklist.remove(i.name)
+            else:
+                shp = list(i.shape)
+                if prev_batchsize is None:
+                    prev_batchsize = shp[0]
+                else:
+                    assert prev_batchsize == shp[0], (
+                        "batchsize mismatch: batchsize={} "
+                        "shape={} dp={}".format(prev_batchsize, shp, i.name)
+                    )
+                shp[0] = batchsize
+                i.shape = tuple(shp)
+
+        assert prev_batchsize is not None, "no data provider found"
+        assert not blacklist, "unused items in blacklist: {}".format(blacklist)
+
+    def replace_vars(self, repl_dict: Dict[VarNode, VarNode]):
+        """
+        Replaces vars in the graph.
+        :param repl_dict: the map {old_var: new_var} that specifies how to replace the vars.
+        """
+        for var in self.all_vars:
+            if var in repl_dict:
+                repl_var = repl_dict[var]
+                owner = repl_var.owner
+                idx = owner.outputs.index(repl_var)
+                owner.outputs[idx] = var
+                var.__dict__.update(repl_var.__dict__)
+
+    def replace_oprs(self, repl_dict: Dict[OpNode, OpNode]):
+        """
+        Replaces operators in the graph.
+        :param oprmap: the map {old_opr: new_opr} that specifies how to replace the operators.
+        """
+        for opr in self.all_oprs:
+            if opr in repl_dict:
+                assert len(opr.outputs) == len(
+                    repl_dict[opr].outputs
+                ), "can not replace {} with {}".format(type(opr), type(repl_dict[opr]))
+                repl_dict[opr].outputs = opr.outputs
+                for ind, var in enumerate(opr.outputs):
+                    var.owner = repl_dict[opr]
+                    var.__dict__.update(repl_dict[opr].outputs[ind].__dict__)
+
+    def get_opr_by_type(self, oprcls, unique=True):
+        assert issubclass(oprcls, OpNode)
+        rst = self.opr_filter.type(oprcls).as_list()
+        if unique:
+            assert len(rst) == 1, "{} operators of type {} found".format(
+                len(rst), oprcls
+            )
+            (rst,) = rst
+        return rst
+
+    def get_opr_by_name(self, name, unique=True):
+        rst = self.opr_filter.name(name).as_list()
+        if unique:
+            assert len(rst) == 1, "{} operators of type {} found".format(len(rst), name)
+            (rst,) = rst
+        return rst
+
+    def get_var_by_name(self, name, unique=True):
+        rst = self.var_filter.name(name).as_list()
+        if unique:
+            assert len(rst) == 1, "{} operators of type {} found".format(len(rst), name)
+            (rst,) = rst
+        return rst
+
+    def get_var_receive_oprs(self, var):
+        """ Gets all oprs which use var as input
+        """
+        return self.opr_filter.has_input(var).as_list()
+
+    def get_dep_oprs(self, var):
+        """Gets dependent oprs of var
+        """
+        return get_oprs_seq(var, False, False)
+
+    @property
+    def opr_filter(self):
+        """Filter on all opnodes of the Network.
+        """
+        oprs = self.all_oprs
+        return NodeFilter(itertools.islice(oprs, len(oprs)))
+
+    @property
+    def var_filter(self):
+        """Filter on all varnode of the Network.
+        """
+        vars = self.all_vars
+        return NodeFilter(itertools.islice(vars, len(vars)))
+
+    @property
+    def params_filter(self):  # all immutable tensor
+        """Filter on all parameters (ImmutableTensor Opr) of the Network
+        """
+        return self.opr_filter.param_provider()
+
+    @property
+    def data_providers_filter(self):  # all host2devicecopy
+        """Filter on all input nodes (Host2DeviceCopy Opr) of the Network
+        """
+        return self.opr_filter.data_provider()
+
+    @property
+    def dest_vars(self):
+        """Output varnodes of the Network.
+        """
+        return self.output_vars
+
+    @property
+    def all_oprs(self):
+        return get_oprs_seq(self.output_vars, False, False)
+
+    @property
+    def all_vars(self):
+        return get_dep_vars(self.output_vars)
+
+    @property
+    def all_vars_dict(self):
+        return self.var_filter.as_dict()
+
+    @property
+    def all_oprs_dict(self):
+        return self.opr_filter.as_dict()
+
+    # used for loading and building graph
+    def _add_opr(self, x):
+        # TODO: use megbrain C++ RTTI to replace type string
+        if x.id not in self.all_oprs_map:
+            self.all_oprs_map[x.id] = str_to_mge_class(get_opr_type(x)).load(x)
+            return self.all_oprs_map[x.id]
+        else:
+            return None
+
+    def _get_opr(self, x):
+        if x.id in self.all_oprs_map:
+            return self.all_oprs_map[x.id]
+        else:
+            return None
+
+    def _get_var(self, x):
+        # auto convert to VarNode of Network
+        if x.id not in self.all_vars_map:
+            self.all_vars_map[x.id] = VarNode.load(x, self._get_opr(x.owner))
+        return self.all_vars_map[x.id]
+
+
+def as_varnode(obj):
+    """convert a :class:`.VarNode` compatible object to :class:`.VarNode`.
+
+    :param obj: it must be one of the following:
+
+        1. a :class:`.VarNode` object
+        2. a :class:`.OpNode` object that has unique output
+        3. an iterable that produces either type 1 or 2, with length 1
+
+    :rtype: :class:`.VarNode`
+    """
+    if type(obj) is VarNode:
+        return obj
+
+    if isinstance(obj, OpNode):
+        assert len(obj.outputs) == 1, (
+            "operator {} must have one output to be converted to VarNode; "
+            "got {} actually".format(obj, len(obj.outputs))
+        )
+        ret = obj.outputs[0]
+        assert type(ret) is VarNode
+        return ret
+
+    assert isinstance(
+        obj, collections.Iterable
+    ), "{} is not compatible with VarNode".format(obj)
+
+    val = list(obj)
+    assert (
+        len(val) == 1
+    ), "can not convert sequence of length {} to VarNode ({})".format(
+        len(val), (lambda s: s if len(s) < 50 else s[:50] + " ...")(str(val))
+    )
+    return as_varnode(val[0])
+
+
+def as_oprnode(obj):
+    """convert a :class:`.OpNode` compatible object to
+    :class:`.OpNode`; it works like :func:`as_varnode`."""
+    if type(obj) is VarNode:
+        return obj.owner
+
+    if isinstance(obj, OpNode):
+        return obj
+
+    assert isinstance(
+        obj, collections.Iterable
+    ), "{} is not compatible with OpNode".format(obj)
+
+    val = list(obj)
+    assert (
+        len(val) == 1
+    ), "can not convert sequence of length {} to " "OpNode({})".format(len(val), val)
+    return as_oprnode(val[0])
+
+
+class NodeFilter:
+    """Filter on node iterator. This class is an iterator of
+    :class:`.NetworkNode` objects and multiple filtering conditions and
+    mappers can be chained.
+
+    Example::
+
+        # find all :class:`.ImmutableTensor` nodes
+        for i in NodeFilter(node_iter).param_provider():
+            print(i)
+
+        # find all :class:`.ImmutableTensor` nodes that end with ':W'
+        for i in NodeFilter(node_iter).param_provider().name('*:W'):
+            print(i)
+
+        # number of inputs
+        nr_input = NodeFilter(node_iter).data_provider().as_count()
+
+    """
+
+    _iter = None
+
+    def __init__(self, node_iter):
+        """
+        :param node_iter: iterator to :class:`.NetworkNode`, or a
+            :class:`.VarNode`-compatible object; in the later case, its
+            dependent oprs would be used
+        """
+        if isinstance(node_iter, VarNode):
+            oprs = get_oprs_seq(node_iter, False, False)
+            node_iter = itertools.islice(oprs, len(oprs) - 1)
+        if isinstance(node_iter, OpNode):
+            oprs = get_oprs_seq(node_iter.inputs, False, False)
+            node_iter = itertools.islice(oprs, len(oprs) - 1)
+
+        assert isinstance(node_iter, collections.Iterable)
+        if (not isinstance(node_iter, NodeFilter)) and type(
+            self
+        ) is not NodeFilterCheckType:
+            node_iter = NodeFilterCheckType(node_iter, NetworkNode)
+        self._iter = node_iter
+
+    @classmethod
+    def make_all_deps(cls, *dest_vars):
+        """make a :class:`NodeFilter` that contains all deps of given vars"""
+        return cls(list(get_oprs_seq(dest_vars, False, False)))
+
+    def __iter__(self):
+        """to be overwritten by subclass to implement filters"""
+        return iter(self._iter)
+
+    def type(self, node_type):
+        """filter by specific node type
+
+        :param node_type: node type class
+        :return: a new :class:`NodeFilter` object
+        """
+        return NodeFilterType(self, node_type)
+
+    def check_type(self, node_type):
+        """assert that all oprs produced by this iterator are instances of
+        certain type
+
+        :param node_type: node type class
+        :return: a new :class:`NodeFilter` object
+        :raises TypeError: if type check failed
+        """
+        return NodeFilterCheckType(self, node_type)
+
+    def not_type(self, node_type):
+        """remove oprs of specific type
+
+        :param node_type: node type class
+        :return: a new :class:`NodeFilter` object
+        """
+        return NodeFilterNotType(self, node_type)
+
+    def param_provider(self):
+        """get :class:`.ParamProvider` oprs; shorthand for
+        ``.type(ParamProvider)``"""
+
+        return self.type(ImmutableTensor)
+
+    def data_provider(self):
+        """get :class:`.DataProvider` oprs; shorthand for
+        ``.type(DataProvider)``"""
+
+        return self.type(Host2DeviceCopy)
+
+    def name(self, pattern, ignorecase=True):
+        """filter by node name
+
+        :param pattern: a string in glob syntax that can contain ``?`` and
+            ``*`` to match a single or arbitrary characters.
+        :type pattern: :class:`str`
+        :param ignorecase: whether to ignroe case
+        :type ignorecase: bool
+        :return: a new :class:`NodeFilter` object
+        """
+        return NodeFilterName(self, pattern, ignorecase)
+
+    def has_input(self, var):
+        """an opr is kept if it has given var as one of its inputs
+
+        :param var: var node to checked
+        :return: a new :class:`NodeFilter` object
+        """
+        return NodeFilterHasInput(self, var)
+
+    def as_list(self):
+        """consume this iterator and return its content as a list
+
+        :rtype: [:class:`.GraphNodeBase`]
+        """
+        return list(self)
+
+    def as_unique(self):
+        """assert that this iterator yields only one node and return it
+
+        :return: the unique node
+        :rtype: :class:`.GraphNodeBase`
+        :raises ValueError: if this iterator does not yield a unique node
+        """
+        (opr,) = self
+        return opr
+
+    def as_dict(self):
+        """construct an ordered dict to map from node names to objects in
+        this iterator
+
+        :rtype: :class:`OrderedDict`
+        """
+        return collections.OrderedDict((i.name, i) for i in self)
+
+    def as_count(self):
+        """consume this iterator and get the number of elements
+
+        :rtype: int
+        """
+        return sum(1 for _ in self)
+
+
+class NodeFilterType(NodeFilter):
+    """see :meth:`NodeFilter.type`"""
+
+    _node_type = None
+
+    def __init__(self, node_iter, node_type):
+        assert issubclass(node_type, NetworkNode), "bad opr type: {}".format(
+            node_type
+        )
+        super().__init__(node_iter)
+        self._node_type = node_type
+
+    def __iter__(self):
+        for i in self._iter:
+            if isinstance(i, self._node_type):
+                yield i
+
+
+class NodeFilterNotType(NodeFilterType):
+    """see :meth:`NodeFilter.not_type`"""
+
+    def __iter__(self):
+        for i in self._iter:
+            if not isinstance(i, self._node_type):
+                yield i
+
+
+class NodeFilterCheckType(NodeFilterType):
+    """see :meth:`NodeFilter.check_type`"""
+
+    def __iter__(self):
+        for i in self._iter:
+            if not isinstance(i, self._node_type):
+                raise TypeError(
+                    "all nodes should be {}; got {!r}".format(self._node_type, i)
+                )
+            yield i
+
+
+class NodeFilterHasInput(NodeFilter):
+    """see :meth:`NodeFilter.has_input`"""
+
+    _var = None
+
+    def __init__(self, node_iter, var):
+        var = as_varnode(var)
+        super().__init__(node_iter)
+        self.var = var
+
+    def __iter__(self):
+        for i in self._iter:
+            assert isinstance(
+                i, OpNode
+            ), "has_input() must be used with OpNode; " "got {!r}".format(i)
+            if self.var in i.inputs:
+                yield i
+
+
+class NodeFilterName(NodeFilter):
+    """see :meth:`NodeFilter.name`"""
+
+    _re = None
+
+    def __init__(self, node_iter, pattern, ignorecase):
+        super().__init__(node_iter)
+        self._re = self.make_re(pattern, ignorecase)
+
+    @classmethod
+    def make_re(cls, pattern, ignorecase=True):
+        assert isinstance(pattern, str), "bad pattern: {!r}".format(pattern)
+        assert isinstance(ignorecase, bool)
+        flags = 0
+        if ignorecase:
+            flags |= re.IGNORECASE
+        return re.compile(fnmatch.translate(pattern), flags=flags)
+
+    def __iter__(self):
+        for i in self._iter:
+            if self._re.match(i.name):
+                yield i

imperative/python/megengine/utils/network_node.py

@@ -0,0 +1,628 @@
+# -*- coding: utf-8 -*-
+# MegEngine is Licensed under the Apache License, Version 2.0 (the "License")
+#
+# Copyright (c) 2014-2021 Megvii Inc. All rights reserved.
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT ARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+import json
+import sys
+from typing import Callable
+
+from ..core import _imperative_rt as rt
+from ..core._wrap import Device
+from ..core.ops import builtin
+from ..core.tensor.megbrain_graph import InputNode
+from ..tensor import Tensor
+from .comp_graph_tools import replace_vars
+
+
+class NetworkNode:
+    pass
+
+
+class VarNode(NetworkNode):
+    def __init__(self, owner_opr=None, name=None):
+        self.var = None
+        self.owner = owner_opr
+        self.name = name
+        self.id = id(self)
+
+    @classmethod
+    def load(cls, sym_var, owner_opr):
+        obj = cls()
+        obj.var = sym_var  # mgb varnode
+        obj.name = sym_var.name
+        obj.owner = owner_opr
+        return obj
+
+    @property
+    def shape(self):
+        rst = None
+        if self.var:
+            try:
+                rst = self.var.shape
+            except:
+                rst = None
+        return rst
+
+    @property
+    def dtype(self):
+        return self.var.dtype if self.var else None
+
+    def set_owner_opr(self, owner_opr):
+        self.owner_opr = owner_opr
+
+
+class OpNode(NetworkNode):
+
+    opdef = None
+    type = None
+
+    def __init__(self):
+        self.inputs = []
+        self.outputs = []
+        self.params = {}
+        self._opr = None  # mgb opnode
+        self.id = id(self)
+
+    @classmethod
+    def load(cls, opr):
+        obj = cls()
+        obj.params = json.loads(opr.params)
+        obj.name = opr.name
+        obj._opr = opr
+        return obj
+
+    def compile(self, graph=None):
+        op = self.opdef(**self.params)
+        args = [i.var for i in self.inputs]
+        outputs = rt.invoke_op(op, args)
+        assert len(outputs) == len(self.outputs)
+        self._opr = outputs[0].owner
+        for i in range(len(self.outputs)):
+            self.outputs[i].var = outputs[i]
+            self.outputs[i].var.name = self.outputs[i].name
+            assert self.outputs[i].owner is self
+
+    def add_inp_var(self, x):
+        self.inputs.append(x)
+
+    def add_out_var(self, x):
+        self.outputs.append(x)
+
+
+def str_to_mge_class(classname):
+    # TODO: use megbrain C++ RTTI to replace type string
+    if classname == "RNGOpr<MegDNNOpr>":
+        classname = "RNGOpr"
+    oprcls = getattr(sys.modules[__name__], classname, None)
+    return oprcls if oprcls else ReadOnlyOpNode
+
+
+class Host2DeviceCopy(OpNode):
+    type = "Host2DeviceCopy"
+
+    def __init__(self, shape=None, dtype=None, name=None, device=None):
+        super().__init__()
+        self.shape = shape
+        self.dtype = dtype
+        self.name = name
+        self.device = Device(device).to_c() if device else Device("xpux").to_c()
+        self.outputs = []
+
+    @classmethod
+    def load(cls, opr):
+        self = cls()
+        self.outputs = []
+        assert len(opr.outputs) == 1, "wrong number of outputs"
+        self.shape = opr.outputs[0].shape
+        self.dtype = opr.outputs[0].dtype
+        self.name = opr.outputs[0].name
+        self.device = opr.outputs[0].comp_node
+        self._opr = opr
+        return self
+
+    def compile(self, graph):
+        outputs = rt.make_h2d(graph, self.device, self.dtype, self.shape, self.name)
+        self._opr = outputs.owner
+        if len(self.outputs) == 0:
+            self.outputs.append(VarNode(self, self.name))
+        self.outputs[0].var = outputs
+        assert self.outputs[0].owner is self
+
+
+class ImmutableTensor(OpNode):
+    type = "ImmutableTensor"
+
+    def __init__(self, data=None, name=None, device=None, graph=None):
+        super().__init__()
+        self.name = name
+        self.outputs = []
+        self.graph = graph
+        if data is not None:
+            self.set_value(data, device)
+
+    @property
+    def device(self):
+        return self._opr.outputs[0].comp_node if self._opr else None
+
+    @device.setter
+    def device(self, device):
+        self.set_value(self.numpy(), device)
+
+    @property
+    def shape(self):
+        return self.outputs[0].shape
+
+    @property
+    def dtype(self):
+        return self._opr.outputs[0].dtype if self._opr else None
+
+    def numpy(self):
+        return self._opr.outputs[0].value if self._opr else None
+
+    def set_value(self, data, device=None):
+        assert self.graph is not None
+        cn = device if device else self.device
+        assert isinstance(data, (int, float, np.ndarray))
+        if isinstance(data, (int, float)):
+            data = np.array(data)
+        if data.dtype == np.float64:
+            data = data.astype(np.float32)
+        elif data.dtype == np.int64:
+            data = data.astype(np.int32)
+        varnode = rt.make_const(self.graph, data, cn, data.dtype, self.name)
+        if len(self.outputs) == 0:
+            self.outputs.append(VarNode(self, self.name))
+        self.outputs[0].var = varnode
+        self._opr = varnode.owner
+
+    @classmethod
+    def load(cls, opr):
+        self = cls()
+        self.outputs = []
+        self._opr = opr
+        self.name = opr.outputs[0].name
+        self.graph = opr.graph
+        return self
+
+    def compile(self, graph):
+        assert self.outputs[0].var is self._opr.outputs[0]
+        assert self.outputs[0].owner is self
+        if self.graph != graph:
+            self.graph = graph
+            self.set_value(self.numpy())
+        if self.name is not None:
+            self.outputs[0].var.name = self.name
+
+
+class ReadOnlyOpNode(OpNode):
+    @classmethod
+    def load(cls, opr):
+        obj = super(ReadOnlyOpNode, cls).load(opr)
+        obj.type = opr.type
+        return obj
+
+    def compile(self):
+        assert self._opr is not None
+        assert len(self.inputs) == len(self._opr.inputs)
+        assert len(self.outputs) == len(self._opr.outputs)
+        repl_dict = {}
+        for ind, i in enumerate(self.inputs):
+            if i.var != self._opr.inputs[ind]:
+                repl_dict[self._opr.inputs[ind]] = i.var
+        if bool(repl_dict):
+            out_vars = replace_vars(self._opr.outputs, repl_dict)
+            for ind, o in enumerate(self.outputs):
+                o.var = out_vars[ind]
+
+
+class Elemwise(OpNode):
+    type = "Elemwise"
+    opdef = builtin.Elemwise
+
+
+class Reduce(OpNode):
+    type = "Reduce"
+    opdef = builtin.Reduce
+
+
+class TypeCvt(OpNode):
+    type = "TypeCvt"
+    opdef = builtin.TypeCvt
+
+    @classmethod
+    def load(cls, opr):
+        obj = super(TypeCvt, cls).load(opr)
+        t_dtype = opr.outputs[0].dtype
+        obj.params["dtype"] = t_dtype
+        return obj
+
+
+class MatrixInverse(OpNode):
+    type = "MatrixInverse"
+    opdef = builtin.MatrixInverse
+
+
+class MatrixMul(OpNode):
+    type = "MatrixMul"
+    opdef = builtin.MatrixMul
+
+
+class BatchedMatrixMul(OpNode):
+    type = "BatchedMatmul"
+    opdef = builtin.BatchedMatrixMul
+
+
+class Dot(OpNode):
+    type = "Dot"
+    opdef = builtin.Dot
+
+
+class SVD(OpNode):
+    type = "SVD"
+    opdef = builtin.SVD
+
+
+class ConvolutionForward(OpNode):
+    type = "Convolution"
+    opdef = builtin.Convolution
+
+
+class ConvolutionBackwardData(OpNode):
+    type = "ConvTranspose"
+    opdef = builtin.ConvolutionBackwardData
+
+
+class DeformableConvForward(OpNode):
+    type = "DeformableConv"
+    opdef = builtin.DeformableConv
+
+
+class GroupLocalForward(OpNode):
+    type = "GroupLocal"
+    opdef = builtin.GroupLocal
+
+
+class PoolingForward(OpNode):
+    type = "Pooling"
+    opdef = builtin.Pooling
+
+
+class AdaptivePoolingForward(OpNode):
+    type = "AdaptivePooling"
+    opdef = builtin.AdaptivePooling
+
+
+class ROIPoolingForward(OpNode):
+    type = "ROIPooling"
+    opdef = builtin.ROIPooling
+
+
+class DeformablePSROIPoolingForward(OpNode):
+    type = "DeformablePSROIPooling"
+    opdef = builtin.DeformablePSROIPooling
+
+
+class ConvBiasForward(OpNode):
+    type = "ConvBias"
+    opdef = builtin.ConvBias
+
+    @classmethod
+    def load(cls, opr):
+        obj = super(ConvBiasForward, cls).load(opr)
+        obj.params["dtype"] = opr.outputs[0].dtype
+        return obj
+
+
+class BatchConvBiasForward(OpNode):
+    type = "BatchConvBias"
+    opdef = builtin.BatchConvBias
+
+    @classmethod
+    def load(cls, opr):
+        obj = super(BatchConvBiasForward, cls).load(opr)
+        obj.params["dtype"] = opr.outputs[0].dtype
+        return obj
+
+
+class BatchNormForward(OpNode):
+    type = "BatchNorm"
+    opdef = builtin.BatchNorm
+
+
+class ROIAlignForward(OpNode):
+    type = "ROIAlign"
+    opdef = builtin.ROIAlign
+
+
+class WarpPerspectiveForward(OpNode):
+    type = "WarpPerspective"
+    opdef = builtin.WarpPerspective
+
+
+class WarpAffineForward(OpNode):
+    type = "WarpAffine"
+    opdef = builtin.WarpAffine
+
+
+class RemapForward(OpNode):
+    type = "Remap"
+    opdef = builtin.Remap
+
+
+class ResizeForward(OpNode):
+    type = "Resize"
+    opdef = builtin.Resize
+
+
+class IndexingOneHot(OpNode):
+    type = "IndexingOneHot"
+    opdef = builtin.IndexingOneHot
+
+
+class IndexingSetOneHot(OpNode):
+    type = "IndexingSetOneHot"
+    opdef = builtin.IndexingSetOneHot
+
+
+class Copy(OpNode):
+    type = "Copy"
+    opdef = builtin.Copy
+
+    @classmethod
+    def load(cls, opr):
+        obj = super(Copy, cls).load(opr)
+        obj.params["comp_node"] = opr.outputs[0].comp_node
+        return obj
+
+
+class ArgsortForward(OpNode):
+    type = "Argsort"
+    opdef = builtin.Argsort
+
+
+class Argmax(OpNode):
+    type = "Argmax"
+    opdef = builtin.Argmax
+
+
+class Argmin(OpNode):
+    type = "Argmin"
+    opdef = builtin.Argmin
+
+
+class CondTake(OpNode):
+    type = "CondTake"
+    opdef = builtin.CondTake
+
+
+class TopK(OpNode):
+    type = "TopK"
+    opdef = builtin.TopK
+
+
+class NvOf(OpNode):
+    type = "NvOf"
+    opdef = builtin.NvOf
+
+
+class RNGOpr(OpNode):
+    @classmethod
+    def load(cls, opr):
+        obj = super(RNGOpr, cls).load(opr)
+        if len(obj.params) == 3:
+            obj.opdef = builtin.GaussianRNG
+            obj.type = "GaussianRNG"
+        else:
+            obj.opdef = builtin.UniformRNG
+            obj.type = "UniformRNG"
+        return obj
+
+
+class Linspace(OpNode):
+    type = "Linspace"
+    opdef = builtin.Linspace
+
+    @classmethod
+    def load(cls, opr):
+        obj = super(Linspace, cls).load(opr)
+        obj.params["comp_node"] = opr.outputs[0].comp_node
+        return obj
+
+
+class Eye(OpNode):
+    type = "Eye"
+    opdef = builtin.Eye
+
+    @classmethod
+    def load(cls, opr):
+        obj = super(Eye, cls).load(opr)
+        obj.params["dtype"] = opr.outputs[0].dtype
+        obj.params["comp_node"] = opr.outputs[0].comp_node
+        return obj
+
+
+class GetVarShape(OpNode):
+    type = "GetVarShape"
+    opdef = builtin.GetVarShape
+
+
+class Concat(OpNode):
+    type = "Concat"
+    opdef = builtin.Concat
+
+    @classmethod
+    def load(cls, opr):
+        obj = super(Concat, cls).load(opr)
+        obj.params["comp_node"] = Device("xpux").to_c()
+        return obj
+
+
+class Broadcast(OpNode):
+    type = "Broadcast"
+    opdef = builtin.Broadcast
+
+
+class Identity(OpNode):
+    type = "Identity"
+    opdef = builtin.Identity
+
+
+class NMSKeep(OpNode):
+    type = "NMSKeep"
+    opdef = builtin.NMSKeep
+
+
+# class ParamPackSplit
+# class ParamPackConcat
+
+
+class Dimshuffle(OpNode):
+    type = "Dimshuffle"
+    opdef = builtin.Dimshuffle
+
+    @classmethod
+    def load(cls, opr):
+        obj = super(Dimshuffle, cls).load(opr)
+        del obj.params["ndim"]
+        return obj
+
+
+class Reshape(OpNode):
+    type = "Reshape"
+    opdef = builtin.Reshape
+
+
+class AxisAddRemove(OpNode):
+    type = "AxisAddRemove"
+
+    @classmethod
+    def load(cls, opr):
+        obj = cls()
+        obj.name = opr.name
+        obj._opr = opr
+        params = json.loads(opr.params)
+        desc = params["desc"]
+        method = None
+        axis = []
+        for i in desc:
+            if method is None:
+                method = i["method"]
+            assert method == i["method"]
+            axis.append(i["axisnum"])
+        obj.params = {"axis": axis}
+        obj.opdef = builtin.AddAxis if desc[0]["method"] == 0 else builtin.RemoveAxis
+        return obj
+
+
+class IndexingBase(OpNode):
+    @classmethod
+    def load(cls, opr):
+        obj = cls()
+        obj.name = opr.name
+        obj._opr = opr
+        params = json.loads(opr.params)
+        items = [
+            [
+                p["axis"],
+                bool(p["begin"]),
+                bool(p["end"]),
+                bool(p["step"]),
+                bool(p["idx"]),
+            ]
+            for p in params
+        ]
+        obj.params["items"] = items
+        return obj
+
+
+class Subtensor(IndexingBase):
+    type = "Subtensor"
+    opdef = builtin.Subtensor
+
+
+class SetSubtensor(IndexingBase):
+    type = "SetSubtensor"
+    opdef = builtin.SetSubtensor
+
+
+class IncrSubtensor(IndexingBase):
+    type = "IncrSubtensor"
+    opdef = builtin.IncrSubtensor
+
+
+class IndexingMultiAxisVec(IndexingBase):
+    type = "IndexingMultiAxisVec"
+    opdef = builtin.IndexingMultiAxisVec
+
+
+class IndexingSetMultiAxisVec(IndexingBase):
+    type = "IndexingSetMultiAxisVec"
+    opdef = builtin.IndexingSetMultiAxisVec
+
+
+class IndexingIncrMultiAxisVec(IndexingBase):
+    type = "IndexingIncrMultiAxisVec"
+    opdef = builtin.IndexingIncrMultiAxisVec
+
+
+class MeshIndexing(IndexingBase):
+    type = "MeshIndexing"
+    opdef = builtin.MeshIndexing
+
+
+class SetMeshIndexing(IndexingBase):
+    type = "SetMeshIndexing"
+    opdef = builtin.SetMeshIndexing
+
+
+class IncrMeshIndexing(IndexingBase):
+    type = "IncrMeshIndexing"
+    opdef = builtin.IncrMeshIndexing
+
+
+class BatchedMeshIndexing(IndexingBase):
+    type = "BatchedMeshIndexing"
+    opdef = builtin.BatchedMeshIndexing
+
+
+class BatchedSetMeshIndexing(IndexingBase):
+    type = "BatchedSetMeshIndexing"
+    opdef = builtin.BatchedSetMeshIndexing
+
+
+class BatchedIncrMeshIndexing(IndexingBase):
+    type = "BatchedIncrMeshIndexing"
+    opdef = builtin.BatchedIncrMeshIndexing
+
+
+# class CollectiveComm
+# class RemoteSend
+# class RemoteRecv
+# class TQT
+# class FakeQuant
+# class InplaceAdd
+
+
+class AssertEqual(OpNode):
+    type = "AssertEqual"
+    opdef = builtin.AssertEqual
+
+
+class ElemwiseMultiType(OpNode):
+    type = "ElemwiseMultiType"
+    opdef = builtin.ElemwiseMultiType
+
+    @classmethod
+    def load(cls, opr):
+        obj = super(ElemwiseMultiType, cls).load(opr)
+        obj.params["dtype"] = opr.outputs[0].dtype
+        return obj
+
+
+class CvtColorForward(OpNode):
+    type = "CvtColor"
+    opdef = builtin.CvtColor

imperative/python/src/tensor.cpp

@@ -160,6 +160,16 @@ PyObject* py_apply(PyObject* self, PyObject*const* args, size_t nargs/* , PyObje
         if (ctx.op->same_type<BackwardGraph>()) {
             ctx.backward = true;
         }
+        
+        
+       if (py::isinstance<cg::VarNode>(py::handle(args[0]))){
+           SmallVector<cg::VarNode*> vinputs(nargs);
+           for (size_t i = 0; i < nargs; ++i) {
+               vinputs[i] = py::handle(args[i]).cast<cg::VarNode *>();
+           }
+           auto op = ctx.op.get();
+           return to_tuple(OpDef::apply_on_var_node(*op, vinputs)).release().ptr();
+       }
 
         for (size_t i = 0; i < nargs; ++i) {
             if (TensorWrapper* tw = TensorWrapper::try_cast(args[i])) {
@@ -675,6 +685,16 @@ PyArray_Descr* _dtype_promotion(PyObject*const* args, size_t nargs) {
                 tensors.emplace_back(descr);
                 continue;
             }
+
+            if (py::isinstance<cg::VarNode>(py::handle(handle))){
+                auto var = py::handle(handle).cast<cg::VarNode *>();
+                mgb::DType type = var->dtype();
+                auto && descr = npy::dtype_mgb2np_descr(type);
+                Py_INCREF(descr.get());
+                tensors.emplace_back(descr.get());
+                continue;
+            }
+
             PyArray_Descr* descr = scalar2dtype(handle);
             if (descr) {
                 scalars.emplace_back(descr);
@@ -719,12 +739,14 @@ CompNode _get_device(PyObject*const* args, size_t nargs) {
     for (size_t i = 0; i < nargs; ++i) {
         PyObject* handle = is_tuple ? PyTuple_GetItem(tuple, i): args[i];
         TensorWrapper* tw = TensorWrapper::try_cast(handle);
-        if (tw) {
+
+        bool is_var = py::isinstance<cg::VarNode>(py::handle(handle));
+        if (tw || is_var) {
             if (!valid) {
-                cn = tw->m_tensor->comp_node();
+                    cn = tw ? tw->m_tensor->comp_node() : py::handle(handle).cast<cg::VarNode *>()->comp_node();
                 valid = true;
             } else {
-                CompNode cn1 = tw->m_tensor->comp_node();
+                CompNode cn1 = tw ? tw->m_tensor->comp_node() : py::handle(handle).cast<cg::VarNode *>()->comp_node();
                 if (cn1 != cn) {
                     throw py::value_error(ssprintf("ambiguous device: %s vs %s",
                         cn.to_string().c_str(), cn1.to_string().c_str()));

imperative/python/test/unit/utils/test_network.py

@@ -0,0 +1,351 @@
+import io
+
+import numpy as np
+
+import megengine.core.tensor.megbrain_graph as G
+import megengine.functional as F
+import megengine.module as M
+import megengine.utils.network_node as N
+from megengine.jit.tracing import trace
+from megengine.tensor import Tensor
+from megengine.utils.comp_graph_tools import GraphInference
+from megengine.utils.network import Network as Net
+from megengine.utils.network import as_oprnode
+from megengine.utils.network_node import Host2DeviceCopy, VarNode
+
+
+def test_replace_var():
+
+    a = Tensor([1, 2])
+    b = Tensor([3, 4])
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(a, b):
+        return (a + b) * 2
+
+    fwd(a, b)
+    orig_model = io.BytesIO()
+    fwd.dump(
+        orig_model, arg_names=["a", "b"], output_names="o", optimize_for_inference=False
+    )
+    orig_model.seek(0)
+
+    graph = Net.load(orig_model)
+    vara = graph.var_filter.name("a").as_unique()
+    varb = graph.var_filter.name("b").as_unique()
+
+    out = F.mul(vara.var, varb.var)
+    out = F.relu(out)
+
+    var_list = graph.add_dep_oprs(out)
+
+    opnode = list(graph.opr_filter.has_input(vara))
+    repl_dict = {opnode[0].outputs[0]: var_list[0]}
+    graph.replace_vars(repl_dict)
+
+    modified_model = io.BytesIO()
+    graph.dump(modified_model)
+    modified_model.seek(0)
+    load_graph = GraphInference(modified_model)
+
+    out = load_graph.run(a, b)
+    np.testing.assert_equal(out["o"], [6, 16])
+
+
+def test_replace_opr():
+
+    a = Tensor([1, 2])
+    b = Tensor([3, 4])
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(a, b):
+        return (a + b) * 2
+
+    fwd(a, b)
+    orig_model = io.BytesIO()
+    fwd.dump(
+        orig_model, arg_names=["a", "b"], output_names="o", optimize_for_inference=False
+    )
+    orig_model.seek(0)
+
+    graph = Net.load(orig_model)
+    vara = graph.var_filter.name("a").as_unique()
+    varb = graph.var_filter.name("b").as_unique()
+
+    out1 = F.sub(vara.var, varb.var)
+    out1 = F.relu(out1)
+
+    var_list = graph.add_dep_oprs(out1)
+    repl_opr = as_oprnode(var_list)
+    orig_opr = graph.opr_filter.has_input(vara).as_unique()
+
+    repl_dict = {orig_opr: repl_opr}
+    graph.replace_oprs(repl_dict)
+    modified_model1 = io.BytesIO()
+    graph.dump(modified_model1)
+    modified_model1.seek(0)
+
+    load_graph = GraphInference(modified_model1)
+    out = load_graph.run(a, b)
+    np.testing.assert_equal(out["o"], [0, 0])
+
+
+def test_modify_params():
+
+    a = Tensor([1, 2])
+    b = Tensor([3, 4])
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(a, b):
+        return (a + b) * 2
+
+    fwd(a, b)
+    orig_model = io.BytesIO()
+    fwd.dump(
+        orig_model, arg_names=["a", "b"], output_names="o", optimize_for_inference=False
+    )
+    orig_model.seek(0)
+
+    graph = Net.load(orig_model)
+    param_const = graph.params_filter.as_unique()
+    param_const.set_value(3)
+
+    modified_model = io.BytesIO()
+    graph.dump(modified_model)
+    modified_model.seek(0)
+    load_graph = GraphInference(modified_model)
+
+    out = load_graph.run(a, b)
+    np.testing.assert_equal(out["o"], [12, 18])
+
+
+def test_make_const():
+
+    a = Tensor([1, 2])
+    b = Tensor([3, 4])
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(a, b):
+        return (a + b) * 2
+
+    fwd(a, b)
+    orig_model = io.BytesIO()
+    fwd.dump(
+        orig_model, arg_names=["a", "b"], output_names="o", optimize_for_inference=False
+    )
+    orig_model.seek(0)
+
+    graph = Net.load(orig_model)
+    const_b = graph.make_const(np.array([0.0, 0.0]), name="b")
+    varb = graph.var_filter.name("b").as_unique()
+
+    repl_dict = {varb: const_b}
+    graph.replace_vars(repl_dict)
+
+    modified_model = io.BytesIO()
+    graph.dump(modified_model)
+    modified_model.seek(0)
+    load_graph = GraphInference(modified_model)
+
+    out = load_graph.run(a)
+    np.testing.assert_equal(out["o"], [2, 4])
+
+
+def test_add_input():
+
+    a = Tensor([1, 2])
+    b = Tensor([3, 4])
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(a, b):
+        return (a + b) * 2
+
+    fwd(a, b)
+    orig_model = io.BytesIO()
+    fwd.dump(
+        orig_model, arg_names=["a", "b"], output_names="o", optimize_for_inference=False
+    )
+    orig_model.seek(0)
+
+    graph = Net.load(orig_model)
+    inp_c = graph.make_input_node((2,), np.int32, name="c")
+    varo = graph.var_filter.name("o").as_unique()
+
+    out = F.add(varo.var, inp_c.var)
+    out = graph.add_dep_oprs(out)[0]
+    out.name = "o1"
+    graph.remove_output(varo)
+    graph.add_output(out)
+    modified_model = io.BytesIO()
+
+    graph.dump(modified_model)
+    modified_model.seek(0)
+    load_graph = GraphInference(modified_model)
+
+    out = load_graph.run(a, b, a)
+    np.testing.assert_equal(out["o1"], ((a + b) * 2 + a).numpy())
+
+
+def test_add_output():
+
+    a = Tensor([1.0, 2.0])
+    b = Tensor([3.0, 4.0])
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(a, b):
+        return (a + b) * 2
+
+    fwd(a, b)
+    orig_model = io.BytesIO()
+    fwd.dump(
+        orig_model, arg_names=["a", "b"], output_names="o", optimize_for_inference=False
+    )
+    orig_model.seek(0)
+
+    net = Net.load(orig_model)
+    var_a = net.var_filter.name("a").as_unique()
+    var_b = net.var_filter.name("b").as_unique()
+
+    y = F.add(var_a.var, var_b.var)
+    y = F.sigmoid(y)
+
+    new_vars = net.add_dep_oprs(y)[0]
+    new_vars.name = "o1"
+    net.add_output(new_vars)
+
+    modified_model = io.BytesIO()
+    net.dump(modified_model)
+    modified_model.seek(0)
+
+    g = GraphInference(modified_model)
+    out = g.run(a.numpy(), b.numpy())
+
+    np.testing.assert_equal(out["o"], ((a + b) * 2).numpy())
+    np.testing.assert_equal(out["o1"], (F.sigmoid((a + b))).numpy())
+
+
+def test_query():
+    class Model(M.Module):
+        def __init__(self):
+            super().__init__()
+            self.conv1 = M.Conv2d(3, 32, 3)
+            self.conv2 = M.Conv2d(32, 32, 3)
+            self.conv3 = M.Conv2d(32, 32, 3)
+
+        def forward(self, data):
+            x = self.conv1(data)
+            x = self.conv2(x)
+            x = self.conv3(x)
+            return x
+
+    n = Model()
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data):
+        return n(data)
+
+    fwd(Tensor(np.random.random((1, 3, 224, 224))))
+    orig_model = io.BytesIO()
+    fwd.dump(
+        orig_model,
+        arg_names=["data"],
+        output_names="o",
+        keep_opr_name=True,
+        keep_var_name=True,
+        optimize_for_inference=False,
+    )
+    orig_model.seek(0)
+
+    graph = Net.load(orig_model)
+
+    r = graph.data_providers_filter.as_count()
+    assert r == 1
+
+    opr = graph.get_opr_by_type(Host2DeviceCopy)
+    assert isinstance(opr, Host2DeviceCopy)
+
+    r1 = graph.params_filter.as_count()
+    assert r1 == 6
+
+    r2 = graph.opr_filter.type(N.ConvolutionForward).as_count()
+    assert r2 == 3
+
+    r3 = graph.opr_filter.not_type(N.ConvolutionForward).as_count()
+    assert r3 == len(graph.all_oprs) - r2
+
+    var = graph.var_filter.name("data").as_unique()
+    r4 = graph.opr_filter.has_input(var).as_count()
+    assert r4 == 1
+
+    r5 = graph.opr_filter.name("data").as_count()
+    assert r5 == 1
+
+    opr = graph.get_opr_by_name("data")
+    assert isinstance(opr, Host2DeviceCopy)
+
+    var = graph.get_var_by_name("data")
+    assert isinstance(var, VarNode)
+
+    r6 = graph.var_filter.name("*bias").as_count()
+    assert r6 == 3
+
+
+def test_optimize_for_inference():
+    @trace(symbolic=True, capture_as_const=True)
+    def f(x):
+        return F.exp(x)
+
+    orig_model = io.BytesIO()
+    f(Tensor(5.0))
+    f.dump(orig_model, optimize_for_inference=False)
+    orig_model.seek(0)
+
+    optimize_model = io.BytesIO()
+    net = Net.load(orig_model)
+    net.dump(optimize_model, enable_io16xc32=True)
+    optimize_model.seek(0)
+
+    res = G.load_graph(optimize_model)
+    computing_input = res.output_vars_list[0].owner.inputs[0]
+    assert computing_input.dtype == np.float16
+
+
+def test_reset_batchsize():
+    @trace(symbolic=True, capture_as_const=True)
+    def f(x):
+        return F.exp(x)
+
+    orig_model = io.BytesIO()
+    f(Tensor(np.random.random((3, 3, 224, 224))))
+    f.dump(orig_model, optimize_for_inference=False)
+    orig_model.seek(0)
+
+    modified_model = io.BytesIO()
+    net = Net.load(orig_model)
+    net.reset_batch_size(1)
+    net.dump(modified_model, optimize_for_inference=False)
+    modified_model.seek(0)
+
+    net1 = Net.load(modified_model)
+    assert net1.data_providers_filter.as_unique().shape[0] == 1
+
+
+def test_modify_opr_name():
+    @trace(symbolic=True, capture_as_const=True)
+    def f(x):
+        return F.exp(x)
+
+    orig_model = io.BytesIO()
+    f(Tensor(np.random.random((3, 3, 224, 224))))
+    f.dump(orig_model, arg_names=["a"], optimize_for_inference=False)
+    orig_model.seek(0)
+
+    modified_model = io.BytesIO()
+    net = Net.load(orig_model)
+    net.modify_opr_names("net")
+    net.modify_opr_names(lambda x: "net1." + x)
+    net.dump(modified_model, optimize_for_inference=False)
+    modified_model.seek(0)
+
+    net1 = Net.load(modified_model)
+    assert net1.data_providers_filter.as_unique().name == "net1.net.a"

imperative/python/test/unit/utils/test_opr.py

@@ -0,0 +1,712 @@
+import io
+import os
+import platform
+
+import numpy as np
+import pytest
+
+import megengine.core.tensor.dtype as dtype
+import megengine.core.tensor.megbrain_graph as G
+import megengine.functional as F
+import megengine.module as M
+import megengine.random as rand
+from megengine.core._imperative_rt.core2 import apply
+from megengine.core._wrap import Device
+from megengine.core.ops import builtin
+from megengine.device import is_cuda_available
+from megengine.functional.external import tensorrt_runtime_opr
+from megengine.jit.tracing import trace
+from megengine.tensor import Tensor
+from megengine.utils.comp_graph_tools import GraphInference
+from megengine.utils.network import Network as Net
+
+
+def check_pygraph_dump(trace_func, inp_data, expect_results):
+    orig_model = io.BytesIO()
+    inp_size = len(inp_data)
+    out_size = len(expect_results)
+    arg_names = ["arg_{}".format(i) for i in range(inp_size)]
+    output_names = ["out_{}".format(i) for i in range(out_size)]
+    trace_func.dump(
+        orig_model,
+        arg_names=arg_names,
+        output_names=output_names,
+        optimize_for_inference=False,
+    )
+    orig_model.seek(0)
+
+    net = Net.load(orig_model)
+    file = io.BytesIO()
+    net.dump(file, optimize_for_inference=False)
+    file.seek(0)
+    graph = GraphInference(file)
+
+    inp_dict = dict([(arg_names[i], inp_data[i].numpy()) for i in range(inp_size)])
+    results = graph.run(inp_dict=inp_dict)
+
+    for ind, tensor in enumerate(expect_results):
+        np.testing.assert_equal(tensor.numpy(), results[output_names[ind]])
+        assert tensor.dtype == results[output_names[ind]].dtype
+
+
+def test_elemwise():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(x, y):
+        z1 = x * y
+        z2 = x + y
+        z3 = z1 / z2
+        z3 = z3 ** 3
+        return z3
+
+    x = Tensor([1.0, 2.0])
+    y = Tensor([3.0, 5.0])
+    result = fwd(x, y)
+    check_pygraph_dump(fwd, [x, y], [result])
+
+
+def test_reduce():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data):
+        x = data.sum(axis=2)
+        x = x.mean(axis=1)
+        return x
+
+    data = Tensor(np.random.random((1, 32, 32)))
+    result = fwd(data)
+    check_pygraph_dump(fwd, [data], [result])
+
+
+def test_typecvt():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data):
+        return data.astype(dtype.qint8(0.8))
+
+    x = Tensor(np.random.random((2, 3)) * 255)
+    result = fwd(x)
+    check_pygraph_dump(fwd, [x], [result])
+
+
+def test_matinv():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data):
+        return F.matinv(data)
+
+    data = Tensor(np.random.random((5, 5)))
+    result = fwd(data)
+    check_pygraph_dump(fwd, [data], [result])
+
+
+def test_matmul():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data1, data2):
+        return F.matmul(data1, data2)
+
+    data1 = Tensor(np.random.random((32, 64)))
+    data2 = Tensor(np.random.random((64, 16)))
+    result = fwd(data1, data2)
+    check_pygraph_dump(fwd, [data1, data2], [result])
+
+
+def test_batchmatmul():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(x, y):
+        return F.matmul(x, y)
+
+    x = Tensor(np.random.random((3, 3, 5)))
+    y = Tensor(np.random.random((3, 5, 3)))
+    result = fwd(x, y)
+    check_pygraph_dump(fwd, [x, y], [result])
+
+
+def test_dot():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(x, y):
+        return F.dot(x, y)
+
+    x = Tensor([1.0, 2.0, 3.0])
+    y = Tensor([3.0, 4.0, 5.0])
+    result = fwd(x, y)
+    check_pygraph_dump(fwd, [x, y], [result])
+
+
+def test_svd():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data):
+        _, out, _ = F.svd(data)
+        return out
+
+    input = Tensor(np.random.random((1, 1, 3, 3)))
+    result = fwd(input)
+    check_pygraph_dump(fwd, [input], [result])
+
+
+def test_conv():
+    conv = M.Conv2d(3, 32, 3)
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data):
+        return conv(data)
+
+    data = Tensor(np.random.random((1, 3, 32, 32)))
+    result = fwd(data)
+    check_pygraph_dump(fwd, [data], [result])
+
+
+def test_deformable_conv():
+    if not is_cuda_available():
+        return
+    conv = M.DeformableConv2d(3, 32, 3)
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data, offset, mask):
+        return conv(data, offset, mask)
+
+    data = Tensor(np.random.random((1, 3, 32, 32)))
+    offset = Tensor(np.ones((32, 3 * 3 * 2, 30, 30)).astype("int32") * 5)
+    mask = Tensor(np.ones((32, 3 * 3, 30, 30)).astype("int32"))
+    out = fwd(data, offset, mask)
+    check_pygraph_dump(fwd, [data, offset, mask], [out])
+
+
+def test_convtranspose():
+    deconv = M.ConvTranspose2d(32, 32, 3)
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data):
+        return deconv(data)
+
+    data = Tensor(np.random.random((1, 32, 32, 32)))
+    result = fwd(data)
+    check_pygraph_dump(fwd, [data], [result])
+
+
+@pytest.mark.skip(reason="pytest aborted")
+def test_grouplocal():
+    n = M.LocalConv2d(3, 32, 32, 32, 3)
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data):
+        return n(data)
+
+    input = Tensor(np.random.random((1, 3, 32, 32)))
+    result = fwd(input)
+    check_pygraph_dump(fwd, [input], [result])
+
+
+def test_pooling():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data):
+        out = F.max_pool2d(data, 2, 2)
+        out = F.avg_pool2d(out, 2, 2)
+        return out
+
+    data = Tensor(np.random.random((1, 3, 64, 64)))
+    result = fwd(data)
+    check_pygraph_dump(fwd, [data], [result])
+
+
+def test_adaptivepooling():
+    pool1 = M.AdaptiveMaxPool2d((2, 2))
+    pool2 = M.AdaptiveAvgPool2d((2, 2))
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data):
+        out = pool1(data)
+        out = pool2(out)
+        return out
+
+    input = Tensor(np.random.random((1, 3, 32, 32)))
+    result = fwd(input)
+    check_pygraph_dump(fwd, [input], [result])
+
+
+def test_roipooling():
+    inp = Tensor(np.random.random((1, 1, 128, 128)))
+    rois = Tensor(np.random.random((4, 5)))
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(inp, rois):
+        return F.nn.roi_pooling(inp, rois, (2, 2), scale=2.0)
+
+    output = fwd(inp, rois)
+    check_pygraph_dump(fwd, [inp, rois], [output])
+
+
+def test_deformable_ps_roi_pooling():
+    inp = Tensor(np.random.random((1, 256, 64, 64)).astype("float32"))
+    rois = Tensor(np.random.random((1, 5)).astype("float32"))
+    trans = Tensor(np.random.random((24, 2, 7, 7)).astype("float32"))
+
+    pooled_h = 7
+    pooled_w = 7
+    sample_per_part = 4
+    no_trans = False
+    part_size = 7
+    spatial_scale = 1.0 / 64
+    trans_std = 0.1
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(inp, rois, trans):
+        y = F.deformable_psroi_pooling(
+            inp,
+            rois,
+            trans,
+            no_trans,
+            part_size,
+            pooled_h,
+            pooled_w,
+            sample_per_part,
+            spatial_scale,
+            trans_std,
+        )
+        return y
+
+    result = fwd(inp, rois, trans)
+    check_pygraph_dump(fwd, [inp, rois, trans], [result])
+
+
+def test_convbias():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(inp, weight, bias):
+        return F.quantized.conv_bias_activation(
+            inp, weight, bias, dtype=dtype.qint8(scale=1.0), nonlinear_mode="RELU"
+        )
+
+    inp = Tensor(np.random.random((1, 3, 64, 64)), dtype=dtype.qint8(scale=1.0))
+    weight = Tensor(np.random.random((32, 3, 3, 3)), dtype=dtype.qint8(scale=1.0))
+    bias = Tensor(np.random.random((1, 32, 1, 1)), dtype=dtype.qint32(scale=1.0))
+    result = fwd(inp, weight, bias)
+    check_pygraph_dump(fwd, [inp, weight, bias], [result])
+
+
+def test_batch_convbias():
+    if is_cuda_available():
+        return
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(inp, weight, bias):
+        return F.quantized.batch_conv_bias_activation(
+            inp, weight, bias, dtype=dtype.qint8(scale=1.0), nonlinear_mode="RELU"
+        )
+
+    inp = Tensor(np.random.random((1, 3, 64, 64)), dtype=dtype.qint8(scale=1.0))
+    weight = Tensor(np.random.random((1, 32, 3, 3, 3)), dtype=dtype.qint8(scale=1.0))
+    bias = Tensor(np.random.random((1, 32, 1, 1)), dtype=dtype.qint32(scale=1.0))
+    result = fwd(inp, weight, bias)
+    check_pygraph_dump(fwd, [inp, weight, bias], [result])
+
+
+def test_batchnorm():
+    bn = M.BatchNorm2d(32)
+    bn.eval()
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data):
+        return bn(data)
+
+    data = Tensor(np.random.random((1, 32, 32, 32)))
+    result = fwd(data)
+    check_pygraph_dump(fwd, [data], [result])
+
+
+def test_roialign():
+    inp = Tensor(np.random.randn(1, 1, 128, 128))
+    rois = Tensor(np.random.random((4, 5)))
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(inp, rois):
+        return F.nn.roi_align(inp, rois, (2, 2))
+
+    output = fwd(inp, rois)
+    check_pygraph_dump(fwd, [inp, rois], [output])
+
+
+def test_warpperspective():
+    inp_shape = (1, 1, 4, 4)
+    x = Tensor(np.arange(16, dtype=np.float32).reshape(inp_shape))
+    M_shape = (1, 3, 3)
+    # M defines a translation: dst(1, 1, h, w) = rst(1, 1, h+1, w+1)
+    M = Tensor(
+        np.array(
+            [[1.0, 0.0, 1.0], [0.0, 1.0, 1.0], [0.0, 0.0, 1.0]], dtype=np.float32
+        ).reshape(M_shape)
+    )
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(x, M):
+        return F.warp_perspective(x, M, (2, 2))
+
+    result = fwd(x, M)
+    check_pygraph_dump(fwd, [x, M], [result])
+
+
+def test_warpaffine():
+    inp_shape = (1, 3, 3, 3)
+    x = Tensor(np.arange(27, dtype=np.float32).reshape(inp_shape))
+    weightv = Tensor([[[1.26666667, 0.6, -83.33333333], [-0.33333333, 1, 66.66666667]]])
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(x, weightv):
+        return F.warp_affine(x, weightv, (2, 2), border_mode="WRAP")
+
+    outp = fwd(x, weightv)
+    check_pygraph_dump(fwd, [x, weightv], [outp])
+
+
+def test_remap():
+    inp_shape = (1, 1, 4, 4)
+    inp = Tensor(np.arange(16, dtype=np.float32).reshape(inp_shape))
+    map_xy_shape = (1, 2, 2, 2)
+    map_xy = Tensor(
+        np.array(
+            [[[1.0, 0.0], [0.0, 1.0]], [[0.0, 1.0], [0.0, 1.0]]], dtype=np.float32
+        ).reshape(map_xy_shape)
+    )
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(inp, map_xy):
+        return F.remap(inp, map_xy)
+
+    out = fwd(inp, map_xy)
+    check_pygraph_dump(fwd, [inp, map_xy], [out])
+
+
+def test_resize():
+    x = Tensor(np.random.randn(10, 3, 32, 32))
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(x):
+        return F.nn.interpolate(x, size=(16, 16), mode="BILINEAR")
+
+    out = fwd(x)
+    check_pygraph_dump(fwd, [x], [out])
+
+
+def test_index_onehot():
+    src = Tensor([[1.0, 2.0]])
+    index = Tensor([0])
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(src, index):
+        return F.indexing_one_hot(src, index)
+
+    out = fwd(src, index)
+    check_pygraph_dump(fwd, [src, index], [out])
+
+
+def test_set_onehot():
+    x = Tensor(np.arange(1, 4, dtype=np.int32))
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(x):
+        return F.one_hot(x, num_classes=4)
+
+    out = fwd(x)
+    check_pygraph_dump(fwd, [x], [out])
+
+
+def test_copy():
+    x = Tensor([1, 2, 3])
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(x):
+        return x.to("cpu0:0")
+
+    o = fwd(x)
+    check_pygraph_dump(fwd, [x], [o])
+
+
+def test_argsort():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data):
+        return F.argsort(data, True)
+
+    data = Tensor([1.0, 2.0, 3.0, 5.0])
+    result = fwd(data)
+    check_pygraph_dump(fwd, [data], [result])
+
+
+def test_argmax_min():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data):
+        return F.argmax(data), F.argmin(data)
+
+    data = Tensor(np.random.random((10, 10)))
+    result = fwd(data)
+    check_pygraph_dump(fwd, [data], result)
+
+
+def test_condtake():
+    mask = Tensor(np.array([[True, False], [False, True]], dtype=np.bool_))
+    x = Tensor(np.array([[1, np.inf], [np.nan, 4]], dtype=np.float32))
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(mask, x):
+        v, index = F.cond_take(mask, x)
+        return v, index
+
+    v, index = fwd(mask, x)
+    check_pygraph_dump(fwd, [mask, x], [v, index])
+
+
+def test_topk():
+    x = Tensor(np.array([2, 4, 6, 8, 7, 5, 3, 1], dtype=np.float32))
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(x):
+        top, indices = F.topk(x, 5)
+        return top, indices
+
+    top, indices = fwd(x)
+    check_pygraph_dump(fwd, [x], [top, indices])
+
+
+
+
+def test_random():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd():
+        x = rand.uniform(size=(2, 2))
+        y = rand.normal(size=(1, 3, 3, 3))
+        return x, y
+
+    x, y = fwd()
+    check_pygraph_dump(fwd, [], [x, y])
+
+
+def test_tensor_gen():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd():
+        a = F.linspace(3, 10, 3, device=Device("xpux").to_c())
+        b = F.eye(3, device=Device("xpux").to_c())
+        return a, b
+
+    a, b = fwd()
+    check_pygraph_dump(fwd, [], [a, b])
+
+
+def test_getvarshape():
+    op = builtin.GetVarShape(axis=1)
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data):
+        return apply(op, data)[0]
+
+    data = Tensor(np.random.random((1, 2, 3, 4)))
+    result = fwd(data)
+    check_pygraph_dump(fwd, [data], [result])
+
+
+def test_concat():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data1, data2):
+        return F.concat([data1, data2], axis=1)
+
+    x = Tensor(np.random.random((2, 3)))
+    y = Tensor(np.random.random((2, 5)))
+    result = fwd(x, y)
+    check_pygraph_dump(fwd, [x, y], [result])
+
+
+def test_broadcast():
+    inp = Tensor([[1], [2], [3], [4]])
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(inp):
+        return F.broadcast_to(inp, (4, 4))
+
+    out = fwd(inp)
+    check_pygraph_dump(fwd, [inp], [out])
+
+
+def test_identity():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data):
+        return F.copy(data)
+
+    data = Tensor([1.0, 2.0])
+    result = fwd(data)
+    check_pygraph_dump(fwd, [data], [result])
+
+
+@pytest.mark.skip(reason="advance indexing trace error")
+def test_nms():
+    x = np.zeros((100, 4))
+    np.random.seed(42)
+    x[:, :2] = np.random.rand(100, 2) * 20
+    x[:, 2:] = np.random.rand(100, 2) * 20 + 100
+    scores = Tensor(np.random.rand(100))
+    inp = Tensor(x)
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(inp, scores):
+        return F.nn.nms(inp, scores, iou_thresh=0.7, max_output=3)
+
+    result = fwd(inp, scores)
+    check_pygraph_dump(fwd, [inp, scores], [result])
+
+
+def test_dimshuffle():
+    inp = Tensor([1, 2, 3, 4])
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(inp):
+        return inp.T
+
+    out = fwd(inp)
+    check_pygraph_dump(fwd, [inp], [out])
+
+
+def test_reshape():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data):
+        return data.reshape((1, 8))
+
+    data = Tensor(np.random.random((1, 2, 2, 2)))
+    result = fwd(data)
+    check_pygraph_dump(fwd, [data], [result])
+
+
+def test_add_remove_axis():
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(data):
+        x = F.expand_dims(data, [0, 0])
+        y = F.squeeze(x, 0)
+        return y
+
+    data = Tensor([1.0, 2.0])
+    result = fwd(data)
+    check_pygraph_dump(fwd, [data], [result])
+
+
+@pytest.mark.parametrize("mode", ["get", "set", "inc"])
+def test_subtensor(mode):
+    items = [[0, True, True, True, False], [1, False, False, False, True]]
+    data = [Tensor(np.random.random((5, 5))), Tensor(np.random.random(2))]
+    if mode == "get":
+        op = builtin.Subtensor(items)
+        data = data[:1]
+    if mode == "set":
+        op = builtin.SetSubtensor(items)
+    if mode == "inc":
+        op = builtin.IncrSubtensor(items)
+    tensors = [Tensor(0), Tensor(4), Tensor(2), Tensor(3)]
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(*tensors):
+        return apply(op, *tensors)[0]
+
+    result = fwd(*data, *tensors)
+    check_pygraph_dump(fwd, data + tensors, [result])
+
+
+@pytest.mark.parametrize("mode", ["get", "set", "inc"])
+def test_advance_indexing(mode):
+    items = [[0, False, False, False, True]]
+    tensors = [Tensor([0, 4, 2])]
+    data = [Tensor(np.random.random((5, 5))), Tensor(np.random.random((3, 5)))]
+    if mode == "get":
+        op = builtin.IndexingMultiAxisVec(items)
+        data = data[:1]
+    if mode == "set":
+        op = builtin.IndexingSetMultiAxisVec(items)
+    if mode == "inc":
+        op = builtin.IndexingIncrMultiAxisVec(items)
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(*tensors):
+        return apply(op, *tensors)[0]
+
+    result = fwd(*data, *tensors)
+    check_pygraph_dump(fwd, data + tensors, [result])
+
+
+@pytest.mark.parametrize("mode", ["get", "set", "inc"])
+def test_mesh_indexing(mode):
+    items = [[0, True, True, True, False], [1, False, False, False, True]]
+    tensors = [Tensor(0), Tensor(5), Tensor(2), Tensor([1, 3])]
+    data = [Tensor(np.random.random((5, 5))), Tensor(np.random.random((3, 2)))]
+    if mode == "get":
+        op = builtin.IndexingMultiAxisVec(items)
+        data = data[:1]
+    if mode == "set":
+        op = builtin.IndexingSetMultiAxisVec(items)
+    if mode == "inc":
+        op = builtin.IndexingIncrMultiAxisVec(items)
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(*tensors):
+        return apply(op, *tensors)[0]
+
+    result = fwd(*data, *tensors)
+    check_pygraph_dump(fwd, data + tensors, [result])
+
+
+@pytest.mark.parametrize("mode", ["get", "set", "inc"])
+def test_batch_mesh_indexing(mode):
+    items = [[1, False, False, False, True], [2, False, False, False, True]]
+    tensors = [Tensor([[0, 2], [0, 2]]), Tensor([[0, 1, 2], [1, 2, 3]])]
+    data = [Tensor(np.random.random((2, 3, 4))), Tensor(np.random.random((2, 2, 3)))]
+    if mode == "get":
+        op = builtin.BatchedMeshIndexing(items)
+        data = data[:1]
+    if mode == "set":
+        op = builtin.BatchedSetMeshIndexing(items)
+    if mode == "inc":
+        op = builtin.BatchedIncrMeshIndexing(items)
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(*tensors):
+        return apply(op, *tensors)[0]
+
+    result = fwd(*data, *tensors)
+    check_pygraph_dump(fwd, data + tensors, [result])
+
+
+@pytest.mark.skip(reason="tmp skip")
+def test_assert_equal():
+    g = G.Graph()
+    inp1 = g.make_h2d(dtype=np.float32, device="xpux")
+    inp2 = g.make_h2d(dtype=np.float32, device="xpux")
+    op = builtin.AssertEqual(maxerr=1e-5)
+    out = G.apply_normal_varnode(op, inp1._node, inp2._node)[0]
+    print(out)
+    g.compile(out)
+    file = io.BytesIO()
+    out_model = G.dump_graph([out])
+    file.write(out_model[0])
+    file.seek(0)
+    net = Net.load(file)
+
+    dump_file = io.BytesIO()
+    net.dump(dump_file)
+    dump_file.seek(0)
+    g = GraphInference(dump_file)
+    g.run(np.array([1.0, 2.0]), np.array([1.0, 2.0]))
+
+
+def test_elemwise_multitype():
+    op = builtin.ElemwiseMultiType(mode="QADD", dtype=dtype.qint32(2.0))
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(x, y):
+        return apply(op, x, y)[0]
+
+    x = Tensor(np.random.random(10) * 10, dtype=dtype.qint8(2.0))
+    y = Tensor(np.random.random(10) * 10, dtype=dtype.qint8(2.0))
+    result = fwd(x, y)
+    check_pygraph_dump(fwd, [x, y], [result])
+
+
+
+
+def test_cvtcolor():
+    inp = np.random.randn(3, 3, 3, 3).astype(np.float32)
+    x = Tensor(inp)
+
+    @trace(symbolic=True, capture_as_const=True)
+    def fwd(inp):
+        return F.img_proc.cvt_color(inp, mode="RGB2GRAY")
+
+    result = fwd(x)
+    check_pygraph_dump(fwd, [x], [result])

src/plugin/impl/opr_footprint.cpp

@@ -17,9 +17,20 @@
 #include "megbrain/opr/dnn/local.h"
 #include "megbrain/opr/dnn/lrn.h"
 #include "megbrain/opr/dnn/pooling.h"
+#include "megbrain/opr/dnn/adaptive_pooling.h"
+#include "megbrain/opr/dnn/roi_pooling.h"
+#include "megbrain/opr/dnn/roi_align.h"
 #include "megbrain/opr/imgproc.h"
+#include "megbrain/opr/standalone/nms_opr.h"
 #include "megbrain/opr/io.h"
 #include "megbrain/opr/tensor_manip.h"
+#include "megbrain/opr/rand.h"
+#include "megbrain/opr/dnn/batch_norm.h"
+#include "megbrain/opr/misc.h"
+#include "megbrain/opr/indexing.h"
+#include "megbrain/opr/internal/indexing_helper.h"
+#include "megbrain/opr/nn_int.h"
+#include "megbrain/opr/tensor_gen.h"
 #if MGB_ENABLE_JSON
 #include "megdnn/opr_param_json.h"
 #endif
@@ -354,7 +365,7 @@ uint64_t opr_footprint_func<opr::DeformableConvForward>(
     auto&& out_shape = opr->output()[0]->shape();
     auto&& filter_shape = opr->input()[1]->shape();
     using Param = opr::DeformableConvForward::Param;
-    auto&& param = opr->cast_final_safe<opr::Convolution>().param();
+    auto&& param = opr->cast_final_safe<opr::DeformableConvForward>().param();
     size_t fh, fw, icpg;
     mgb_assert(param.format == Param::Format::NCHW);
     if (param.sparse == Param::Sparse::GROUP) {
@@ -425,9 +436,11 @@ uint64_t opr_footprint_func<opr::BatchConvBiasForward>(
     auto&& filter_shape = opr->input()[1]->shape();
     using Param = opr::BatchConvBiasForward::Param;
     auto&& param = opr->cast_final_safe<opr::BatchConvBiasForward>().param();
-    mgb_assert(param.format == Param::Format::NCHW4);
-    size_t packed_channels = 4;
+    size_t packed_channels = 1;
     size_t kern_spatial_pos = 3;
+    if (param.format == Param::Format::NCHW4) {
+        packed_channels = 4;
+    }
     size_t fh = filter_shape[kern_spatial_pos],
            fw = filter_shape[kern_spatial_pos + 1];
     return out_shape.total_nr_elems() * fh * fw * src_shape[1] *
@@ -508,7 +521,29 @@ REGISTE_PARAM_JSON_FUNC(LocalShareBackwardFilter)
 REGISTE_PARAM_JSON_FUNC(DeformableConvForward)
 REGISTE_PARAM_JSON_FUNC(DeformableConvBackwardFilter)
 REGISTE_PARAM_JSON_FUNC(DeformableConvBackwardData)
+REGISTE_PARAM_JSON_FUNC(DeformablePSROIPoolingForward)
 REGISTE_PARAM_JSON_FUNC(BatchConvBiasForward)
+REGISTE_PARAM_JSON_FUNC(BatchNormForward)
+REGISTE_PARAM_JSON_FUNC(ElemwiseMultiType)
+REGISTE_PARAM_JSON_FUNC(Argsort)
+REGISTE_PARAM_JSON_FUNC(Argmax)
+REGISTE_PARAM_JSON_FUNC(Argmin)
+REGISTE_PARAM_JSON_FUNC(AdaptivePooling)
+REGISTE_PARAM_JSON_FUNC(ROIPooling)
+REGISTE_PARAM_JSON_FUNC(ROIAlign)
+REGISTE_PARAM_JSON_FUNC(WarpPerspective)
+REGISTE_PARAM_JSON_FUNC(WarpAffine)
+REGISTE_PARAM_JSON_FUNC(Remap)
+REGISTE_PARAM_JSON_FUNC(Resize)
+REGISTE_PARAM_JSON_FUNC(IndexingOneHot)
+REGISTE_PARAM_JSON_FUNC(IndexingSetOneHot)
+REGISTE_PARAM_JSON_FUNC(TopK)
+REGISTE_PARAM_JSON_FUNC(UniformRNG)
+REGISTE_PARAM_JSON_FUNC(GaussianRNG)
+REGISTE_PARAM_JSON_FUNC(Linspace)
+REGISTE_PARAM_JSON_FUNC(Eye)
+REGISTE_PARAM_JSON_FUNC(CvtColor)
+
 
 template <>
 std::shared_ptr<json::Value> opr_param_json_func<opr::Dimshuffle>(
@@ -547,24 +582,83 @@ std::shared_ptr<json::Value> opr_param_json_func<opr::AxisAddRemove>(
         });
     }
 
+std::shared_ptr<json::Value> indexing_param_to_json(
+        const std::vector<opr::indexing::AxisIndexer>& indices) {
+    auto desc = json::Array::make();
+    for (auto& index : indices) {
+        desc->add(json::Object::make({
+                {"axis", json::NumberInt::make(index.axis.get_raw())},
+                {"begin",
+                    json::NumberInt::make(index.begin.node() != nullptr)},
+                {"end", json::NumberInt::make(index.end.node() != nullptr)},
+                {"step",
+                    json::NumberInt::make(index.step.node() != nullptr)},
+                {"idx", json::NumberInt::make(index.idx.node() != nullptr)},
+        }));
+    }
+    return desc;
+}
+
+#define REGISTE_INDEXING_PARAM_JSON_FUNC(cls)                         \
+    template <>                                                       \
+    std::shared_ptr<json::Value> opr_param_json_func<opr::cls>(       \
+            cg::OperatorNodeBase * opr) {                             \
+        auto indices = opr->cast_final_safe<opr::cls>().index_desc(); \
+        return indexing_param_to_json(indices);                  \
+    }
+
+REGISTE_INDEXING_PARAM_JSON_FUNC(Subtensor);
+REGISTE_INDEXING_PARAM_JSON_FUNC(SetSubtensor);
+REGISTE_INDEXING_PARAM_JSON_FUNC(IncrSubtensor);
+REGISTE_INDEXING_PARAM_JSON_FUNC(IndexingMultiAxisVec);
+REGISTE_INDEXING_PARAM_JSON_FUNC(IndexingSetMultiAxisVec);
+REGISTE_INDEXING_PARAM_JSON_FUNC(IndexingIncrMultiAxisVec);
+REGISTE_INDEXING_PARAM_JSON_FUNC(MeshIndexing);
+REGISTE_INDEXING_PARAM_JSON_FUNC(IncrMeshIndexing);
+REGISTE_INDEXING_PARAM_JSON_FUNC(SetMeshIndexing);
+REGISTE_INDEXING_PARAM_JSON_FUNC(BatchedMeshIndexing);
+REGISTE_INDEXING_PARAM_JSON_FUNC(BatchedIncrMeshIndexing);
+REGISTE_INDEXING_PARAM_JSON_FUNC(BatchedSetMeshIndexing);
+
 template <>
-std::shared_ptr<json::Value> opr_param_json_func<opr::Subtensor>(
+std::shared_ptr<json::Value> opr_param_json_func<opr::Reshape>(
     cg::OperatorNodeBase * opr) {
         auto desc = json::Array::make();
-        auto indices = opr->cast_final_safe<opr::Subtensor>().index_desc();
-        for (auto &index : indices){
-            desc->add(
-                json::Object::make({
-                    {"axis", json::NumberInt::make(index.axis.get_raw())},
-                    {"begin", json::NumberInt::make(index.begin.node() != nullptr)},
-                    {"end", json::NumberInt::make(index.end.node() != nullptr)},
-                    {"step", json::NumberInt::make(index.step.node() != nullptr)},
-                    {"idx", json::NumberInt::make(index.idx.node() != nullptr)},
-                }));
+        auto axis_param = opr->cast_final_safe<opr::Reshape>().param();
+        if (axis_param.axis != axis_param.MAX_NDIM){
+            return json::Object::make({
+                {"axis", json::NumberInt::make(axis_param.axis)},
+            });
+        } else {
+            return json::Object::make();
         }
+    }
 
-        return desc;
+template <>
+std::shared_ptr<json::Value> opr_param_json_func<opr::GetVarShape>(
+    cg::OperatorNodeBase * opr) {
+        auto desc = json::Array::make();
+        auto axis_param = opr->cast_final_safe<opr::GetVarShape>().param();
+        if (axis_param.axis != axis_param.MAX_NDIM){
+            return json::Object::make({
+                {"axis", json::NumberInt::make(axis_param.axis)},
+            });
+        } else {
+            return json::Object::make();
+        }
     }
+
+template <>
+std::shared_ptr<json::Value> opr_param_json_func<opr::standalone::NMSKeep>(
+    cg::OperatorNodeBase * opr) {
+        auto nms_param = opr->cast_final_safe<opr::standalone::NMSKeep>().param();
+        return json::Object::make({
+                {"iou_thresh", json::Number::make(nms_param.iou_thresh)},
+                {"max_output", json::Number::make(nms_param.max_output)},
+            });
+    }
+    
+
 #endif // MGB_ENABLE_JSON
 
 }  // namespace
@@ -632,6 +726,17 @@ void OprFootprint::init_all_footprints() {
     add_single_param_json<opr::Dimshuffle>();
     add_single_param_json<opr::AxisAddRemove>();
     add_single_param_json<opr::Subtensor>();
+    add_single_param_json<opr::SetSubtensor>();
+    add_single_param_json<opr::IncrSubtensor>();
+    add_single_param_json<opr::IndexingMultiAxisVec>();
+    add_single_param_json<opr::IndexingSetMultiAxisVec>();
+    add_single_param_json<opr::IndexingIncrMultiAxisVec>();
+    add_single_param_json<opr::MeshIndexing>();
+    add_single_param_json<opr::SetMeshIndexing>();
+    add_single_param_json<opr::IncrMeshIndexing>();
+    add_single_param_json<opr::BatchedMeshIndexing>();
+    add_single_param_json<opr::BatchedSetMeshIndexing>();
+    add_single_param_json<opr::BatchedIncrMeshIndexing>();
     add_single_param_json<opr::Reduce>();
     add_single_param_json<opr::LocalShareForward>();
     add_single_param_json<opr::LocalShareBackwardData>();
@@ -639,7 +744,31 @@ void OprFootprint::init_all_footprints() {
     add_single_param_json<opr::DeformableConvForward>();
     add_single_param_json<opr::DeformableConvBackwardFilter>();
     add_single_param_json<opr::DeformableConvBackwardData>();
+    add_single_param_json<opr::DeformablePSROIPoolingForward>();
     add_single_param_json<opr::BatchConvBiasForward>();
+    add_single_param_json<opr::BatchNormForward>();
+    add_single_param_json<opr::Reshape>();
+    add_single_param_json<opr::GetVarShape>();
+    add_single_param_json<opr::Argsort>();
+    add_single_param_json<opr::Argmin>();
+    add_single_param_json<opr::Argmax>();
+    add_single_param_json<opr::ElemwiseMultiType>();
+    add_single_param_json<opr::AdaptivePooling>();
+    add_single_param_json<opr::ROIPooling>();
+    add_single_param_json<opr::ROIAlign>();
+    add_single_param_json<opr::WarpPerspective>();
+    add_single_param_json<opr::Remap>();
+    add_single_param_json<opr::Resize>();
+    add_single_param_json<opr::IndexingOneHot>();
+    add_single_param_json<opr::IndexingSetOneHot>();
+    add_single_param_json<opr::WarpAffine>();
+    add_single_param_json<opr::TopK>();
+    add_single_param_json<opr::UniformRNG>();
+    add_single_param_json<opr::GaussianRNG>();
+    add_single_param_json<opr::Linspace>();
+    add_single_param_json<opr::Eye>();
+    add_single_param_json<opr::standalone::NMSKeep>();
+    add_single_param_json<opr::CvtColor>();
 
 #endif
 }