【他山之石】在C++平台上部署PyTorch模型流程+踩坑实录
最近因为工作需要,要把pytorch的模型部署到c++平台上,基本过程主要参照官网的教学示例,期间发现了不少坑,特此记录。
01
模型转换
import torchimport torchvision# An instance of your model.model = torchvision.models.resnet18()# An example input you would normally provide to your model's forward() method.example = torch.rand(1, 3, 224, 224)# Use torch.jit.trace to generate a torch.jit.ScriptModule via tracing.traced_script_module = torch.jit.trace(model, example)
class MyModule(torch.nn.Module):def __init__(self, N, M):super(MyModule, self).__init__()self.weight = torch.nn.Parameter(torch.rand(N, M))def forward(self, input):if input.sum() > 0:output = self.weight.mv(input)else:output = self.weight + inputreturn outputmy_module = MyModule(10,20)sm = torch.jit.script(my_module)
-
forward方法会被默认编译,forward中被调用的方法也会按照被调用的顺序被编译 -
如果想要编译一个forward以外且未被forward调用的方法,可以添加 @torch.jit.export. -
如果想要方法不被编译,可使用@torch.jit.ignore[1] 或者 @torch.jit.unused[2]
# Same behavior as pre-PyTorch 1.2def some_fn():return 2# Marks a function as ignored, if nothing# ever calls it then this has no effectdef some_fn2():return 2# As with ignore, if nothing calls it then it has no effect.# If it is called in script it is replaced with an exception.def some_fn3():import pdb; pdb.set_trace()return 4# Doesn't do anything, this function is already# the main entry pointdef some_fn4():return 2
def __init__(self, **kwargs):
或者
if output_flag == 0:return reshape_logitselse:loss = self.loss(reshape_logits, term_mask, labels_id)return reshape_logits, loss
layers = [int(a) for a in layers]
报错torch.jit.frontend.UnsupportedNodeError: ListComp aren’t supported
for k in range(len(layers)):layers[k] = int(layers[k])
seq_iter = enumerate(scores)try:_, inivalues = seq_iter.__next__()except:_, inivalues = seq_iter.next()
line = next(infile)
3)不支持的语句
from typing import Dictclass MyModule(torch.nn.Module):my_dict: Dict[str, int]def __init__(self):super(MyModule, self).__init__()# This type cannot be inferred and must be specifiedself.my_dict = {}# The attribute type here is inferred to be `int`self.my_int = 20def forward(self):passm = torch.jit.script(MyModule())
try:from typing_extensions import Finalexcept:# If you don't have `typing_extensions` installed, you can use a# polyfill from `torch.jit`.from torch.jit import Finalclass MyModule(torch.nn.Module):my_constant: Final[int]def __init__(self):super(MyModule, self).__init__()self.my_constant = 2def forward(self):passm = torch.jit.script(MyModule())
def forward(self, batch_size:int, seq_len:int, use_cuda:bool):
方法三:Tracing and Scriptin混合
import torchdef foo(x, y):if x.max() > y.max():r = xelse:r = yreturn rdef bar(x, y, z):return foo(x, y) + ztraced_bar = torch.jit.trace(bar, (torch.rand(3), torch.rand(3), torch.rand(3)))
import torchimport torchvisionclass MyScriptModule(torch.nn.Module):def __init__(self):super(MyScriptModule, self).__init__()self.means = torch.nn.Parameter(torch.tensor([103.939, 116.779, 123.68]).resize_(1, 3, 1, 1))self.resnet = torch.jit.trace(torchvision.models.resnet18(),torch.rand(1, 3, 224, 224))def forward(self, input):return self.resnet(input - self.means)my_script_module = torch.jit.script(MyScriptModule())
02
保存序列化模型
gpu_model.eval()cpu_model = gpu_model.cpu()sample_input_cpu = sample_input_gpu.cpu()traced_cpu = torch.jit.trace(traced_cpu, sample_input_cpu)torch.jit.save(traced_cpu, "cpu.pth")traced_gpu = torch.jit.trace(traced_gpu, sample_input_gpu)torch.jit.save(traced_gpu, "gpu.pth")
03
C++ load训练好的模型
libtorch/bin/include/lib/share/
example-app/CMakeLists.txtexample-app.cpp
int main(int argc, const char* argv[]) {if (argc != 2) {std::cerr << "usage: example-app <path-to-exported-script-module>\n";return -1;}torch::jit::script::Module module;try {// Deserialize the ScriptModule from a file using torch::jit::load().module = torch::jit::load(argv[1]);}catch (const c10::Error& e) {std::cerr << "error loading the model\n";return -1;}std::cout << "ok\n";}
cmake_minimum_required(VERSION 3.0 FATAL_ERROR)project(custom_ops)find_package(Torch REQUIRED)add_executable(example-app example-app.cpp)target_link_libraries(example-app "${TORCH_LIBRARIES}")set_property(TARGET example-app PROPERTY CXX_STANDARD 14)
mkdir buildcd buildcmake -DCMAKE_PREFIX_PATH=/path/to/libtorch ..cmake --build . --config Release
04
执行Script Module
// Create a vector of inputs.std::vector<torch::jit::IValue> inputs;inputs.push_back(torch::ones({1, 3, 224, 224}));// Execute the model and turn its output into a tensor.at::Tensor output = module.forward(inputs).toTensor();std::cout << output.slice(/*dim=*/1, /*start=*/0, /*end=*/5) << '\n';
torch::tensor(input_list[j]).to(at::kLong).resize_({batch, 128}).clone()//torch::tensor对应pytorch的torch.tensor; at::kLong对应torch.int64;resize_对应resize
[1] https://pytorch.org/docs/master/generated/torch.jit.ignore.html#torch.jit.ignore
[2] https://pytorch.org/docs/master/generated/torch.jit.unused.html#torch.jit.unused
[3] https://pytorch.org/docs/master/jit_unsupported.html#jit-unsupported
https://pytorch.org/cppdocs/
https://pytorch.org/tutorials/advanced/cpp_export.html
直播预告
历史文章推荐
分享、点赞、在看,给个三连击呗!
