[WIP]

Author: Will Feng

cpp/autograd/CMakeLists.txt

@@ -0,0 +1,21 @@
+cmake_minimum_required(VERSION 2.8)
+
+project(autograd)
+set(CMAKE_CXX_STANDARD 14)
+
+find_package(Torch REQUIRED)
+
+add_executable(${PROJECT_NAME} "autograd.cpp")
+target_link_libraries(${PROJECT_NAME} "${TORCH_LIBRARIES}")
+
+# The following code block is suggested to be used on Windows.
+# According to https://github.com/pytorch/pytorch/issues/25457,
+# the DLLs need to be copied to avoid memory errors.
+if (MSVC)
+  file(GLOB TORCH_DLLS "${TORCH_INSTALL_PREFIX}/lib/*.dll")
+  add_custom_command(TARGET ${PROJECT_NAME}
+                     POST_BUILD
+                     COMMAND ${CMAKE_COMMAND} -E copy_if_different
+                     ${TORCH_DLLS}
+                     $<TARGET_FILE_DIR:${PROJECT_NAME}>)
+endif (MSVC)

cpp/autograd/README.md

@@ -0,0 +1,78 @@
+# C++ autograd example
+
+`autograd.cpp` contains several examples of doing autograd in PyTorch C++ frontend.
+
+To build the code, run the following commands from your terminal:
+
+```shell
+$ cd autograd
+$ mkdir build
+$ cd build
+$ cmake -DCMAKE_PREFIX_PATH=/path/to/libtorch ..
+$ make
+```
+
+where `/path/to/libtorch` should be the path to the unzipped *LibTorch*
+distribution, which you can get from the [PyTorch
+homepage](https://pytorch.org/get-started/locally/).
+
+Execute the compiled binary to run:
+
+```shell
+$ ./autograd
+====== Running: "Basic autograd operations" ======
+ 1  1
+ 1  1
+[ CPUFloatType{2,2} ]
+ 3  3
+ 3  3
+[ CPUFloatType{2,2} ]
+AddBackward1
+ 27  27
+ 27  27
+[ CPUFloatType{2,2} ]
+MulBackward1
+27
+[ CPUFloatType{} ]
+MeanBackward0
+false
+true
+SumBackward0
+ 4.5000  4.5000
+ 4.5000  4.5000
+[ CPUFloatType{2,2} ]
+ -731.0470
+  963.0721
+ 1236.4192
+[ CPUFloatType{3} ]
+MulBackward1
+  102.4000
+ 1024.0000
+    0.1024
+[ CPUFloatType{3} ]
+true
+true
+false
+true
+false
+true
+
+====== Running "Computing higher-order gradients in C++" ======
+-0.0384  0.1510 -0.0288  0.0872
+-0.0105 -0.0936 -0.0553 -0.0222
+ 0.0589 -0.0848 -0.0730  0.0070
+[ CPUFloatType{3,4} ]
+
+====== Running "Using custom autograd function in C++" ======
+-0.6962 -1.7728  1.4167
+-0.6962 -1.7728  1.4167
+[ CPUFloatType{2,3} ]
+ 1.5162  1.6421  1.2691
+ 1.5162  1.6421  1.2691
+ 1.5162  1.6421  1.2691
+ 1.5162  1.6421  1.2691
+[ CPUFloatType{4,3} ]
+ 5.5000
+ 5.5000
+[ CPUFloatType{2} ]
+```

cpp/autograd/autograd.cpp

@@ -0,0 +1,175 @@
+#include <torch/torch.h>
+#include <iostream>
+
+using namespace torch::autograd;
+
+void basic_autograd_operations_example() {
+  std::cout << "====== Running: \"Basic autograd operations\" ======" << std::endl;
+
+  auto x = torch::ones({2, 2}, torch::requires_grad());
+  std::cout << x << std::endl;
+
+  auto y = x + 2;
+  std::cout << y << std::endl;
+
+  std::cout << y.grad_fn()->name() << std::endl;
+
+  auto z = y * y * 3;
+  auto out = z.mean();
+
+  std::cout << z << std::endl;
+  std::cout << z.grad_fn()->name() << std::endl;
+  std::cout << out << std::endl;
+  std::cout << out.grad_fn()->name() << std::endl;
+
+  auto a = torch::randn({2, 2});
+  a = ((a * 3) / (a - 1));
+  std::cout << a.requires_grad() << std::endl;
+
+  a.requires_grad_(true);
+  std::cout << a.requires_grad() << std::endl;
+
+  auto b = (a * a).sum();
+  std::cout << b.grad_fn()->name() << std::endl;
+
+  out.backward();
+  std::cout << x.grad() << std::endl;
+
+  x = torch::randn(3, torch::requires_grad());
+
+  y = x * 2;
+  while (y.norm().item<double>() < 1000) {
+    y = y * 2;
+  }
+
+  std::cout << y << std::endl;
+  std::cout << y.grad_fn()->name() << std::endl;
+
+  auto v = torch::tensor({0.1, 1.0, 0.0001}, torch::kFloat);
+  y.backward(v);
+
+  std::cout << x.grad() << std::endl;
+
+  std::cout << x.requires_grad() << std::endl;
+  std::cout << x.pow(2).requires_grad() << std::endl;
+
+  {
+    torch::NoGradGuard no_grad;
+    std::cout << x.pow(2).requires_grad() << std::endl;
+  }
+
+  std::cout << x.requires_grad() << std::endl;
+  y = x.detach();
+  std::cout << y.requires_grad() << std::endl;
+  std::cout << x.eq(y).all().item<bool>() << std::endl;
+}
+
+void compute_higher_order_gradients_example() {
+  std::cout << "====== Running \"Computing higher-order gradients in C++\" ======" << std::endl;
+
+  auto model = torch::nn::Linear(4, 3);
+
+  auto input = torch::randn({3, 4}).requires_grad_(true);
+  auto output = model(input);
+
+  // Calculate loss
+  auto target = torch::randn({3, 3});
+  auto loss = torch::nn::MSELoss()(output, target);
+
+  // Use norm of gradients as penalty
+  auto grad_output = torch::ones_like(output);
+  auto gradient = torch::autograd::grad({output}, {input}, /*grad_outputs=*/{grad_output},
+                  /*retain_graph=*/true, /*create_graph=*/true,
+                  /*allow_unused=*/true)[0];
+  gradient = gradient.view({-1, 1});
+  auto gradient_penalty = torch::pow((gradient.norm(2, /*dim=*/1) - 1), 2).mean();
+
+  // Add gradient penalty to loss
+  auto combined_loss = loss + gradient_penalty;
+  combined_loss.backward();
+
+  std::cout << input.grad() << std::endl;
+}
+
+// Inherit from Function
+class LinearFunction : public Function<LinearFunction> {
+ public:
+  // Note that both forward and backward are static functions
+
+  // bias is an optional argument
+  static Variable forward(AutogradContext *ctx, Variable input, Variable weight, Variable bias = Variable()) {
+    ctx->save_for_backward({input, weight, bias});
+    auto output = input.mm(weight.t());
+    if (bias.defined()) {
+      output += bias.unsqueeze(0).expand_as(output);
+    }
+    return output;
+  }
+
+  static variable_list backward(AutogradContext *ctx, variable_list grad_outputs) {
+    auto saved = ctx->get_saved_variables();
+    auto input = saved[0];
+    auto weight = saved[1];
+    auto bias = saved[2];
+
+    auto grad_output = grad_outputs[0];
+    auto grad_input = grad_output.mm(weight);
+    auto grad_weight = grad_output.t().mm(input);
+    auto grad_bias = Variable();
+    if (bias.defined()) {
+      grad_bias = grad_output.sum(0);
+    }
+
+    return {grad_input, grad_weight, grad_bias};
+  }
+};
+
+class MulConstant : public Function<MulConstant> {
+ public:
+  static Variable forward(AutogradContext *ctx, Variable variable, double constant) {
+    // ctx is a context object that can be used to stash information
+    // for backward computation
+    ctx->saved_data["constant"] = constant;
+    return variable * constant;
+  }
+
+  static variable_list backward(AutogradContext *ctx, variable_list grad_outputs) {
+    // We return as many input gradients as there were arguments.
+    // Gradients of non-Tensor arguments to forward must be `Variable()`.
+    return {grad_outputs[0] * ctx->saved_data["constant"].toDouble(), Variable()};
+  }
+};
+
+void custom_autograd_function_example() {
+  std::cout << "====== Running \"Using custom autograd function in C++\" ======" << std::endl;
+  {
+    auto x = torch::randn({2, 3}).requires_grad_();
+    auto weight = torch::randn({4, 3}).requires_grad_();
+    auto y = LinearFunction::apply(x, weight);
+    y.sum().backward();
+
+    std::cout << x.grad() << std::endl;
+    std::cout << weight.grad() << std::endl;
+  }
+  {
+    auto x = torch::randn({2}).requires_grad_();
+    auto y = MulConstant::apply(x, 5.5);
+    y.sum().backward();
+
+    std::cout << x.grad() << std::endl;
+  }
+}
+
+int main() {
+  std::cout << std::boolalpha;
+
+  basic_autograd_operations_example();
+
+  std::cout << "\n";
+
+  compute_higher_order_gradients_example();
+
+  std::cout << "\n";
+
+  custom_autograd_function_example();
+}