From cff62bdbd20bab3ebcc27213be1e6d79da00c842 Mon Sep 17 00:00:00 2001 From: Zhi <5145158+zhiics@users.noreply.github.com> Date: Sun, 10 Nov 2019 21:57:43 -0800 Subject: [PATCH] [tutorial] Relay pass infra tutorial (#4083) MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit * Add pass manager tutorial * fix some examples * retrigger ci * Update tutorials/dev/relay_pass_infra.py Co-Authored-By: 雾雨魔理沙 * Add ToANormalForm link --- include/tvm/relay/transform.h | 4 +- python/tvm/relay/transform.py | 9 +- src/relay/pass/print_ir.cc | 4 +- tutorials/dev/relay_pass_infra.py | 255 ++++++++++++++++++++++++++++++++++++++ 4 files changed, 267 insertions(+), 5 deletions(-) create mode 100644 tutorials/dev/relay_pass_infra.py diff --git a/include/tvm/relay/transform.h b/include/tvm/relay/transform.h index 82144d7..10de087 100644 --- a/include/tvm/relay/transform.h +++ b/include/tvm/relay/transform.h @@ -567,9 +567,11 @@ TVM_DLL Pass EtaExpand(); /*! * \brief Print the IR for a module to help debugging. * + * \param show_meta_data The flag to control if meta data needs to be printed. + * * \return the pass. */ -TVM_DLL Pass PrintIR(); +TVM_DLL Pass PrintIR(bool show_meta_data = true); } // namespace transform diff --git a/python/tvm/relay/transform.py b/python/tvm/relay/transform.py index ad1aecf..d3509da 100644 --- a/python/tvm/relay/transform.py +++ b/python/tvm/relay/transform.py @@ -594,16 +594,21 @@ def LambdaLift(): return _transform.LambdaLift() -def PrintIR(): +def PrintIR(show_meta_data=True): """ Print the IR for a module to help debugging. + Parameters + ---------- + show_meta_data : bool + A boolean flag to indicate if meta data should be printed. + Returns ------- ret : tvm.relay.Pass The registered pass that prints the module IR. """ - return _transform.PrintIR() + return _transform.PrintIR(show_meta_data) def gradient(expr, mod=None, mode='higher_order'): diff --git a/src/relay/pass/print_ir.cc b/src/relay/pass/print_ir.cc index e244547..5542d26 100644 --- a/src/relay/pass/print_ir.cc +++ b/src/relay/pass/print_ir.cc @@ -32,10 +32,10 @@ namespace relay { namespace transform { -Pass PrintIR() { +Pass PrintIR(bool show_meta_data) { runtime::TypedPackedFunc pass_func = [=](Module m, PassContext pc) { - LOG(INFO) << "Dumping the module IR: " << std::endl << AsText(m); + LOG(INFO) << "Dumping the module IR: " << std::endl << AsText(m, show_meta_data); return m; }; return CreateModulePass(pass_func, 0, "PrintIR", {}); diff --git a/tutorials/dev/relay_pass_infra.py b/tutorials/dev/relay_pass_infra.py new file mode 100644 index 0000000..2a2d1f5 --- /dev/null +++ b/tutorials/dev/relay_pass_infra.py @@ -0,0 +1,255 @@ +# Licensed to the Apache Software Foundation (ASF) under one +# or more contributor license agreements. See the NOTICE file +# distributed with this work for additional information +# regarding copyright ownership. The ASF licenses this file +# to you under the Apache License, Version 2.0 (the +# "License"); you may not use this file except in compliance +# with the License. You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, +# software distributed under the License is distributed on an +# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY +# KIND, either express or implied. See the License for the +# specific language governing permissions and limitations +# under the License. +# pylint: disable=line-too-long +""" +.. _tutorial-relay-pass-infra + +How to Use Relay Pass Infra +=========================== +**Author**: `Zhi Chen `_ + +As the number of optimization passes increases in Relay, it becomes intractable to +execute them and maintain their dependencies manually. Therefore, we have +introduced an infrastructure to manage the optimization passes. + +The optimizations of a Relay program could be applied at various granularity, +namely function-level and module-level using `FunctionPass`_ and `ModulePass`_ +respectively. Or users can rely on `Sequential`_ to apply a sequence of passes +on a Relay program where the dependencies between passes can be resolved by the +pass infra. For more details about each type of these passes, please refer to +the `pass infra doc`_. + +This tutorial demostrates how developers can use the Relay pass infra to perform +a certain optimization and create an optimization pipeline. + +.. _FunctionPass: https://docs.tvm.ai/api/python/relay/transform.html#tvm.relay.transform.FunctionPass + +.. _ModulePass: https://docs.tvm.ai/api/python/relay/transform.html#tvm.relay.transform.ModulePass + +.. _Sequential: https://docs.tvm.ai/api/python/relay/transform.html#tvm.relay.transform.Sequential + +.. _pass infra doc: https://docs.tvm.ai/dev/relay_pass_infra.html + +.. _ToANormalForm: https://docs.tvm.ai/api/python/relay/transform.html#tvm.relay.transform.ToANormalForm +""" + +import numpy as np +import tvm +import tvm.relay as relay + +############################################################################### +# Create An Example Relay Program +# ------------------------------- +# First of all, we create a simple Relay program for the tutorial. This program +# will be used by various optimizations of the examples in this tutorial. + +def example(): + shape = (1, 64, 54, 54) + c_data = np.empty(shape).astype("float32") + c = relay.const(c_data) + weight = relay.var('weight', shape=(64, 64, 3, 3)) + x = relay.var("x", relay.TensorType((1, 64, 56, 56), "float32")) + conv = relay.nn.conv2d(x, weight) + y = relay.add(c, c) + y = relay.multiply(y, relay.const(2, "float32")) + y = relay.add(conv, y) + z = relay.add(y, c) + z1 = relay.add(y, c) + z2 = relay.add(z, z1) + return relay.Function([x], z2) + +############################################################################### +# Let us register layout alteration for a conv2d op so that we can apply the +# layout alteration pass on the example. How alter layout pass works is out +# the scope of this tutorial. + +@relay.op.register_alter_op_layout("nn.conv2d", level=101) +def alter_conv2d(attrs, inputs, tinfos): + data, weight = inputs + new_attrs = dict(attrs) + new_attrs['data_layout'] = 'NCHW16c' + return relay.nn.conv2d(data, weight, **new_attrs) + +############################################################################### +# Optimize the Program +# -------------------- +# Now we would like to optimize the program. Relay features a host of +# optimizations. We will select some of them to apply on this example program. +# +# There are multiple ways to optimize a Relay program. Below we will provide +# examples for each of them. +# +# Manually Apply Optimization Passes +# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ + +# Let's first create a relay Module which contains one or multiple Relay +# functions for optimization. +f = example() +mod = relay.Module.from_expr(f) + +# Now we can apply constant folding on the module. +# fold_const here is a callback that doesn't take any parameters. +fold_const = relay.transform.FoldConstant() +# Then, we can invoke the pass on the given module. Note that the constant +# folding pass works at the function-level. That being said, each function in +# the module will be applied with the optimization. Users don't need to iterate +# through individual functions manually to apply this pass. +mod = fold_const(mod) +# We can see from the updated program that the constants are folded. +print(mod) + +############################################################################### +# More optimizations can be applied in the similar manner. For instance, we can +# eliminate the common expressions that used by `z` and `z1`. +mod = relay.transform.EliminateCommonSubexpr()(mod) +print(mod) + +############################################################################### +# Some optimizations, such as fusion, are parameteric as well. For example, +# opt level 0 will not allow operators to be fused together. Users can pass the +# `fuse_opt_level` to enable this. +mod = relay.transform.FuseOps(fuse_opt_level=0)(mod) + +# We can observe that the optimized module contains functions that only have +# a signle primitive op. +print(mod) + +############################################################################### +# Use `Sequential`_ to Apply a Sequence of Passes +# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +# Applying passes as above is actually tedious and it may require users to have +# better understanding about the dependencies between them. For example, fusion +# currently doesn't work well on let bindings. Therefore, we would not be able +# to fuse operators that were fusable if `ToANormalForm`_ is applied before +# fusion, as this pass generates let bindings for each expression to +# canonicalize a Relay program. +# +# Relay, hence, provides `Sequential`_ to alleviate developers from handling +# these issues explicitly by specifying the required passes of each pass and +# packing them as a whole to execute. For example, the same passes can now be +# applied using the sequential style as the following. `Sequential`_ is +# similiar to `torch.nn.sequential `_ +# and `mxnet.gluon.block `_. +# For example, `torch.nn.sequential` is used to contain a sequence of PyTorch +# `Modules` that will be added to build a network. It focuses on the network +# layers. Instead, the `Sequential`_ in our pass infra works on the optimizing +# pass. + +# Now let's execute some passes through `Sequential`_ +f = example() +mod = relay.Module.from_expr(f) +# Glob the interested passes. +seq = relay.transform.Sequential([relay.transform.FoldConstant(), + relay.transform.EliminateCommonSubexpr(), + relay.transform.FuseOps(fuse_opt_level=2)]) +mod1 = seq(mod) +print(mod1) + +############################################################################### +# From the transformed Relay program, we can see that there are still two +# identical addition operations. This is because `EliminateCommonSubexpr` +# was not actually performed. The reason is because only the passes that have +# optimization level less or equal to 2 will be executed by default under +# `Sequential`_. The pass infra, however, provides a configuration interface +# for users to customize the optimization level that they want to execute. + +with relay.build_config(opt_level=3): + mod2 = seq(mod) +print(mod2) + +############################################################################### +# Now we can see that only one of the two identical additions is kept. +# +# In addition, users can selectively disable some passes using the +# `disabled_pass` config, which is similar to the `-fno-xxx` option used the +# general purpose compilers, such as Clang and GCC. For example, we can disable +# EliminateCommonSubexpr as following. The printed module will again show two +# identical addition operations. + +with relay.build_config(opt_level=3, disabled_pass=["EliminateCommonSubexpr"]): + mod3 = seq(mod) +print(mod3) + +############################################################################### +# The passes applied so far are target independent. The pass infra also +# provides a means to make pass target-aware. For example, the layout +# alteration pass falls in such category. + +with relay.build_config(opt_level=3): + mod4 = seq(mod) +print(mod4) + +seq1 = relay.transform.Sequential([relay.transform.AlterOpLayout()]) +with relay.build_config(opt_level=3): + with tvm.target.create("llvm"): + mod5 = seq1(mod) +print(mod5) + +############################################################################## +# Implement a Pass Using Python Decorator +# ------------------------------------------ +# The next example illustrates how we can orchestrate a customized optimization +# pipeline through the pass infra using Python decorators. This functionality +# greatly eases the implementation of passes. For example, users can simply +# define a decorated class to do function-level optimizations as the following +# example shows. `transform_function` wraps a class to replace all constants +# with a multiple of `c`. Later on, each function in a given module will be +# visited and each constant in the function will be replaced when we invoke the +# customized pass. + +@relay.transform.function_pass(opt_level=1) +class CustomPipeline: + """Simple test function to replace one argument to another.""" + + def __init__(self, multiplier): + self.multiplier = multiplier + + # This function can define a pass. + def transform_function(self, func, mod, ctx): + obj = self + + class ReplaceConstant(tvm.relay.ExprMutator): + def visit_const(self, c): + return relay.multiply(obj.multiplier, c) + return ReplaceConstant().visit(func) + +f = example() +mod = relay.Module.from_expr(f) +custom_pass = CustomPipeline(multiplier=relay.const(3, "float")) +assert custom_pass.info.name == "CustomPipeline" +mod3 = custom_pass(mod) +print(mod3) + +############################################################################## +# Debug a Pass +# ------------ +# Relay provides users a plug-and-play style debugging pass that print the IR +# after a certain pass is done. For example, we can print out the IR on the +# completion of constant folding and fusion by adding the debugging pass after +# them. + +f = example() +mod = relay.Module.from_expr(f) +seq = relay.transform.Sequential([relay.transform.FoldConstant(), + relay.transform.PrintIR(), + relay.transform.EliminateCommonSubexpr(), + relay.transform.FuseOps(), + relay.transform.PrintIR()]) +with relay.build_config(opt_level=3): + mod = seq(mod) + +print("done") -- 2.7.4