在 Relay 中使用管道执行器#

原作者: Hua Jiang

这是关于如何在 Relay 中使用“管道执行器”（Pipeline Executor）的简短教程。

import env # 加载 TVM
import tvm
from tvm import te
import numpy as np
from tvm.contrib import graph_executor as runtime
from tvm.relay.op.contrib.cutlass import partition_for_cutlass
from tvm import relay
from tvm.relay import testing
import tvm.testing
from tvm.contrib.cutlass import (
    has_cutlass,
    num_cutlass_partitions,
    finalize_modules,
    finalize_modules_vm,
)

img_size = 8

/media/pc/data/4tb/lxw/home/lxw/tvm/xinetzone/src

创建简单网络，可以是预训练的模型#

创建非常简单的网络进行演示。它由卷积、batch normalization、dense 和 ReLU 激活组成。

def get_network():
    out_channels = 16
    batch_size = 1
    data = relay.var("data", relay.TensorType((batch_size, 3, img_size, img_size), "float16"))
    dense_weight = relay.var(
        "dweight", relay.TensorType((batch_size, 16 * img_size * img_size), "float16")
    )
    weight = relay.var("weight")
    second_weight = relay.var("second_weight")
    bn_gamma = relay.var("bn_gamma")
    bn_beta = relay.var("bn_beta")
    bn_mmean = relay.var("bn_mean")
    bn_mvar = relay.var("bn_var")
    simple_net = relay.nn.conv2d(
        data=data, weight=weight, kernel_size=(3, 3), channels=out_channels, padding=(1, 1)
    )
    simple_net = relay.nn.batch_norm(simple_net, bn_gamma, bn_beta, bn_mmean, bn_mvar)[0]
    simple_net = relay.nn.relu(simple_net)
    simple_net = relay.nn.batch_flatten(simple_net)
    simple_net = relay.nn.dense(simple_net, dense_weight)
    simple_net = relay.Function(relay.analysis.free_vars(simple_net), simple_net)
    data_shape = (batch_size, 3, img_size, img_size)
    net, params = testing.create_workload(simple_net)
    return net, params, data_shape


net, params, data_shape = get_network()

将网络分成两个子图#

单元测试中的 ‘graph_split’ 函数只是一个例子。用户可以创建自定义的逻辑来分割计算图。

import os
from env import TVM_ROOT
os.sys.path.append(f"{TVM_ROOT}/tests/python/relay")

from test_pipeline_executor import graph_split

将网络分成两个子图。

split_config = [{"op_name": "nn.relu", "op_index": 0}]
subgraphs = graph_split(net["main"], split_config, params)

生成的子图应该如下所示。

for k, m in enumerate(subgraphs):
    print(f"子图 {k}:\n", m["main"])

子图 0:
 fn (%data: Tensor[(1, 3, 8, 8), float16] /* ty=Tensor[(1, 3, 8, 8), float16] */) {
  %0 = nn.conv2d(%data, meta[relay.Constant][0] /* ty=Tensor[(16, 3, 3, 3), float16] */, padding=[1, 1, 1, 1], channels=16, kernel_size=[3, 3]) /* ty=Tensor[(1, 16, 8, 8), float16] */;
  %1 = nn.batch_norm(%0, meta[relay.Constant][1] /* ty=Tensor[(16), float16] */, meta[relay.Constant][2] /* ty=Tensor[(16), float16] */, meta[relay.Constant][3] /* ty=Tensor[(16), float16] */, meta[relay.Constant][4] /* ty=Tensor[(16), float16] */) /* ty=(Tensor[(1, 16, 8, 8), float16], Tensor[(16), float16], Tensor[(16), float16]) */;
  %2 = %1.0 /* ty=Tensor[(1, 16, 8, 8), float16] */;
  nn.relu(%2) /* ty=Tensor[(1, 16, 8, 8), float16] */
}

子图 1:
 fn (%data_n_0: Tensor[(1, 16, 8, 8), float16] /* ty=Tensor[(1, 16, 8, 8), float16] */) {
  %0 = nn.batch_flatten(%data_n_0) /* ty=Tensor[(1, 1024), float16] */;
  nn.dense(%0, meta[relay.Constant][0] /* ty=Tensor[(1, 1024), float16] */, units=None) /* ty=Tensor[(1, 1), float16] */
}

使用 cutlass target 构建子图#

cutlass = tvm.target.Target(
    {
        "kind": "cutlass",
        "sm": int(tvm.target.Target("cuda").arch.split("_")[1]),
        "use_3xtf32": True,
        "split_k_slices": [1],
        "profile_all_alignments": False,
        "find_first_valid": True,
        "use_multiprocessing": True,
        "use_fast_math": False,
        "tmp_dir": "./tmp",
    },
    host=tvm.target.Target("llvm"),
)


def cutlass_build(mod, target, params=None, target_host=None, mod_name="default"):
    target = [target, cutlass]
    lib = relay.build_module.build(
        mod, target=target, params=params, target_host=target_host, mod_name=mod_name
    )
    return lib

使用管道执行器在管道中运行两个子图#

在 cmake 中将 USE_PIPELINE_EXECUTOR 设置为 ON，并将 USE_CUTLASS 设置为 ON。

from tvm.contrib import graph_executor, pipeline_executor, pipeline_executor_build

Create subgraph pipeline configuration. Associate a subgraph module with a target. Use CUTLASS BYOC to build the second subgraph module.

mod0, mod1 = subgraphs[0], subgraphs[1]
# Use cutlass as the codegen.
mod1 = partition_for_cutlass(mod1)

Get the pipeline executor configuration object.

pipe_config = pipeline_executor_build.PipelineConfig()

Set the compile target of the subgraph module.

pipe_config[mod0].target = "llvm"
pipe_config[mod0].dev = tvm.cpu(0)

Set the compile target of the second subgraph module as cuda.

pipe_config[mod1].target = "cuda"
pipe_config[mod1].dev = tvm.device("cuda", 0)
pipe_config[mod1].build_func = cutlass_build
pipe_config[mod1].export_cc = "nvcc"
# Create the pipeline by connecting the subgraph modules.
# The global input will be forwarded to the input interface of the first module named mod0
pipe_config["input"]["data"].connect(pipe_config[mod0]["input"]["data"])
# The first output of mod0 will be forwarded to the input interface of mod1
pipe_config[mod0]["output"][0].connect(pipe_config[mod1]["input"]["data_n_0"])
# The first output of mod1 will be the first global output.
pipe_config[mod1]["output"][0].connect(pipe_config["output"][0])

The pipeline configuration as below.

"""
print(pipe_config)
 Inputs
  |data: mod0:data

 output
  |output(0) : mod1.output(0)

 connections
  |mod0.output(0)-> mod1.data_n_0
"""

'\nprint(pipe_config)\n Inputs\n  |data: mod0:data\n\n output\n  |output(0) : mod1.output(0)\n\n connections\n  |mod0.output(0)-> mod1.data_n_0\n'

Build the pipeline executor.#

with tvm.transform.PassContext(opt_level=3):
    pipeline_mod_factory = pipeline_executor_build.build(pipe_config)

/media/pc/data/4tb/lxw/home/lxw/tvm/python/tvm/driver/build_module.py:266: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
  warnings.warn(
One or more operators have not been tuned. Please tune your model for better performance. Use DEBUG logging level to see more details.

Export the parameter configuration to a file.

directory_path = tvm.contrib.utils.tempdir().temp_dir
os.makedirs(directory_path, exist_ok=True)
config_file_name = pipeline_mod_factory.export_library(directory_path)

Use the load function to create and initialize PipelineModule.#

pipeline_module = pipeline_executor.PipelineModule.load_library(config_file_name)

Run the pipeline executor.#

Allocate input data.

data = np.random.uniform(-1, 1, size=data_shape).astype("float16")
pipeline_module.set_input("data", tvm.nd.array(data))

Run the two subgraph in the pipeline mode to get the output asynchronously or synchronously. In the following example, it is synchronous.

pipeline_module.run()
outputs = pipeline_module.get_output()

Use graph_executor for verification.#

Run these two subgraphs in sequence with graph_executor to get the output.

target = "llvm"
dev0 = tvm.device(target, 0)
lib0 = relay.build_module.build(mod0, target, params=params)
module0 = runtime.GraphModule(lib0["default"](dev0))
cuda = tvm.target.Target("cuda", host=tvm.target.Target("llvm"))
lib1 = relay.build_module.build(mod1, [cuda, cutlass], params=params)
lib1 = finalize_modules(lib1, "compile.so", "./tmp")

dev1 = tvm.device("cuda", 0)

module1 = runtime.GraphModule(lib1["default"](dev1))

module0.set_input("data", data)
module0.run()
out_shape = (1, 16, img_size, img_size)
out = module0.get_output(0, tvm.nd.empty(out_shape, "float16"))
module1.set_input("data_n_0", out)
module1.run()
out_shape = (1, 1)
out = module1.get_output(0, tvm.nd.empty(out_shape, "float16"))

/media/pc/data/4tb/lxw/home/lxw/tvm/python/tvm/driver/build_module.py:266: UserWarning: target_host parameter is going to be deprecated. Please pass in tvm.target.Target(target, host=target_host) instead.
  warnings.warn(

Verify the result.

tvm.testing.assert_allclose(outputs[0].numpy(), out.numpy())

TVM 文档

在 Relay 中使用管道执行器

导航