How to Use TVM Pass Instrument

Author: Chi-Wei Wang

As more and more passes are implemented, it becomes useful to instrument pass execution, analyze per-pass effects, and observe various events.

We can instrument passes by providing a list of tvm.ir.instrument.PassInstrument instances to tvm.transform.PassContext. We provide a pass instrument for collecting timing information (tvm.ir.instrument.PassTimingInstrument), but an extension mechanism is available via the tvm.instrument.pass_instrument() decorator.

This tutorial demonstrates how developers can use PassContext to instrument passes. Please also refer to the Pass Infrastructure.

import tvm
import tvm.relay as relay
from tvm.relay.testing import resnet
from tvm.contrib.download import download_testdata
from tvm.relay.build_module import bind_params_by_name
from tvm.ir.instrument import (
    PassTimingInstrument,
    pass_instrument,
)

Create An Example Relay Program

We use pre-defined resnet-18 network in Relay.

batch_size = 1
num_of_image_class = 1000
image_shape = (3, 224, 224)
output_shape = (batch_size, num_of_image_class)
relay_mod, relay_params = resnet.get_workload(num_layers=18, batch_size=1, image_shape=image_shape)
print("Printing the IR module...")
print(relay_mod.astext(show_meta_data=False))

Printing the IR module...
#[version = "0.0.5"]
def @main(%data: Tensor[(1, 3, 224, 224), float32] /* ty=Tensor[(1, 3, 224, 224), float32] */, %bn_data_gamma: Tensor[(3), float32] /* ty=Tensor[(3), float32] */, %bn_data_beta: Tensor[(3), float32] /* ty=Tensor[(3), float32] */, %bn_data_moving_mean: Tensor[(3), float32] /* ty=Tensor[(3), float32] */, %bn_data_moving_var: Tensor[(3), float32] /* ty=Tensor[(3), float32] */, %conv0_weight: Tensor[(64, 3, 7, 7), float32] /* ty=Tensor[(64, 3, 7, 7), float32] */, %bn0_gamma: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %bn0_beta: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %bn0_moving_mean: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %bn0_moving_var: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage1_unit1_bn1_gamma: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage1_unit1_bn1_beta: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage1_unit1_bn1_moving_mean: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage1_unit1_bn1_moving_var: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage1_unit1_conv1_weight: Tensor[(64, 64, 3, 3), float32] /* ty=Tensor[(64, 64, 3, 3), float32] */, %stage1_unit1_bn2_gamma: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage1_unit1_bn2_beta: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage1_unit1_bn2_moving_mean: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage1_unit1_bn2_moving_var: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage1_unit1_conv2_weight: Tensor[(64, 64, 3, 3), float32] /* ty=Tensor[(64, 64, 3, 3), float32] */, %stage1_unit1_sc_weight: Tensor[(64, 64, 1, 1), float32] /* ty=Tensor[(64, 64, 1, 1), float32] */, %stage1_unit2_bn1_gamma: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage1_unit2_bn1_beta: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage1_unit2_bn1_moving_mean: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage1_unit2_bn1_moving_var: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage1_unit2_conv1_weight: Tensor[(64, 64, 3, 3), float32] /* ty=Tensor[(64, 64, 3, 3), float32] */, %stage1_unit2_bn2_gamma: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage1_unit2_bn2_beta: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage1_unit2_bn2_moving_mean: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage1_unit2_bn2_moving_var: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage1_unit2_conv2_weight: Tensor[(64, 64, 3, 3), float32] /* ty=Tensor[(64, 64, 3, 3), float32] */, %stage2_unit1_bn1_gamma: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage2_unit1_bn1_beta: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage2_unit1_bn1_moving_mean: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage2_unit1_bn1_moving_var: Tensor[(64), float32] /* ty=Tensor[(64), float32] */, %stage2_unit1_conv1_weight: Tensor[(128, 64, 3, 3), float32] /* ty=Tensor[(128, 64, 3, 3), float32] */, %stage2_unit1_bn2_gamma: Tensor[(128), float32] /* ty=Tensor[(128), float32] */, %stage2_unit1_bn2_beta: Tensor[(128), float32] /* ty=Tensor[(128), float32] */, %stage2_unit1_bn2_moving_mean: Tensor[(128), float32] /* ty=Tensor[(128), float32] */, %stage2_unit1_bn2_moving_var: Tensor[(128), float32] /* ty=Tensor[(128), float32] */, %stage2_unit1_conv2_weight: Tensor[(128, 128, 3, 3), float32] /* ty=Tensor[(128, 128, 3, 3), float32] */, %stage2_unit1_sc_weight: Tensor[(128, 64, 1, 1), float32] /* ty=Tensor[(128, 64, 1, 1), float32] */, %stage2_unit2_bn1_gamma: Tensor[(128), float32] /* ty=Tensor[(128), float32] */, %stage2_unit2_bn1_beta: Tensor[(128), float32] /* ty=Tensor[(128), float32] */, %stage2_unit2_bn1_moving_mean: Tensor[(128), float32] /* ty=Tensor[(128), float32] */, %stage2_unit2_bn1_moving_var: Tensor[(128), float32] /* ty=Tensor[(128), float32] */, %stage2_unit2_conv1_weight: Tensor[(128, 128, 3, 3), float32] /* ty=Tensor[(128, 128, 3, 3), float32] */, %stage2_unit2_bn2_gamma: Tensor[(128), float32] /* ty=Tensor[(128), float32] */, %stage2_unit2_bn2_beta: Tensor[(128), float32] /* ty=Tensor[(128), float32] */, %stage2_unit2_bn2_moving_mean: Tensor[(128), float32] /* ty=Tensor[(128), float32] */, %stage2_unit2_bn2_moving_var: Tensor[(128), float32] /* ty=Tensor[(128), float32] */, %stage2_unit2_conv2_weight: Tensor[(128, 128, 3, 3), float32] /* ty=Tensor[(128, 128, 3, 3), float32] */, %stage3_unit1_bn1_gamma: Tensor[(128), float32] /* ty=Tensor[(128), float32] */, %stage3_unit1_bn1_beta: Tensor[(128), float32] /* ty=Tensor[(128), float32] */, %stage3_unit1_bn1_moving_mean: Tensor[(128), float32] /* ty=Tensor[(128), float32] */, %stage3_unit1_bn1_moving_var: Tensor[(128), float32] /* ty=Tensor[(128), float32] */, %stage3_unit1_conv1_weight: Tensor[(256, 128, 3, 3), float32] /* ty=Tensor[(256, 128, 3, 3), float32] */, %stage3_unit1_bn2_gamma: Tensor[(256), float32] /* ty=Tensor[(256), float32] */, %stage3_unit1_bn2_beta: Tensor[(256), float32] /* ty=Tensor[(256), float32] */, %stage3_unit1_bn2_moving_mean: Tensor[(256), float32] /* ty=Tensor[(256), float32] */, %stage3_unit1_bn2_moving_var: Tensor[(256), float32] /* ty=Tensor[(256), float32] */, %stage3_unit1_conv2_weight: Tensor[(256, 256, 3, 3), float32] /* ty=Tensor[(256, 256, 3, 3), float32] */, %stage3_unit1_sc_weight: Tensor[(256, 128, 1, 1), float32] /* ty=Tensor[(256, 128, 1, 1), float32] */, %stage3_unit2_bn1_gamma: Tensor[(256), float32] /* ty=Tensor[(256), float32] */, %stage3_unit2_bn1_beta: Tensor[(256), float32] /* ty=Tensor[(256), float32] */, %stage3_unit2_bn1_moving_mean: Tensor[(256), float32] /* ty=Tensor[(256), float32] */, %stage3_unit2_bn1_moving_var: Tensor[(256), float32] /* ty=Tensor[(256), float32] */, %stage3_unit2_conv1_weight: Tensor[(256, 256, 3, 3), float32] /* ty=Tensor[(256, 256, 3, 3), float32] */, %stage3_unit2_bn2_gamma: Tensor[(256), float32] /* ty=Tensor[(256), float32] */, %stage3_unit2_bn2_beta: Tensor[(256), float32] /* ty=Tensor[(256), float32] */, %stage3_unit2_bn2_moving_mean: Tensor[(256), float32] /* ty=Tensor[(256), float32] */, %stage3_unit2_bn2_moving_var: Tensor[(256), float32] /* ty=Tensor[(256), float32] */, %stage3_unit2_conv2_weight: Tensor[(256, 256, 3, 3), float32] /* ty=Tensor[(256, 256, 3, 3), float32] */, %stage4_unit1_bn1_gamma: Tensor[(256), float32] /* ty=Tensor[(256), float32] */, %stage4_unit1_bn1_beta: Tensor[(256), float32] /* ty=Tensor[(256), float32] */, %stage4_unit1_bn1_moving_mean: Tensor[(256), float32] /* ty=Tensor[(256), float32] */, %stage4_unit1_bn1_moving_var: Tensor[(256), float32] /* ty=Tensor[(256), float32] */, %stage4_unit1_conv1_weight: Tensor[(512, 256, 3, 3), float32] /* ty=Tensor[(512, 256, 3, 3), float32] */, %stage4_unit1_bn2_gamma: Tensor[(512), float32] /* ty=Tensor[(512), float32] */, %stage4_unit1_bn2_beta: Tensor[(512), float32] /* ty=Tensor[(512), float32] */, %stage4_unit1_bn2_moving_mean: Tensor[(512), float32] /* ty=Tensor[(512), float32] */, %stage4_unit1_bn2_moving_var: Tensor[(512), float32] /* ty=Tensor[(512), float32] */, %stage4_unit1_conv2_weight: Tensor[(512, 512, 3, 3), float32] /* ty=Tensor[(512, 512, 3, 3), float32] */, %stage4_unit1_sc_weight: Tensor[(512, 256, 1, 1), float32] /* ty=Tensor[(512, 256, 1, 1), float32] */, %stage4_unit2_bn1_gamma: Tensor[(512), float32] /* ty=Tensor[(512), float32] */, %stage4_unit2_bn1_beta: Tensor[(512), float32] /* ty=Tensor[(512), float32] */, %stage4_unit2_bn1_moving_mean: Tensor[(512), float32] /* ty=Tensor[(512), float32] */, %stage4_unit2_bn1_moving_var: Tensor[(512), float32] /* ty=Tensor[(512), float32] */, %stage4_unit2_conv1_weight: Tensor[(512, 512, 3, 3), float32] /* ty=Tensor[(512, 512, 3, 3), float32] */, %stage4_unit2_bn2_gamma: Tensor[(512), float32] /* ty=Tensor[(512), float32] */, %stage4_unit2_bn2_beta: Tensor[(512), float32] /* ty=Tensor[(512), float32] */, %stage4_unit2_bn2_moving_mean: Tensor[(512), float32] /* ty=Tensor[(512), float32] */, %stage4_unit2_bn2_moving_var: Tensor[(512), float32] /* ty=Tensor[(512), float32] */, %stage4_unit2_conv2_weight: Tensor[(512, 512, 3, 3), float32] /* ty=Tensor[(512, 512, 3, 3), float32] */, %bn1_gamma: Tensor[(512), float32] /* ty=Tensor[(512), float32] */, %bn1_beta: Tensor[(512), float32] /* ty=Tensor[(512), float32] */, %bn1_moving_mean: Tensor[(512), float32] /* ty=Tensor[(512), float32] */, %bn1_moving_var: Tensor[(512), float32] /* ty=Tensor[(512), float32] */, %fc1_weight: Tensor[(1000, 512), float32] /* ty=Tensor[(1000, 512), float32] */, %fc1_bias: Tensor[(1000), float32] /* ty=Tensor[(1000), float32] */) -> Tensor[(1, 1000), float32] {
  %0 = nn.batch_norm(%data, %bn_data_gamma, %bn_data_beta, %bn_data_moving_mean, %bn_data_moving_var, epsilon=2e-05f, scale=False) /* ty=(Tensor[(1, 3, 224, 224), float32], Tensor[(3), float32], Tensor[(3), float32]) */;
  %1 = %0.0 /* ty=Tensor[(1, 3, 224, 224), float32] */;
  %2 = nn.conv2d(%1, %conv0_weight, strides=[2, 2], padding=[3, 3, 3, 3], channels=64, kernel_size=[7, 7]) /* ty=Tensor[(1, 64, 112, 112), float32] */;
  %3 = nn.batch_norm(%2, %bn0_gamma, %bn0_beta, %bn0_moving_mean, %bn0_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 64, 112, 112), float32], Tensor[(64), float32], Tensor[(64), float32]) */;
  %4 = %3.0 /* ty=Tensor[(1, 64, 112, 112), float32] */;
  %5 = nn.relu(%4) /* ty=Tensor[(1, 64, 112, 112), float32] */;
  %6 = nn.max_pool2d(%5, pool_size=[3, 3], strides=[2, 2], padding=[1, 1, 1, 1]) /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %7 = nn.batch_norm(%6, %stage1_unit1_bn1_gamma, %stage1_unit1_bn1_beta, %stage1_unit1_bn1_moving_mean, %stage1_unit1_bn1_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 64, 56, 56), float32], Tensor[(64), float32], Tensor[(64), float32]) */;
  %8 = %7.0 /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %9 = nn.relu(%8) /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %10 = nn.conv2d(%9, %stage1_unit1_conv1_weight, padding=[1, 1, 1, 1], channels=64, kernel_size=[3, 3]) /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %11 = nn.batch_norm(%10, %stage1_unit1_bn2_gamma, %stage1_unit1_bn2_beta, %stage1_unit1_bn2_moving_mean, %stage1_unit1_bn2_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 64, 56, 56), float32], Tensor[(64), float32], Tensor[(64), float32]) */;
  %12 = %11.0 /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %13 = nn.relu(%12) /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %14 = nn.conv2d(%13, %stage1_unit1_conv2_weight, padding=[1, 1, 1, 1], channels=64, kernel_size=[3, 3]) /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %15 = nn.conv2d(%9, %stage1_unit1_sc_weight, padding=[0, 0, 0, 0], channels=64, kernel_size=[1, 1]) /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %16 = add(%14, %15) /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %17 = nn.batch_norm(%16, %stage1_unit2_bn1_gamma, %stage1_unit2_bn1_beta, %stage1_unit2_bn1_moving_mean, %stage1_unit2_bn1_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 64, 56, 56), float32], Tensor[(64), float32], Tensor[(64), float32]) */;
  %18 = %17.0 /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %19 = nn.relu(%18) /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %20 = nn.conv2d(%19, %stage1_unit2_conv1_weight, padding=[1, 1, 1, 1], channels=64, kernel_size=[3, 3]) /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %21 = nn.batch_norm(%20, %stage1_unit2_bn2_gamma, %stage1_unit2_bn2_beta, %stage1_unit2_bn2_moving_mean, %stage1_unit2_bn2_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 64, 56, 56), float32], Tensor[(64), float32], Tensor[(64), float32]) */;
  %22 = %21.0 /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %23 = nn.relu(%22) /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %24 = nn.conv2d(%23, %stage1_unit2_conv2_weight, padding=[1, 1, 1, 1], channels=64, kernel_size=[3, 3]) /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %25 = add(%24, %16) /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %26 = nn.batch_norm(%25, %stage2_unit1_bn1_gamma, %stage2_unit1_bn1_beta, %stage2_unit1_bn1_moving_mean, %stage2_unit1_bn1_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 64, 56, 56), float32], Tensor[(64), float32], Tensor[(64), float32]) */;
  %27 = %26.0 /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %28 = nn.relu(%27) /* ty=Tensor[(1, 64, 56, 56), float32] */;
  %29 = nn.conv2d(%28, %stage2_unit1_conv1_weight, strides=[2, 2], padding=[1, 1, 1, 1], channels=128, kernel_size=[3, 3]) /* ty=Tensor[(1, 128, 28, 28), float32] */;
  %30 = nn.batch_norm(%29, %stage2_unit1_bn2_gamma, %stage2_unit1_bn2_beta, %stage2_unit1_bn2_moving_mean, %stage2_unit1_bn2_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 128, 28, 28), float32], Tensor[(128), float32], Tensor[(128), float32]) */;
  %31 = %30.0 /* ty=Tensor[(1, 128, 28, 28), float32] */;
  %32 = nn.relu(%31) /* ty=Tensor[(1, 128, 28, 28), float32] */;
  %33 = nn.conv2d(%32, %stage2_unit1_conv2_weight, padding=[1, 1, 1, 1], channels=128, kernel_size=[3, 3]) /* ty=Tensor[(1, 128, 28, 28), float32] */;
  %34 = nn.conv2d(%28, %stage2_unit1_sc_weight, strides=[2, 2], padding=[0, 0, 0, 0], channels=128, kernel_size=[1, 1]) /* ty=Tensor[(1, 128, 28, 28), float32] */;
  %35 = add(%33, %34) /* ty=Tensor[(1, 128, 28, 28), float32] */;
  %36 = nn.batch_norm(%35, %stage2_unit2_bn1_gamma, %stage2_unit2_bn1_beta, %stage2_unit2_bn1_moving_mean, %stage2_unit2_bn1_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 128, 28, 28), float32], Tensor[(128), float32], Tensor[(128), float32]) */;
  %37 = %36.0 /* ty=Tensor[(1, 128, 28, 28), float32] */;
  %38 = nn.relu(%37) /* ty=Tensor[(1, 128, 28, 28), float32] */;
  %39 = nn.conv2d(%38, %stage2_unit2_conv1_weight, padding=[1, 1, 1, 1], channels=128, kernel_size=[3, 3]) /* ty=Tensor[(1, 128, 28, 28), float32] */;
  %40 = nn.batch_norm(%39, %stage2_unit2_bn2_gamma, %stage2_unit2_bn2_beta, %stage2_unit2_bn2_moving_mean, %stage2_unit2_bn2_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 128, 28, 28), float32], Tensor[(128), float32], Tensor[(128), float32]) */;
  %41 = %40.0 /* ty=Tensor[(1, 128, 28, 28), float32] */;
  %42 = nn.relu(%41) /* ty=Tensor[(1, 128, 28, 28), float32] */;
  %43 = nn.conv2d(%42, %stage2_unit2_conv2_weight, padding=[1, 1, 1, 1], channels=128, kernel_size=[3, 3]) /* ty=Tensor[(1, 128, 28, 28), float32] */;
  %44 = add(%43, %35) /* ty=Tensor[(1, 128, 28, 28), float32] */;
  %45 = nn.batch_norm(%44, %stage3_unit1_bn1_gamma, %stage3_unit1_bn1_beta, %stage3_unit1_bn1_moving_mean, %stage3_unit1_bn1_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 128, 28, 28), float32], Tensor[(128), float32], Tensor[(128), float32]) */;
  %46 = %45.0 /* ty=Tensor[(1, 128, 28, 28), float32] */;
  %47 = nn.relu(%46) /* ty=Tensor[(1, 128, 28, 28), float32] */;
  %48 = nn.conv2d(%47, %stage3_unit1_conv1_weight, strides=[2, 2], padding=[1, 1, 1, 1], channels=256, kernel_size=[3, 3]) /* ty=Tensor[(1, 256, 14, 14), float32] */;
  %49 = nn.batch_norm(%48, %stage3_unit1_bn2_gamma, %stage3_unit1_bn2_beta, %stage3_unit1_bn2_moving_mean, %stage3_unit1_bn2_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 256, 14, 14), float32], Tensor[(256), float32], Tensor[(256), float32]) */;
  %50 = %49.0 /* ty=Tensor[(1, 256, 14, 14), float32] */;
  %51 = nn.relu(%50) /* ty=Tensor[(1, 256, 14, 14), float32] */;
  %52 = nn.conv2d(%51, %stage3_unit1_conv2_weight, padding=[1, 1, 1, 1], channels=256, kernel_size=[3, 3]) /* ty=Tensor[(1, 256, 14, 14), float32] */;
  %53 = nn.conv2d(%47, %stage3_unit1_sc_weight, strides=[2, 2], padding=[0, 0, 0, 0], channels=256, kernel_size=[1, 1]) /* ty=Tensor[(1, 256, 14, 14), float32] */;
  %54 = add(%52, %53) /* ty=Tensor[(1, 256, 14, 14), float32] */;
  %55 = nn.batch_norm(%54, %stage3_unit2_bn1_gamma, %stage3_unit2_bn1_beta, %stage3_unit2_bn1_moving_mean, %stage3_unit2_bn1_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 256, 14, 14), float32], Tensor[(256), float32], Tensor[(256), float32]) */;
  %56 = %55.0 /* ty=Tensor[(1, 256, 14, 14), float32] */;
  %57 = nn.relu(%56) /* ty=Tensor[(1, 256, 14, 14), float32] */;
  %58 = nn.conv2d(%57, %stage3_unit2_conv1_weight, padding=[1, 1, 1, 1], channels=256, kernel_size=[3, 3]) /* ty=Tensor[(1, 256, 14, 14), float32] */;
  %59 = nn.batch_norm(%58, %stage3_unit2_bn2_gamma, %stage3_unit2_bn2_beta, %stage3_unit2_bn2_moving_mean, %stage3_unit2_bn2_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 256, 14, 14), float32], Tensor[(256), float32], Tensor[(256), float32]) */;
  %60 = %59.0 /* ty=Tensor[(1, 256, 14, 14), float32] */;
  %61 = nn.relu(%60) /* ty=Tensor[(1, 256, 14, 14), float32] */;
  %62 = nn.conv2d(%61, %stage3_unit2_conv2_weight, padding=[1, 1, 1, 1], channels=256, kernel_size=[3, 3]) /* ty=Tensor[(1, 256, 14, 14), float32] */;
  %63 = add(%62, %54) /* ty=Tensor[(1, 256, 14, 14), float32] */;
  %64 = nn.batch_norm(%63, %stage4_unit1_bn1_gamma, %stage4_unit1_bn1_beta, %stage4_unit1_bn1_moving_mean, %stage4_unit1_bn1_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 256, 14, 14), float32], Tensor[(256), float32], Tensor[(256), float32]) */;
  %65 = %64.0 /* ty=Tensor[(1, 256, 14, 14), float32] */;
  %66 = nn.relu(%65) /* ty=Tensor[(1, 256, 14, 14), float32] */;
  %67 = nn.conv2d(%66, %stage4_unit1_conv1_weight, strides=[2, 2], padding=[1, 1, 1, 1], channels=512, kernel_size=[3, 3]) /* ty=Tensor[(1, 512, 7, 7), float32] */;
  %68 = nn.batch_norm(%67, %stage4_unit1_bn2_gamma, %stage4_unit1_bn2_beta, %stage4_unit1_bn2_moving_mean, %stage4_unit1_bn2_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 512, 7, 7), float32], Tensor[(512), float32], Tensor[(512), float32]) */;
  %69 = %68.0 /* ty=Tensor[(1, 512, 7, 7), float32] */;
  %70 = nn.relu(%69) /* ty=Tensor[(1, 512, 7, 7), float32] */;
  %71 = nn.conv2d(%70, %stage4_unit1_conv2_weight, padding=[1, 1, 1, 1], channels=512, kernel_size=[3, 3]) /* ty=Tensor[(1, 512, 7, 7), float32] */;
  %72 = nn.conv2d(%66, %stage4_unit1_sc_weight, strides=[2, 2], padding=[0, 0, 0, 0], channels=512, kernel_size=[1, 1]) /* ty=Tensor[(1, 512, 7, 7), float32] */;
  %73 = add(%71, %72) /* ty=Tensor[(1, 512, 7, 7), float32] */;
  %74 = nn.batch_norm(%73, %stage4_unit2_bn1_gamma, %stage4_unit2_bn1_beta, %stage4_unit2_bn1_moving_mean, %stage4_unit2_bn1_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 512, 7, 7), float32], Tensor[(512), float32], Tensor[(512), float32]) */;
  %75 = %74.0 /* ty=Tensor[(1, 512, 7, 7), float32] */;
  %76 = nn.relu(%75) /* ty=Tensor[(1, 512, 7, 7), float32] */;
  %77 = nn.conv2d(%76, %stage4_unit2_conv1_weight, padding=[1, 1, 1, 1], channels=512, kernel_size=[3, 3]) /* ty=Tensor[(1, 512, 7, 7), float32] */;
  %78 = nn.batch_norm(%77, %stage4_unit2_bn2_gamma, %stage4_unit2_bn2_beta, %stage4_unit2_bn2_moving_mean, %stage4_unit2_bn2_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 512, 7, 7), float32], Tensor[(512), float32], Tensor[(512), float32]) */;
  %79 = %78.0 /* ty=Tensor[(1, 512, 7, 7), float32] */;
  %80 = nn.relu(%79) /* ty=Tensor[(1, 512, 7, 7), float32] */;
  %81 = nn.conv2d(%80, %stage4_unit2_conv2_weight, padding=[1, 1, 1, 1], channels=512, kernel_size=[3, 3]) /* ty=Tensor[(1, 512, 7, 7), float32] */;
  %82 = add(%81, %73) /* ty=Tensor[(1, 512, 7, 7), float32] */;
  %83 = nn.batch_norm(%82, %bn1_gamma, %bn1_beta, %bn1_moving_mean, %bn1_moving_var, epsilon=2e-05f) /* ty=(Tensor[(1, 512, 7, 7), float32], Tensor[(512), float32], Tensor[(512), float32]) */;
  %84 = %83.0 /* ty=Tensor[(1, 512, 7, 7), float32] */;
  %85 = nn.relu(%84) /* ty=Tensor[(1, 512, 7, 7), float32] */;
  %86 = nn.global_avg_pool2d(%85) /* ty=Tensor[(1, 512, 1, 1), float32] */;
  %87 = nn.batch_flatten(%86) /* ty=Tensor[(1, 512), float32] */;
  %88 = nn.dense(%87, %fc1_weight, units=1000) /* ty=Tensor[(1, 1000), float32] */;
  %89 = nn.bias_add(%88, %fc1_bias, axis=-1) /* ty=Tensor[(1, 1000), float32] */;
  nn.softmax(%89) /* ty=Tensor[(1, 1000), float32] */
}

Create PassContext With Instruments

To run all passes with an instrument, pass it via the instruments argument to the PassContext constructor. A built-in PassTimingInstrument is used to profile the execution time of each passes.

timing_inst = PassTimingInstrument()
with tvm.transform.PassContext(instruments=[timing_inst]):
    relay_mod = relay.transform.InferType()(relay_mod)
    relay_mod = relay.transform.FoldScaleAxis()(relay_mod)
    # before exiting the context, get profile results.
    profiles = timing_inst.render()
print("Printing results of timing profile...")
print(profiles)

Printing results of timing profile...
InferType: 24404us [24404us] (48.58%; 48.58%)
FoldScaleAxis: 25831us [9us] (51.42%; 51.42%)
        FoldConstant: 25822us [1783us] (51.40%; 99.96%)
                InferType: 24039us [24039us] (47.85%; 93.09%)

Use Current PassContext With Instruments

One can also use the current PassContext and register PassInstrument instances by override_instruments method. Note that override_instruments executes exit_pass_ctx method if any instrument already exists. Then it switches to new instruments and calls enter_pass_ctx method of new instruments. Refer to following sections and tvm.instrument.pass_instrument() for these methods.

cur_pass_ctx = tvm.transform.PassContext.current()
cur_pass_ctx.override_instruments([timing_inst])
relay_mod = relay.transform.InferType()(relay_mod)
relay_mod = relay.transform.FoldScaleAxis()(relay_mod)
profiles = timing_inst.render()
print("Printing results of timing profile...")
print(profiles)

Printing results of timing profile...
InferType: 24192us [24192us] (48.35%; 48.35%)
FoldScaleAxis: 25845us [9us] (51.65%; 51.65%)
        FoldConstant: 25836us [1849us] (51.63%; 99.96%)
                InferType: 23987us [23987us] (47.94%; 92.84%)

Register empty list to clear existing instruments.

Note that exit_pass_ctx of PassTimingInstrument is called. Profiles are cleared so nothing is printed.

cur_pass_ctx.override_instruments([])
# Uncomment the call to .render() to see a warning like:
# Warning: no passes have been profiled, did you enable pass profiling?
# profiles = timing_inst.render()

Create Customized Instrument Class

A customized instrument class can be created using the tvm.instrument.pass_instrument() decorator.

Let’s create an instrument class which calculates the change in number of occurrences of each operator caused by each pass. We can look at op.name to find the name of each operator. And we do this before and after passes to calculate the difference.

@pass_instrument
class RelayCallNodeDiffer:
    def __init__(self):
        self._op_diff = []
        # Passes can be nested.
        # Use stack to make sure we get correct before/after pairs.
        self._op_cnt_before_stack = []

    def enter_pass_ctx(self):
        self._op_diff = []
        self._op_cnt_before_stack = []

    def exit_pass_ctx(self):
        assert len(self._op_cnt_before_stack) == 0, "The stack is not empty. Something wrong."

    def run_before_pass(self, mod, info):
        self._op_cnt_before_stack.append((info.name, self._count_nodes(mod)))

    def run_after_pass(self, mod, info):
        # Pop out the latest recorded pass.
        name_before, op_to_cnt_before = self._op_cnt_before_stack.pop()
        assert name_before == info.name, "name_before: {}, info.name: {} doesn't match".format(
            name_before, info.name
        )
        cur_depth = len(self._op_cnt_before_stack)
        op_to_cnt_after = self._count_nodes(mod)
        op_diff = self._diff(op_to_cnt_after, op_to_cnt_before)
        # only record passes causing differences.
        if op_diff:
            self._op_diff.append((cur_depth, info.name, op_diff))

    def get_pass_to_op_diff(self):
        """
        return [
          (depth, pass_name, {op_name: diff_num, ...}), ...
        ]
        """
        return self._op_diff

    @staticmethod
    def _count_nodes(mod):
        """Count the number of occurrences of each operator in the module"""
        ret = {}

        def visit(node):
            if isinstance(node, relay.expr.Call):
                if hasattr(node.op, "name"):
                    op_name = node.op.name
                else:
                    # Some CallNode may not have 'name' such as relay.Function
                    return
                ret[op_name] = ret.get(op_name, 0) + 1

        relay.analysis.post_order_visit(mod["main"], visit)
        return ret

    @staticmethod
    def _diff(d_after, d_before):
        """Calculate the difference of two dictionary along their keys.
        The result is values in d_after minus values in d_before.
        """
        ret = {}
        key_after, key_before = set(d_after), set(d_before)
        for k in key_before & key_after:
            tmp = d_after[k] - d_before[k]
            if tmp:
                ret[k] = d_after[k] - d_before[k]
        for k in key_after - key_before:
            ret[k] = d_after[k]
        for k in key_before - key_after:
            ret[k] = -d_before[k]
        return ret

Apply Passes and Multiple Instrument Classes

We can use multiple instrument classes in a PassContext. However, it should be noted that instrument methods are executed sequentially, obeying the order of instruments argument. So for instrument classes like PassTimingInstrument, it is inevitable to count-up the execution time of other instrument classes to the final profile result.

call_node_inst = RelayCallNodeDiffer()
desired_layouts = {
    "nn.conv2d": ["NHWC", "HWIO"],
}
pass_seq = tvm.transform.Sequential(
    [
        relay.transform.FoldConstant(),
        relay.transform.ConvertLayout(desired_layouts),
        relay.transform.FoldConstant(),
    ]
)
relay_mod["main"] = bind_params_by_name(relay_mod["main"], relay_params)
# timing_inst is put after call_node_inst.
# So the execution time of ``call_node.inst.run_after_pass()`` is also counted.
with tvm.transform.PassContext(opt_level=3, instruments=[call_node_inst, timing_inst]):
    relay_mod = pass_seq(relay_mod)
    profiles = timing_inst.render()
# Uncomment the next line to see timing-profile results.
# print(profiles)

We can see how many CallNode increase/decrease per op type.

from pprint import pprint

print("Printing the change in number of occurrences of each operator caused by each pass...")
pprint(call_node_inst.get_pass_to_op_diff())

Printing the change in number of occurrences of each operator caused by each pass...
[(1, 'CanonicalizeOps', {'add': 1, 'nn.bias_add': -1}),
 (1, 'ConvertLayout', {'expand_dims': 1, 'layout_transform': 23}),
 (1, 'FoldConstant', {'expand_dims': -1, 'layout_transform': -21}),
 (0, 'sequential', {'add': 1, 'layout_transform': 2, 'nn.bias_add': -1})]

Exception Handling

Let’s see what happens if an exception occurs in a method of a PassInstrument.

Define PassInstrument classes which raise exceptions in enter/exit PassContext:

class PassExampleBase:
    def __init__(self, name):
        self._name = name

    def enter_pass_ctx(self):
        print(self._name, "enter_pass_ctx")

    def exit_pass_ctx(self):
        print(self._name, "exit_pass_ctx")

    def should_run(self, mod, info):
        print(self._name, "should_run")
        return True

    def run_before_pass(self, mod, pass_info):
        print(self._name, "run_before_pass")

    def run_after_pass(self, mod, pass_info):
        print(self._name, "run_after_pass")


@pass_instrument
class PassFine(PassExampleBase):
    pass


@pass_instrument
class PassBadEnterCtx(PassExampleBase):
    def enter_pass_ctx(self):
        print(self._name, "bad enter_pass_ctx!!!")
        raise ValueError("{} bad enter_pass_ctx".format(self._name))


@pass_instrument
class PassBadExitCtx(PassExampleBase):
    def exit_pass_ctx(self):
        print(self._name, "bad exit_pass_ctx!!!")
        raise ValueError("{} bad exit_pass_ctx".format(self._name))

If an exception occurs in enter_pass_ctx, PassContext will disable the pass instrumentation. And it will run the exit_pass_ctx of each PassInstrument which successfully finished enter_pass_ctx.

In following example, we can see exit_pass_ctx of PassFine_0 is executed after exception.

demo_ctx = tvm.transform.PassContext(
    instruments=[
        PassFine("PassFine_0"),
        PassBadEnterCtx("PassBadEnterCtx"),
        PassFine("PassFine_1"),
    ]
)
try:
    with demo_ctx:
        relay_mod = relay.transform.InferType()(relay_mod)
except ValueError as ex:
    print("Catching", str(ex).split("\n")[-1])

PassFine_0 enter_pass_ctx
PassBadEnterCtx bad enter_pass_ctx!!!
PassFine_0 exit_pass_ctx
Catching PassBadEnterCtx bad enter_pass_ctx

Exceptions in PassInstrument instances cause all instruments of the current PassContext to be cleared, so nothing is printed when override_instruments is called.

demo_ctx.override_instruments([])  # no PassFine_0 exit_pass_ctx printed....etc

If an exception occurs in exit_pass_ctx, then the pass instrument is disabled. Then exception is propagated. That means PassInstrument instances registered after the one throwing the exception do not execute exit_pass_ctx.

demo_ctx = tvm.transform.PassContext(
    instruments=[
        PassFine("PassFine_0"),
        PassBadExitCtx("PassBadExitCtx"),
        PassFine("PassFine_1"),
    ]
)
try:
    # PassFine_1 execute enter_pass_ctx, but not exit_pass_ctx.
    with demo_ctx:
        relay_mod = relay.transform.InferType()(relay_mod)
except ValueError as ex:
    print("Catching", str(ex).split("\n")[-1])

PassFine_0 enter_pass_ctx
PassBadExitCtx enter_pass_ctx
PassFine_1 enter_pass_ctx
PassFine_0 should_run
PassBadExitCtx should_run
PassFine_1 should_run
PassFine_0 run_before_pass
PassBadExitCtx run_before_pass
PassFine_1 run_before_pass
PassFine_0 run_after_pass
PassBadExitCtx run_after_pass
PassFine_1 run_after_pass
PassFine_0 exit_pass_ctx
PassBadExitCtx bad exit_pass_ctx!!!
Catching PassBadExitCtx bad exit_pass_ctx

Exceptions occurred in should_run, run_before_pass, run_after_pass are not handled explicitly – we rely on the context manager (the with syntax) to exit PassContext safely.

We use run_before_pass as an example:

@pass_instrument
class PassBadRunBefore(PassExampleBase):
    def run_before_pass(self, mod, pass_info):
        print(self._name, "bad run_before_pass!!!")
        raise ValueError("{} bad run_before_pass".format(self._name))


demo_ctx = tvm.transform.PassContext(
    instruments=[
        PassFine("PassFine_0"),
        PassBadRunBefore("PassBadRunBefore"),
        PassFine("PassFine_1"),
    ]
)
try:
    # All exit_pass_ctx are called.
    with demo_ctx:
        relay_mod = relay.transform.InferType()(relay_mod)
except ValueError as ex:
    print("Catching", str(ex).split("\n")[-1])

PassFine_0 enter_pass_ctx
PassBadRunBefore enter_pass_ctx
PassFine_1 enter_pass_ctx
PassFine_0 should_run
PassBadRunBefore should_run
PassFine_1 should_run
PassFine_0 run_before_pass
PassBadRunBefore bad run_before_pass!!!
PassFine_0 exit_pass_ctx
PassBadRunBefore exit_pass_ctx
PassFine_1 exit_pass_ctx
Catching PassBadRunBefore bad run_before_pass

Also note that pass instrumentation is not disable. So if we call override_instruments, the exit_pass_ctx of old registered PassInstrument is called.

demo_ctx.override_instruments([])

PassFine_0 exit_pass_ctx
PassBadRunBefore exit_pass_ctx
PassFine_1 exit_pass_ctx

If we don’t wrap pass execution with with syntax, exit_pass_ctx is not called. Let try this with current PassContext:

cur_pass_ctx = tvm.transform.PassContext.current()
cur_pass_ctx.override_instruments(
    [
        PassFine("PassFine_0"),
        PassBadRunBefore("PassBadRunBefore"),
        PassFine("PassFine_1"),
    ]
)

PassFine_0 enter_pass_ctx
PassBadRunBefore enter_pass_ctx
PassFine_1 enter_pass_ctx

Then call passes. exit_pass_ctx is not executed after the exception, as expectation.

try:
    # No ``exit_pass_ctx`` got executed.
    relay_mod = relay.transform.InferType()(relay_mod)
except ValueError as ex:
    print("Catching", str(ex).split("\n")[-1])

PassFine_0 should_run
PassBadRunBefore should_run
PassFine_1 should_run
PassFine_0 run_before_pass
PassBadRunBefore bad run_before_pass!!!
Catching PassBadRunBefore bad run_before_pass

Clear instruments.

cur_pass_ctx.override_instruments([])

PassFine_0 exit_pass_ctx
PassBadRunBefore exit_pass_ctx
PassFine_1 exit_pass_ctx