ڧgf,ddlZddlZddlZddlZddlmZddlmZddlm Z ddl m Z m Z m Z mZmZmZmZddlZddlZddlmZmZddlmZdd gZGd d ejZGd d eZdZdedefdZdeefdZdee e e fde e e ffdZ! d"dej"dee e e fde#dee e e fdeee ee ff dZ$Gddej%Z& d#dej"deeee e e e ffdeeee e e e ffd eeee e e e ffdee e e fde#dee e e fdej%fd!Z'dS)$N) OrderedDict)deepcopy)chain)AnyCallableDictListOptionalTupleUnion)fxnn) _copy_attrcreate_feature_extractorget_graph_node_namescDeZdZdZfdZdejdedeffd Z xZ S)LeafModuleAwareTracerz An fx.Tracer that allows the user to specify a set of leaf modules, i.e. modules that are not to be traced through. The resulting graph ends up having single nodes referencing calls to the leaf modules' forward methods. ci|_d|vr|d}||_tj|i|dS)N leaf_modules)rpopsuper__init__)selfargskwargsr __class__s a/var/www/html/ai-engine/env/lib/python3.11/site-packages/torchvision/models/feature_extraction.pyrzLeafModuleAwareTracer.__init__sQ V # #!::n55L ,D $)&)))))mmodule_qualnamereturnct|t|jrdSt||S)NT) isinstancetuplerris_leaf_module)rrr rs rr%z$LeafModuleAwareTracer.is_leaf_module!s? at011 2 2 4ww%%a999r) __name__ __module__ __qualname____doc__rrModulestrboolr% __classcell__rs@rrrsw *****: :C:D::::::::::rrceZdZdZfdZdejjdefdZ dde de j j d e jjffd Zd e d e j jd e fd ZxZS)NodePathTracerah NodePathTracer is an FX tracer that, for each operation, also records the name of the Node from which the operation originated. A node name here is a `.` separated path walking the hierarchy from top level module down to leaf operation or leaf module. The name of the top level module is not included as part of the node name. For example, if we trace a module whose forward method applies a ReLU module, the name for that node will simply be 'relu'. Some notes on the specifics: - Nodes are recorded to `self.node_to_qualname` which is a dictionary mapping a given Node object to its node name. - Nodes are recorded in the order which they are executed during tracing. - When a duplicate node name is encountered, a suffix of the form _{int} is added. The counter starts from 1. cntj|i|d|_t|_dS)N)rrcurrent_module_qualnamernode_to_qualname)rrrrs rrzNodePathTracer.__init__:s;$)&)))')$ !, rrforwardc|j} ||}||_|||s||i|}|||_S|d|||||_S#||_wxYw)a Override of `fx.Tracer.call_module` This override: 1) Stores away the qualified name of the caller for restoration later 2) Adds the qualified name of the caller to `current_module_qualname` for retrieval by `create_proxy` 3) Once a leaf module is reached, calls `create_proxy` 4) Restores the caller's qualified name into current_module_qualname call_module)r3path_of_moduler% create_proxy)rrr5rr old_qualnamer outs rr7zNodePathTracer.call_moduleDs3  8"11!44O+:D (&&q/:: gt.v..,8D ( ($$]OT6RR+7D ( (z'_is_subseq...s(996V#999999rN)any)r^r`iter_xs @rraz_is_subseq..sAJJfs9999&99999JJJJJJr)iterall)xyrcs @r _is_subseqrhs3!WWF JJJJJJJ J JJr train_tracer eval_tracerct|j}t|j}t|t|kr)t dt ||DrdSd}t ||rd}nt ||rd}nd}tj||zdS)z Utility function for warning the user if there are differences between the train graph nodes and the eval graph nodes. c3(K|] \}}||kVdSr\r])r^tes rraz*_warn_graph_differences..s*2c2cda162c2c2c2c2c2crNztWhen choosing nodes for feature extraction, you may need to specify output nodes for train and eval mode separately.znNOTE: The nodes obtained by tracing the model in eval mode are a subsequence of those obtained in train mode. znNOTE: The nodes obtained by tracing the model in train mode are a subsequence of those obtained in eval mode. zdThe nodes obtained by tracing the model in train mode are different to those obtained in eval mode. ) listr4rNrIreziprhwarningswarn)rirj train_nodes eval_nodessuggestion_msgmsgs r_warn_graph_differencesrws |4;;==>>Kk299;;<` in the `documentation `_. Args: model (nn.Module): model for which we'd like to print node names tracer_kwargs (dict, optional): a dictionary of keyword arguments for ``NodePathTracer`` (they are eventually passed onto `torch.fx.Tracer `_). By default, it will be set to wrap and make leaf nodes all torchvision ops: {"autowrap_modules": (math, torchvision.ops,),"leaf_modules": _get_leaf_modules_for_ops(),} WARNING: In case the user provides tracer_kwargs, above default arguments will be appended to the user provided dictionary. suppress_diff_warning (bool, optional): whether to suppress a warning when there are discrepancies between the train and eval version of the graph. Defaults to False. concrete_args (Optional[Dict[str, any]]): Concrete arguments that should not be treated as Proxies. According to the `Pytorch docs `_, this parameter's API may not be guaranteed. Returns: tuple(list, list): a list of node names from tracing the model in train mode, and another from tracing the model in eval mode. Examples:: >>> model = torchvision.models.resnet18() >>> train_nodes, eval_nodes = get_graph_node_names(model) rr]) rtrainingr0tracetrainevalror4rNrw) rrrr is_trainingrirjrsrts rrrsd/}==M.K!22M22Lu{{}}MBBB 11=11Kejjll-@@@|4;;==>>Kk299;;<)rr rrrr&rrrrdnodesrHr#r=r+ TypeErrortyperrgraph _tracer_clsr()rrrrrr@rs rrzDualGraphModule.__init__s bnd##,,...",&$${0114 8H3I3IJJ 4 4Dw555!$+s33i#$gTXY]YdTeTe$g$ghhh4t{333    ? &$*:*F F FmZ^ZiZummEIEUEammm   : ! 6j 8N8[&[&[#z5D    6 6&[&[rTc|r|js |j|_n|s|jr |j|_t |S)z Swap out the graph depending on the selected training mode. NOTE this should be safe when calling model.eval() because that just calls this with mode == False. )mode)rrrrrr)rrrs rrzDualGraphModule.train?sT  )  ))DJJ )$- )DJww}}$}'''r)r)T) r&r'r(r)rUrr*r Graphr+rrr-r.s@rrr ser)6)6HO)624()6HJ)6^a)6)6)6)6)6)6V ( ( ( ( ( ( ( ( ( (rr return_nodestrain_return_nodeseval_return_nodesc D t|}|j}td|||fDrtd|du|duz rtd|du|duz stddtt t ffd}|*||}t |}t |}n||}||}i} i} ||d} dD]Z} | d kr|n| d kr|tdi|} | || }t|tj r |jjn|j}t!j| j||}t'| j}t-t/|t-|krtd | | D]0t3fd |Dstdd1g}t5|jjD]"}|jdkr||#|std|D]}|j|d|jjD}tA}t5|D]}| j!|}|| | D]|"d}d#|$dd|dzkr.||| | <| | %n}tAt5t'|&}|j'|d5|j(|dddn #1swxYwY|j)|*| | | <|| | <\|stW| d | d tY|| d | d |}|||||S)a2 Creates a new graph module that returns intermediate nodes from a given model as dictionary with user specified keys as strings, and the requested outputs as values. This is achieved by re-writing the computation graph of the model via FX to return the desired nodes as outputs. All unused nodes are removed, together with their corresponding parameters. Desired output nodes must be specified as a ``.`` separated path walking the module hierarchy from top level module down to leaf operation or leaf module. For more details on the node naming conventions used here, please see the :ref:`relevant subheading ` in the `documentation `_. Not all models will be FX traceable, although with some massaging they can be made to cooperate. Here's a (not exhaustive) list of tips: - If you don't need to trace through a particular, problematic sub-module, turn it into a "leaf module" by passing a list of ``leaf_modules`` as one of the ``tracer_kwargs`` (see example below). It will not be traced through, but rather, the resulting graph will hold a reference to that module's forward method. - Likewise, you may turn functions into leaf functions by passing a list of ``autowrap_functions`` as one of the ``tracer_kwargs`` (see example below). - Some inbuilt Python functions can be problematic. For instance, ``int`` will raise an error during tracing. You may wrap them in your own function and then pass that in ``autowrap_functions`` as one of the ``tracer_kwargs``. For further information on FX see the `torch.fx documentation `_. Args: model (nn.Module): model on which we will extract the features return_nodes (list or dict, optional): either a ``List`` or a ``Dict`` containing the names (or partial names - see note above) of the nodes for which the activations will be returned. If it is a ``Dict``, the keys are the node names, and the values are the user-specified keys for the graph module's returned dictionary. If it is a ``List``, it is treated as a ``Dict`` mapping node specification strings directly to output names. In the case that ``train_return_nodes`` and ``eval_return_nodes`` are specified, this should not be specified. train_return_nodes (list or dict, optional): similar to ``return_nodes``. This can be used if the return nodes for train mode are different than those from eval mode. If this is specified, ``eval_return_nodes`` must also be specified, and ``return_nodes`` should not be specified. eval_return_nodes (list or dict, optional): similar to ``return_nodes``. This can be used if the return nodes for train mode are different than those from eval mode. If this is specified, ``train_return_nodes`` must also be specified, and `return_nodes` should not be specified. tracer_kwargs (dict, optional): a dictionary of keyword arguments for ``NodePathTracer`` (which passes them onto it's parent class `torch.fx.Tracer `_). By default, it will be set to wrap and make leaf nodes all torchvision ops: {"autowrap_modules": (math, torchvision.ops,),"leaf_modules": _get_leaf_modules_for_ops(),} WARNING: In case the user provides tracer_kwargs, above default arguments will be appended to the user provided dictionary. suppress_diff_warning (bool, optional): whether to suppress a warning when there are discrepancies between the train and eval version of the graph. Defaults to False. concrete_args (Optional[Dict[str, any]]): Concrete arguments that should not be treated as Proxies. According to the `Pytorch docs `_, this parameter's API may not be guaranteed. Examples:: >>> # Feature extraction with resnet >>> model = torchvision.models.resnet18() >>> # extract layer1 and layer3, giving as names `feat1` and feat2` >>> model = create_feature_extractor( >>> model, {'layer1': 'feat1', 'layer3': 'feat2'}) >>> out = model(torch.rand(1, 3, 224, 224)) >>> print([(k, v.shape) for k, v in out.items()]) >>> [('feat1', torch.Size([1, 64, 56, 56])), >>> ('feat2', torch.Size([1, 256, 14, 14]))] >>> # Specifying leaf modules and leaf functions >>> def leaf_function(x): >>> # This would raise a TypeError if traced through >>> return int(x) >>> >>> class LeafModule(torch.nn.Module): >>> def forward(self, x): >>> # This would raise a TypeError if traced through >>> int(x.shape[0]) >>> return torch.nn.functional.relu(x + 4) >>> >>> class MyModule(torch.nn.Module): >>> def __init__(self): >>> super().__init__() >>> self.conv = torch.nn.Conv2d(3, 1, 3) >>> self.leaf_module = LeafModule() >>> >>> def forward(self, x): >>> leaf_function(x.shape[0]) >>> x = self.conv(x) >>> return self.leaf_module(x) >>> >>> model = create_feature_extractor( >>> MyModule(), return_nodes=['leaf_module'], >>> tracer_kwargs={'leaf_modules': [LeafModule], >>> 'autowrap_functions': [leaf_function]}) c3K|]}|duV dSr\r])r^args rraz+create_feature_extractor..s& X X33$; X X X X X XrzcEither `return_nodes` or `train_return_nodes` and `eval_return_nodes` together, should be specifiedNzcIf any of `train_return_nodes` and `eval_return_nodes` are specified, then both should be specifiedz\If `train_return_nodes` and `eval_return_nodes` are specified, then both should be specifiedr!ct|tr d|DSd|DS)NcHi|]}t|t| Sr]r+)r^is r z@create_feature_extractor..to_strdict..s&...qCFFCFF...rcNi|]"\}}t|t|#Sr]r)r^kvs rrz@create_feature_extractor..to_strdict..s*55541aAA555r)r#roitems)ns r to_strdictz,create_feature_extractor..to_strdictsD a   /..A... .55177995555r)rrrrrzYThere are duplicate nodes! Please raise an issue https://github.com/pytorch/vision/issuescFg|]}tjdd|duS)^z(\.|$)N)rJrK)r^rquerys r z,create_feature_extractor..s6]]]!3e!3!3!3Q77tC]]]rznode: 'z' is not present in model. Hint: use `get_graph_node_names` to make sure the `return_nodes` you specified are present. It may even be that you need to specify `train_return_nodes` and `eval_return_nodes` separately.outputz1No output nodes found in graph_module.graph.nodescg|]}|Sr]r])r^rs rrz,create_feature_extractor..s555q555rrFrG)rr])-rrre ValueErrorrr+rrrr0rr#rr*rr&r rrror4rNrIrkeysrbrMrrrHr erase_nodergetcountjoinsplitrrinserting_afterreliminate_dead_code recompilerwr)rrrrrrrrrtracersgraphsmode_return_nodesrtracerrrA graph_moduleavailable_nodesorig_output_nodesrr output_nodesr depthrs @rrrNsj/}==M.K X X<1CEV"W X X XXX  q    d"'8D'@A  q   T !&8D&@ Aywxxx6c3h6666 !!z,// %l33$\22'Z(:;;&J'899G F=OYj3k3k!HH 7?? KKMMMM V^^ JJLLL 00-00 U- @@+5eRY+G+G[u''U^~fk5$?? v6==??@@ s?## $ $O(<(< < <k 't,1133  E]]]]_]]]^^  6e666 ,,233 , ,Atx!((+++  RPQQ Q" - -A   ) )! , , , ,65L.4555"}} %  A$599!<VZ[[[L KK {### sO55O9 <O9 )NFN)NNNNFN)(ryrrJrq collectionsrcopyr itertoolsrtypingrrrr r r r rUr{r rtorch.fx.graph_moduler__all__Tracerrr0rhrwrrr+rr*r,rrrrr]rrrs ######DDDDDDDDDDDDDDDDDD ,,,,,, &'= >:::::BI:::(`````*```FKKK(.(~((((>4: 8DcN3K PTUXZ]U]P^    &/3"'.2 >#># 9>#DcN+># >#DcN+ >#  49d3i  >#>#>#>#B?(?(?(?(?(bn?(?(?(H@DEIDH.2"'.2nn 9n5cDcN!:;<n!tCy$sCx.'@!ABn d3ic3h&? @A n DcN+ n  nDcN+n^nnnnnnr