.. note:: :class: sphx-glr-download-link-note Click :ref:`here ` to download the full example code .. rst-class:: sphx-glr-example-title .. _sphx_glr_intermediate_fx_profiling_tutorial.py: (beta) Building a Simple CPU Performance Profiler with FX ******************************************************* **Author**: `James Reed `_ In this tutorial, we are going to use FX to do the following: 1) Capture PyTorch Python code in a way that we can inspect and gather statistics about the structure and execution of the code 2) Build out a small class that will serve as a simple performance "profiler", collecting runtime statistics about each part of the model from actual runs. For this tutorial, we are going to use the torchvision ResNet18 model for demonstration purposes. .. code-block:: default import torch import torch.fx import torchvision.models as models rn18 = models.resnet18() rn18.eval() Now that we have our model, we want to inspect deeper into its performance. That is, for the following invocation, which parts of the model are taking the longest? .. code-block:: default input = torch.randn(5, 3, 224, 224) output = rn18(input) A common way of answering that question is to go through the program source, add code that collects timestamps at various points in the program, and compare the difference between those timestamps to see how long the regions between the timestamps take. That technique is certainly applicable to PyTorch code, however it would be nicer if we didn't have to copy over model code and edit it, especially code we haven't written (like this torchvision model). Instead, we are going to use FX to automate this "instrumentation" process without needing to modify any source. First, let's get some imports out of the way (we will be using all of these later in the code). .. code-block:: default import statistics, tabulate, time from typing import Any, Dict, List from torch.fx import Interpreter .. note:: ``tabulate`` is an external library that is not a dependency of PyTorch. We will be using it to more easily visualize performance data. Please make sure you've installed it from your favorite Python package source. Capturing the Model with Symbolic Tracing ----------------------------------------- Next, we are going to use FX's symbolic tracing mechanism to capture the definition of our model in a data structure we can manipulate and examine. .. code-block:: default traced_rn18 = torch.fx.symbolic_trace(rn18) print(traced_rn18.graph) .. rst-class:: sphx-glr-script-out Out: .. code-block:: none graph(): %x : torch.Tensor [#users=1] = placeholder[target=x] %conv1 : [#users=1] = call_module[target=conv1](args = (%x,), kwargs = {}) %bn1 : [#users=1] = call_module[target=bn1](args = (%conv1,), kwargs = {}) %relu : [#users=1] = call_module[target=relu](args = (%bn1,), kwargs = {}) %maxpool : [#users=2] = call_module[target=maxpool](args = (%relu,), kwargs = {}) %layer1_0_conv1 : [#users=1] = call_module[target=layer1.0.conv1](args = (%maxpool,), kwargs = {}) %layer1_0_bn1 : [#users=1] = call_module[target=layer1.0.bn1](args = (%layer1_0_conv1,), kwargs = {}) %layer1_0_relu : [#users=1] = call_module[target=layer1.0.relu](args = (%layer1_0_bn1,), kwargs = {}) %layer1_0_conv2 : [#users=1] = call_module[target=layer1.0.conv2](args = (%layer1_0_relu,), kwargs = {}) %layer1_0_bn2 : [#users=1] = call_module[target=layer1.0.bn2](args = (%layer1_0_conv2,), kwargs = {}) %add : [#users=1] = call_function[target=operator.add](args = (%layer1_0_bn2, %maxpool), kwargs = {}) %layer1_0_relu_1 : [#users=2] = call_module[target=layer1.0.relu](args = (%add,), kwargs = {}) %layer1_1_conv1 : [#users=1] = call_module[target=layer1.1.conv1](args = (%layer1_0_relu_1,), kwargs = {}) %layer1_1_bn1 : [#users=1] = call_module[target=layer1.1.bn1](args = (%layer1_1_conv1,), kwargs = {}) %layer1_1_relu : [#users=1] = call_module[target=layer1.1.relu](args = (%layer1_1_bn1,), kwargs = {}) %layer1_1_conv2 : [#users=1] = call_module[target=layer1.1.conv2](args = (%layer1_1_relu,), kwargs = {}) %layer1_1_bn2 : [#users=1] = call_module[target=layer1.1.bn2](args = (%layer1_1_conv2,), kwargs = {}) %add_1 : [#users=1] = call_function[target=operator.add](args = (%layer1_1_bn2, %layer1_0_relu_1), kwargs = {}) %layer1_1_relu_1 : [#users=2] = call_module[target=layer1.1.relu](args = (%add_1,), kwargs = {}) %layer2_0_conv1 : [#users=1] = call_module[target=layer2.0.conv1](args = (%layer1_1_relu_1,), kwargs = {}) %layer2_0_bn1 : [#users=1] = call_module[target=layer2.0.bn1](args = (%layer2_0_conv1,), kwargs = {}) %layer2_0_relu : [#users=1] = call_module[target=layer2.0.relu](args = (%layer2_0_bn1,), kwargs = {}) %layer2_0_conv2 : [#users=1] = call_module[target=layer2.0.conv2](args = (%layer2_0_relu,), kwargs = {}) %layer2_0_bn2 : [#users=1] = call_module[target=layer2.0.bn2](args = (%layer2_0_conv2,), kwargs = {}) %layer2_0_downsample_0 : [#users=1] = call_module[target=layer2.0.downsample.0](args = (%layer1_1_relu_1,), kwargs = {}) %layer2_0_downsample_1 : [#users=1] = call_module[target=layer2.0.downsample.1](args = (%layer2_0_downsample_0,), kwargs = {}) %add_2 : [#users=1] = call_function[target=operator.add](args = (%layer2_0_bn2, %layer2_0_downsample_1), kwargs = {}) %layer2_0_relu_1 : [#users=2] = call_module[target=layer2.0.relu](args = (%add_2,), kwargs = {}) %layer2_1_conv1 : [#users=1] = call_module[target=layer2.1.conv1](args = (%layer2_0_relu_1,), kwargs = {}) %layer2_1_bn1 : [#users=1] = call_module[target=layer2.1.bn1](args = (%layer2_1_conv1,), kwargs = {}) %layer2_1_relu : [#users=1] = call_module[target=layer2.1.relu](args = (%layer2_1_bn1,), kwargs = {}) %layer2_1_conv2 : [#users=1] = call_module[target=layer2.1.conv2](args = (%layer2_1_relu,), kwargs = {}) %layer2_1_bn2 : [#users=1] = call_module[target=layer2.1.bn2](args = (%layer2_1_conv2,), kwargs = {}) %add_3 : [#users=1] = call_function[target=operator.add](args = (%layer2_1_bn2, %layer2_0_relu_1), kwargs = {}) %layer2_1_relu_1 : [#users=2] = call_module[target=layer2.1.relu](args = (%add_3,), kwargs = {}) %layer3_0_conv1 : [#users=1] = call_module[target=layer3.0.conv1](args = (%layer2_1_relu_1,), kwargs = {}) %layer3_0_bn1 : [#users=1] = call_module[target=layer3.0.bn1](args = (%layer3_0_conv1,), kwargs = {}) %layer3_0_relu : [#users=1] = call_module[target=layer3.0.relu](args = (%layer3_0_bn1,), kwargs = {}) %layer3_0_conv2 : [#users=1] = call_module[target=layer3.0.conv2](args = (%layer3_0_relu,), kwargs = {}) %layer3_0_bn2 : [#users=1] = call_module[target=layer3.0.bn2](args = (%layer3_0_conv2,), kwargs = {}) %layer3_0_downsample_0 : [#users=1] = call_module[target=layer3.0.downsample.0](args = (%layer2_1_relu_1,), kwargs = {}) %layer3_0_downsample_1 : [#users=1] = call_module[target=layer3.0.downsample.1](args = (%layer3_0_downsample_0,), kwargs = {}) %add_4 : [#users=1] = call_function[target=operator.add](args = (%layer3_0_bn2, %layer3_0_downsample_1), kwargs = {}) %layer3_0_relu_1 : [#users=2] = call_module[target=layer3.0.relu](args = (%add_4,), kwargs = {}) %layer3_1_conv1 : [#users=1] = call_module[target=layer3.1.conv1](args = (%layer3_0_relu_1,), kwargs = {}) %layer3_1_bn1 : [#users=1] = call_module[target=layer3.1.bn1](args = (%layer3_1_conv1,), kwargs = {}) %layer3_1_relu : [#users=1] = call_module[target=layer3.1.relu](args = (%layer3_1_bn1,), kwargs = {}) %layer3_1_conv2 : [#users=1] = call_module[target=layer3.1.conv2](args = (%layer3_1_relu,), kwargs = {}) %layer3_1_bn2 : [#users=1] = call_module[target=layer3.1.bn2](args = (%layer3_1_conv2,), kwargs = {}) %add_5 : [#users=1] = call_function[target=operator.add](args = (%layer3_1_bn2, %layer3_0_relu_1), kwargs = {}) %layer3_1_relu_1 : [#users=2] = call_module[target=layer3.1.relu](args = (%add_5,), kwargs = {}) %layer4_0_conv1 : [#users=1] = call_module[target=layer4.0.conv1](args = (%layer3_1_relu_1,), kwargs = {}) %layer4_0_bn1 : [#users=1] = call_module[target=layer4.0.bn1](args = (%layer4_0_conv1,), kwargs = {}) %layer4_0_relu : [#users=1] = call_module[target=layer4.0.relu](args = (%layer4_0_bn1,), kwargs = {}) %layer4_0_conv2 : [#users=1] = call_module[target=layer4.0.conv2](args = (%layer4_0_relu,), kwargs = {}) %layer4_0_bn2 : [#users=1] = call_module[target=layer4.0.bn2](args = (%layer4_0_conv2,), kwargs = {}) %layer4_0_downsample_0 : [#users=1] = call_module[target=layer4.0.downsample.0](args = (%layer3_1_relu_1,), kwargs = {}) %layer4_0_downsample_1 : [#users=1] = call_module[target=layer4.0.downsample.1](args = (%layer4_0_downsample_0,), kwargs = {}) %add_6 : [#users=1] = call_function[target=operator.add](args = (%layer4_0_bn2, %layer4_0_downsample_1), kwargs = {}) %layer4_0_relu_1 : [#users=2] = call_module[target=layer4.0.relu](args = (%add_6,), kwargs = {}) %layer4_1_conv1 : [#users=1] = call_module[target=layer4.1.conv1](args = (%layer4_0_relu_1,), kwargs = {}) %layer4_1_bn1 : [#users=1] = call_module[target=layer4.1.bn1](args = (%layer4_1_conv1,), kwargs = {}) %layer4_1_relu : [#users=1] = call_module[target=layer4.1.relu](args = (%layer4_1_bn1,), kwargs = {}) %layer4_1_conv2 : [#users=1] = call_module[target=layer4.1.conv2](args = (%layer4_1_relu,), kwargs = {}) %layer4_1_bn2 : [#users=1] = call_module[target=layer4.1.bn2](args = (%layer4_1_conv2,), kwargs = {}) %add_7 : [#users=1] = call_function[target=operator.add](args = (%layer4_1_bn2, %layer4_0_relu_1), kwargs = {}) %layer4_1_relu_1 : [#users=1] = call_module[target=layer4.1.relu](args = (%add_7,), kwargs = {}) %avgpool : [#users=1] = call_module[target=avgpool](args = (%layer4_1_relu_1,), kwargs = {}) %flatten : [#users=1] = call_function[target=torch.flatten](args = (%avgpool, 1), kwargs = {}) %fc : [#users=1] = call_module[target=fc](args = (%flatten,), kwargs = {}) return fc This gives us a Graph representation of the ResNet18 model. A Graph consists of a series of Nodes connected to each other. Each Node represents a call-site in the Python code (whether to a function, a module, or a method) and the edges (represented as ``args`` and ``kwargs`` on each node) represent the values passed between these call-sites. More information about the Graph representation and the rest of FX's APIs ca be found at the FX documentation https://pytorch.org/docs/master/fx.html. Creating a Profiling Interpreter -------------------------------- Next, we are going to create a class that inherits from ``torch.fx.Interpreter``. Though the ``GraphModule`` that ``symbolic_trace`` produces compiles Python code that is run when you call a ``GraphModule``, an alternative way to run a ``GraphModule`` is by executing each ``Node`` in the ``Graph`` one by one. That is the functionality that ``Interpreter`` provides: It interprets the graph node- by-node. By inheriting from ``Interpreter``, we can override various functionality and install the profiling behavior we want. The goal is to have an object to which we can pass a model, invoke the model 1 or more times, then get statistics about how long the model and each part of the model took during those runs. Let's define our ``ProfilingInterpreter`` class: .. code-block:: default class ProfilingInterpreter(Interpreter): def __init__(self, mod : torch.nn.Module): # Rather than have the user symbolically trace their model, # we're going to do it in the constructor. As a result, the # user can pass in any ``Module`` without having to worry about # symbolic tracing APIs gm = torch.fx.symbolic_trace(mod) super().__init__(gm) # We are going to store away two things here: # # 1. A list of total runtimes for ``mod``. In other words, we are # storing away the time ``mod(...)`` took each time this # interpreter is called. self.total_runtime_sec : List[float] = [] # 2. A map from ``Node`` to a list of times (in seconds) that # node took to run. This can be seen as similar to (1) but # for specific sub-parts of the model. self.runtimes_sec : Dict[torch.fx.Node, List[float]] = {} ###################################################################### # Next, let's override our first method: ``run()``. ``Interpreter``'s ``run`` # method is the top-level entrypoint for execution of the model. We will # want to intercept this so that we can record the total runtime of the # model. def run(self, *args) -> Any: # Record the time we started running the model t_start = time.time() # Run the model by delegating back into Interpreter.run() return_val = super().run(*args) # Record the time we finished running the model t_end = time.time() # Store the total elapsed time this model execution took in the # ProfilingInterpreter self.total_runtime_sec.append(t_end - t_start) return return_val ###################################################################### # Now, let's override ``run_node``. ``Interpreter`` calls ``run_node`` each # time it executes a single node. We will intercept this so that we # can measure and record the time taken for each individual call in # the model. def run_node(self, n : torch.fx.Node) -> Any: # Record the time we started running the op t_start = time.time() # Run the op by delegating back into Interpreter.run_node() return_val = super().run_node(n) # Record the time we finished running the op t_end = time.time() # If we don't have an entry for this node in our runtimes_sec # data structure, add one with an empty list value. self.runtimes_sec.setdefault(n, []) # Record the total elapsed time for this single invocation # in the runtimes_sec data structure self.runtimes_sec[n].append(t_end - t_start) return return_val ###################################################################### # Finally, we are going to define a method (one which doesn't override # any ``Interpreter`` method) that provides us a nice, organized view of # the data we have collected. def summary(self, should_sort : bool = False) -> str: # Build up a list of summary information for each node node_summaries : List[List[Any]] = [] # Calculate the mean runtime for the whole network. Because the # network may have been called multiple times during profiling, # we need to summarize the runtimes. We choose to use the # arithmetic mean for this. mean_total_runtime = statistics.mean(self.total_runtime_sec) # For each node, record summary statistics for node, runtimes in self.runtimes_sec.items(): # Similarly, compute the mean runtime for ``node`` mean_runtime = statistics.mean(runtimes) # For easier understanding, we also compute the percentage # time each node took with respect to the whole network. pct_total = mean_runtime / mean_total_runtime * 100 # Record the node's type, name of the node, mean runtime, and # percent runtim node_summaries.append( [node.op, str(node), mean_runtime, pct_total]) # One of the most important questions to answer when doing performance # profiling is "Which op(s) took the longest?". We can make this easy # to see by providing sorting functionality in our summary view if should_sort: node_summaries.sort(key=lambda s: s[2], reverse=True) # Use the ``tabulate`` library to create a well-formatted table # presenting our summary information headers : List[str] = [ 'Op type', 'Op', 'Average runtime (s)', 'Pct total runtime' ] return tabulate.tabulate(node_summaries, headers=headers) .. note:: We use Python's ``time.time`` function to pull wall clock timestamps and compare them. This is not the most accurate way to measure performance, and will only give us a first- order approximation. We use this simple technique only for the purpose of demonstration in this tutorial. Investigating the Performance of ResNet18 ----------------------------------------- We can now use ``ProfilingInterpreter`` to inspect the performance characteristics of our ResNet18 model; .. code-block:: default interp = ProfilingInterpreter(rn18) interp.run(input) print(interp.summary(True)) .. rst-class:: sphx-glr-script-out Out: .. code-block:: none Op type Op Average runtime (s) Pct total runtime ------------- --------------------- --------------------- ------------------- call_module maxpool 0.018712 9.38134 call_module conv1 0.0169921 8.51904 call_module layer1_1_conv1 0.0113249 5.67776 call_module layer4_0_conv2 0.0112648 5.64764 call_module layer1_0_conv2 0.0108731 5.45125 call_module layer1_1_conv2 0.0101438 5.0856 call_module layer4_1_conv1 0.00981736 4.92196 call_module layer4_1_conv2 0.00953436 4.78008 call_module layer2_1_conv1 0.00913215 4.57843 call_module layer2_1_conv2 0.0090754 4.54998 call_module layer1_0_conv1 0.00896072 4.49249 call_module layer3_1_conv2 0.0084939 4.25844 call_module layer3_0_conv2 0.0079031 3.96224 call_module layer2_0_conv2 0.00763893 3.8298 call_module layer3_1_conv1 0.00758338 3.80195 call_module layer2_0_conv1 0.00704288 3.53097 call_module layer4_0_conv1 0.00657892 3.29836 call_module bn1 0.00621057 3.11369 call_module layer3_0_conv1 0.00589919 2.95758 call_module layer2_0_downsample_0 0.0027411 1.37426 call_function add_1 0.00153208 0.768112 call_function add 0.0014267 0.715279 call_module layer4_0_downsample_0 0.000902414 0.452428 call_function add_3 0.000860929 0.43163 call_module layer3_0_downsample_0 0.000832081 0.417166 call_module relu 0.000778675 0.390391 call_module layer1_0_bn2 0.000358582 0.179776 call_module layer1_1_bn2 0.000343323 0.172126 call_module layer1_0_bn1 0.000320673 0.16077 call_module layer1_1_bn1 0.000314236 0.157543 call_module fc 0.000309467 0.155152 call_module layer2_0_downsample_1 0.000267029 0.133876 call_module layer2_0_bn1 0.000218153 0.109372 call_module layer2_1_bn1 0.000216246 0.108415 call_module layer1_0_relu 0.000208378 0.104471 call_module layer1_1_relu_1 0.000206709 0.103634 call_module layer2_1_bn2 0.000200272 0.100407 call_module layer2_0_bn2 0.000189543 0.0950278 call_function add_2 0.000189066 0.0947888 call_module layer1_0_relu_1 0.000188589 0.0945497 call_module layer1_1_relu 0.000163078 0.0817598 call_module layer3_1_bn2 0.000156879 0.078652 call_module layer3_0_downsample_1 0.000145674 0.073034 call_module layer4_1_bn1 0.000141382 0.0708824 call_module layer4_0_bn2 0.000140905 0.0706433 call_module layer3_1_bn1 0.000138521 0.069448 call_module layer2_1_relu_1 0.000137806 0.0690894 call_module layer3_0_bn2 0.000137806 0.0690894 call_module avgpool 0.00013566 0.0680136 call_module layer4_1_bn2 0.000135422 0.0678941 call_module layer4_0_bn1 0.000133276 0.0668183 call_module layer3_0_bn1 0.000130892 0.065623 call_module layer4_0_downsample_1 0.000130653 0.0655035 call_module layer2_0_relu 0.000113487 0.0568972 call_module layer2_0_relu_1 0.000107527 0.0539089 call_function add_4 0.000106335 0.0533112 call_function add_5 0.000103951 0.0521159 call_module layer2_1_relu 0.000100613 0.0504424 call_module layer3_0_relu_1 7.86781e-05 0.0394455 call_module layer3_0_relu 7.67708e-05 0.0384893 call_function add_6 7.29561e-05 0.0365768 call_module layer3_1_relu_1 7.27177e-05 0.0364572 call_module layer3_1_relu 6.86646e-05 0.0344252 call_function add_7 6.69956e-05 0.0335885 call_module layer4_0_relu_1 6.53267e-05 0.0327517 call_module layer4_1_relu_1 6.12736e-05 0.0307197 call_module layer4_1_relu 5.4121e-05 0.0271337 call_module layer4_0_relu 5.31673e-05 0.0266556 call_function flatten 3.05176e-05 0.0153001 output output 1.81198e-05 0.00908442 placeholder x 1.21593e-05 0.00609613 There are two things we should call out here: * MaxPool2d takes up the most time. This is a known issue: https://github.com/pytorch/pytorch/issues/51393 * BatchNorm2d also takes up significant time. We can continue this line of thinking and optimize this in the Conv-BN Fusion with FX `tutorial `_. Conclusion ---------- As we can see, using FX we can easily capture PyTorch programs (even ones we don't have the source code for!) in a machine-interpretable format and use that for analysis, such as the performance analysis we've done here. FX opens up an exiciting world of possibilities for working with PyTorch programs. Finally, since FX is still in beta, we would be happy to hear any feedback you have about using it. Please feel free to use the PyTorch Forums (https://discuss.pytorch.org/) and the issue tracker (https://github.com/pytorch/pytorch/issues) to provide any feedback you might have. .. rst-class:: sphx-glr-timing **Total running time of the script:** ( 0 minutes 0.693 seconds) .. _sphx_glr_download_intermediate_fx_profiling_tutorial.py: .. only :: html .. container:: sphx-glr-footer :class: sphx-glr-footer-example .. container:: sphx-glr-download :download:`Download Python source code: fx_profiling_tutorial.py ` .. container:: sphx-glr-download :download:`Download Jupyter notebook: fx_profiling_tutorial.ipynb ` .. only:: html .. rst-class:: sphx-glr-signature `Gallery generated by Sphinx-Gallery `_