feat(//tests): New optional accuracy tests to check INT8 and FP16

Signed-off-by: Naren Dasan <[email protected]>
Signed-off-by: Naren Dasan <[email protected]>

Author: narendasan

.gitignore

@@ -21,5 +21,6 @@ py/.eggs
 cpp/ptq/training/vgg16/data/*
 *.bin
 cpp/ptq/datasets/data/
+tests/accuracy/datasets/data/*
 ._.DS_Store
 *.tar.gz

tests/BUILD

@@ -5,4 +5,11 @@ test_suite(
         "//tests/modules:test_modules"
     ],
 )
-    
+
+test_suite(
+    name = "required_and_optional_tests",
+    tests = [
+        ":tests",
+        "//tests/accuracy:test_accuracy"
+    ]
+)

tests/accuracy/BUILD

@@ -0,0 +1,73 @@
+filegroup(
+    name = "jit_models",
+    srcs = glob(["**/*.jit.pt"])
+)
+
+test_suite(
+    name = "test_accuracy",
+    tests = [
+        ":test_int8_accuracy",
+        ":test_fp16_accuracy",
+        ":test_fp32_accuracy",
+    ]
+)
+
+cc_test(
+    name = "test_int8_accuracy",
+    srcs = ["test_int8_accuracy.cpp"],
+    deps = [
+        ":accuracy_test",
+        "//tests/accuracy/datasets:cifar10"
+    ],
+    data = [
+        ":jit_models",
+    ]
+)
+
+cc_test(
+    name = "test_fp16_accuracy",
+    srcs = ["test_fp16_accuracy.cpp"],
+    deps = [
+        ":accuracy_test",
+        "//tests/accuracy/datasets:cifar10"
+    ],
+    data = [
+        ":jit_models",
+    ]
+)
+
+cc_test(
+    name = "test_fp32_accuracy",
+    srcs = ["test_fp32_accuracy.cpp"],
+    deps = [
+        ":accuracy_test",
+        "//tests/accuracy/datasets:cifar10"
+    ],
+    data = [
+        ":jit_models",
+    ]
+)
+
+cc_binary(
+    name = "test",
+    srcs = ["test.cpp"],
+    deps = [
+        ":accuracy_test",
+        "//tests/accuracy/datasets:cifar10"
+    ],
+    data = [
+        ":jit_models",
+    ]
+)
+
+
+cc_library(
+    name = "accuracy_test",
+    hdrs = ["accuracy_test.h"],
+    deps = [
+        "//cpp/api:trtorch",
+        "//tests/util",
+        "@libtorch//:libtorch",
+        "@googletest//:gtest_main",
+    ],
+)

tests/accuracy/accuracy_test.h

@@ -0,0 +1,30 @@
+#include <utility>
+#include "torch/script.h"
+#include "gtest/gtest.h"
+#include "tests/util/util.h"
+#include "trtorch/trtorch.h"
+#include "c10/cuda/CUDACachingAllocator.h"
+
+// TODO: Extend this to support other datasets
+class AccuracyTests
+    : public testing::TestWithParam<std::string> {
+public:
+    void SetUp() override {
+        auto params = GetParam();
+        auto module_path = params;
+        try {
+            // Deserialize the ScriptModule from a file using torch::jit::load().
+            mod = torch::jit::load(module_path);
+        }
+        catch (const c10::Error& e) {
+            std::cerr << "error loading the model\n";
+            return;
+        }
+    }
+
+    void TearDown() {
+        c10::cuda::CUDACachingAllocator::emptyCache();
+    }
+protected:
+    torch::jit::script::Module mod;
+};

tests/accuracy/datasets/BUILD

@@ -0,0 +1,23 @@
+package(default_visibility = ["//visibility:public"])
+
+cc_library(
+    name = "cifar10",
+    hdrs = [
+        "cifar10.h"
+    ],
+    srcs = [
+        "cifar10.cpp"
+    ],
+    deps = [
+        "@libtorch//:libtorch"
+    ],
+    data = [
+        ":cifar10_data"
+    ]
+
+)
+
+filegroup(
+    name = "cifar10_data",
+    srcs = glob(["data/cifar-10-batches-bin/**/*.bin"])
+)

tests/accuracy/datasets/cifar10.cpp

@@ -0,0 +1,132 @@
+#include "tests/accuracy/datasets/cifar10.h"
+
+#include "torch/torch.h"
+#include "torch/data/example.h"
+#include "torch/types.h"
+
+#include <iostream>
+#include <cstddef>
+#include <fstream>
+#include <string>
+#include <vector>
+#include <utility>
+#include <sstream>
+#include <memory>
+
+namespace datasets {
+namespace {
+constexpr const char* kTrainFilenamePrefix = "data_batch_";
+constexpr const uint32_t kNumTrainFiles = 5;
+constexpr const char* kTestFilename = "test_batch.bin";
+constexpr const size_t kLabelSize = 1; // B
+constexpr const size_t kImageSize = 3072; // B
+constexpr const size_t kImageDim = 32;
+constexpr const size_t kImageChannels = 3;
+constexpr const size_t kBatchSize = 10000;
+
+std::pair<torch::Tensor, torch::Tensor> read_batch(const std::string& path) {
+    std::ifstream batch;
+    batch.open(path, std::ios::in|std::ios::binary|std::ios::ate);
+
+    auto file_size = batch.tellg();
+    std::unique_ptr<char[]> buf(new char[file_size]);
+
+    batch.seekg(0, std::ios::beg);
+    batch.read(buf.get(), file_size);
+    batch.close();
+
+    std::vector<uint8_t> labels;
+    std::vector<torch::Tensor> images;
+    labels.reserve(kBatchSize);
+    images.reserve(kBatchSize);
+
+    for (size_t i = 0; i < kBatchSize; i++) {
+        uint8_t label = buf[i * (kImageSize + kLabelSize)];
+        std::vector<uint8_t> image;
+        image.reserve(kImageSize);
+        std::copy(&buf[i * (kImageSize + kLabelSize) + 1], &buf[i * (kImageSize + kLabelSize) + kImageSize], std::back_inserter(image));
+        labels.push_back(label);
+        auto image_tensor = torch::from_blob(image.data(),
+                                             {kImageChannels, kImageDim, kImageDim},
+                                             torch::TensorOptions().dtype(torch::kU8)).to(torch::kF32);
+        images.push_back(image_tensor);
+    }
+
+    auto labels_tensor = torch::from_blob(labels.data(),
+                                          {kBatchSize},
+                                          torch::TensorOptions().dtype(torch::kU8)).to(torch::kF32);
+    assert(labels_tensor.size(0) == kBatchSize);
+
+    auto images_tensor = torch::stack(images);
+    assert(images_tensor.size(0) == kBatchSize);
+
+    return std::make_pair(images_tensor, labels_tensor);
+}
+
+std::pair<torch::Tensor, torch::Tensor> read_train_data(const std::string& root) {
+    std::vector<torch::Tensor> images, targets;
+    for(uint32_t i = 1; i <= 5; i++) {
+        std::stringstream ss;
+        ss << root << '/' << kTrainFilenamePrefix << i << ".bin";
+        auto batch = read_batch(ss.str());
+        images.push_back(batch.first);
+        targets.push_back(batch.second);
+    }
+
+    torch::Tensor image_tensor = std::accumulate(++images.begin(), images.end(), *images.begin(), [&](torch::Tensor a, torch::Tensor b) {return torch::cat({a, b}, 0);});
+    torch::Tensor target_tensor = std::accumulate(++targets.begin(), targets.end(), *targets.begin(), [&](torch::Tensor a, torch::Tensor b) {return torch::cat({a, b}, 0);});
+
+    return std::make_pair(image_tensor, target_tensor);
+}
+
+std::pair<torch::Tensor, torch::Tensor> read_test_data(const std::string& root) {
+    std::stringstream ss;
+    ss << root << '/' << kTestFilename;
+    return read_batch(ss.str());
+}
+}
+
+CIFAR10::CIFAR10(const std::string& root, Mode mode)
+    : mode_(mode) {
+
+    std::pair<torch::Tensor, torch::Tensor> data;
+    if (mode_ == Mode::kTrain) {
+        data = read_train_data(root);
+    } else {
+        data = read_test_data(root);
+    }
+
+    images_ = std::move(data.first);
+    targets_ = std::move(data.second);
+    assert(images_.sizes()[0] == images_.sizes()[0]);
+}
+
+torch::data::Example<> CIFAR10::get(size_t index) {
+    return {images_[index], targets_[index]};
+}
+
+c10::optional<size_t> CIFAR10::size() const {
+    return images_.size(0);
+}
+
+bool CIFAR10::is_train() const noexcept {
+    return mode_ == Mode::kTrain;
+}
+
+const torch::Tensor& CIFAR10::images() const {
+    return images_;
+}
+
+const torch::Tensor& CIFAR10::targets() const {
+    return targets_;
+}
+
+CIFAR10&& CIFAR10::use_subset(int64_t new_size) {
+    assert(new_size <= images_.sizes()[0]);
+    images_ = images_.slice(0, 0, new_size);
+    targets_ = targets_.slice(0, 0, new_size);
+    return std::move(*this);
+}
+
+} // namespace datasets
+

tests/accuracy/datasets/cifar10.h

@@ -0,0 +1,45 @@
+#pragma once
+
+#include "torch/data/datasets/base.h"
+#include "torch/data/example.h"
+#include "torch/types.h"
+
+#include <cstddef>
+#include <string>
+
+namespace datasets {
+// The CIFAR10 Dataset
+class CIFAR10 : public torch::data::datasets::Dataset<CIFAR10> {
+public:
+    // The mode in which the dataset is loaded
+    enum class Mode { kTrain, kTest };
+
+    // Loads CIFAR10 from un-tarred file
+    // Dataset can be found https://www.cs.toronto.edu/~kriz/cifar-10-binary.tar.gz
+    // Root path should be the directory that contains the content of tarball
+    explicit CIFAR10(const std::string& root, Mode mode = Mode::kTrain);
+
+    // Returns the pair at index in the dataset
+    torch::data::Example<> get(size_t index) override;
+
+    // The size of the dataset
+    c10::optional<size_t> size() const override;
+
+    // The mode the dataset is in
+    bool is_train() const noexcept;
+
+    // Returns all images stacked into a single tensor
+    const torch::Tensor& images() const;
+
+    // Returns all targets stacked into a single tensor
+    const torch::Tensor& targets() const;
+
+    // Trims the dataset to the first n pairs
+    CIFAR10&& use_subset(int64_t new_size);
+
+
+private:
+    Mode mode_;
+    torch::Tensor images_, targets_;
+};
+} // namespace datasets

tests/accuracy/test_fp16_accuracy.cpp

@@ -0,0 +1,56 @@
+#include "accuracy_test.h"
+#include "datasets/cifar10.h"
+#include "torch/torch.h"
+
+TEST_P(AccuracyTests, FP16AccuracyIsClose) {
+    auto eval_dataset = datasets::CIFAR10("tests/accuracy/datasets/data/cifar-10-batches-bin/", datasets::CIFAR10::Mode::kTest)
+                                    .use_subset(3200)
+                                    .map(torch::data::transforms::Normalize<>({0.4914, 0.4822, 0.4465},
+                                                                              {0.2023, 0.1994, 0.2010}))
+                                    .map(torch::data::transforms::Stack<>());
+    auto eval_dataloader = torch::data::make_data_loader(std::move(eval_dataset), torch::data::DataLoaderOptions()
+                                                                                                    .batch_size(32)
+                                                                                                    .workers(2));
+
+    // Check the FP32 accuracy in JIT
+    torch::Tensor jit_correct = torch::zeros({1}, {torch::kCUDA}), jit_total = torch::zeros({1}, {torch::kCUDA});
+    for (auto batch : *eval_dataloader) {
+        auto images = batch.data.to(torch::kCUDA);
+        auto targets = batch.target.to(torch::kCUDA);
+
+        auto outputs = mod.forward({images});
+        auto predictions = std::get<1>(torch::max(outputs.toTensor(), 1, false));
+
+        jit_total += targets.sizes()[0];
+        jit_correct += torch::sum(torch::eq(predictions, targets));
+    }
+    torch::Tensor jit_accuracy = jit_correct / jit_total;
+
+    std::vector<std::vector<int64_t>> input_shape = {{32, 3, 32, 32}};
+    auto extra_info = trtorch::ExtraInfo({input_shape});
+    extra_info.op_precision = torch::kF16;
+
+    auto trt_mod = trtorch::CompileGraph(mod, extra_info);
+
+    torch::Tensor trt_correct = torch::zeros({1}, {torch::kCUDA}), trt_total = torch::zeros({1}, {torch::kCUDA});
+    for (auto batch : *eval_dataloader) {
+        auto images = batch.data.to(torch::kCUDA).to(torch::kF16);
+        auto targets = batch.target.to(torch::kCUDA).to(torch::kF16);
+
+        auto outputs = trt_mod.forward({images});
+        auto predictions = std::get<1>(torch::max(outputs.toTensor(), 1, false));
+        predictions = predictions.reshape(predictions.sizes()[0]);
+
+        trt_total += targets.sizes()[0];
+        trt_correct += torch::sum(torch::eq(predictions, targets));
+    }
+
+    torch::Tensor trt_accuracy = trt_correct / trt_total;
+
+    ASSERT_TRUE(trtorch::tests::util::almostEqual(jit_accuracy, trt_accuracy, 3));
+}
+
+
+INSTANTIATE_TEST_SUITE_P(FP16AccuracyIsCloseSuite,
+                         AccuracyTests,
+                         testing::Values("tests/accuracy/vgg16_cifar10.jit.pt"));