our_knp/cats-dogs/source/parse_arguments.cpp

/**
 * @file parse_arguments.cpp
 * @brief Parsing of command line arguments.
 * @kaspersky_support D. Postnikov
 * @date 03.02.2026
 * @license Apache 2.0
 * @copyright © 2026 AO Kaspersky Lab
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

#include "parse_arguments.h"

#include <knp/framework/logging.h>

#include <iostream>
#include <string>

#include <boost/program_options.hpp>


namespace po = boost::program_options;

std::optional<ModelDescription> parse_arguments(int argc, char** argv)
{
    po::options_description desc("Usage");
    desc.add_options()("help,h", "print available options")(
        "model,m", po::value<std::string>()->default_value("blifat"), "model type: blifat or altai")(
        "train_iters,t", po::value<size_t>()->default_value(60000), "number of images for training")(
        "inference_iters,i", po::value<size_t>()->default_value(10000), "number of images for inference")(
        "images", po::value<std::string>()->default_value("MNIST.bin"), "path to raw images file")(
        "labels", po::value<std::string>()->default_value("MNIST.target"), "path to images labels file")(
        "log_path", po::value<std::string>()->default_value(""),
                "path for saving logs (if empty, no logs are saved)")(
        "inference_only", "skip training and use `model_path` to load a pre-trained model")(
        "training_backend", po::value<std::string>()->default_value("knp-cpu-single-threaded-backend"),
                "path to backend used for training")(
        "inference_backend", po::value<std::string>(),
                "path to backend for inference (if empty, `training_backend` is used)")(
        "extensive_logs_path", po::value<std::string>()->default_value(""),
                "path for storing extensive logs (if empty, no extensive logs are produced)")(
        "model_path", po::value<std::string>()->default_value(""),
                "path for saving trained model (if empty, model is not saved)")(
        "logging_level,l", po::value<std::string>()->default_value("info"),
        "logging level: trace, debug, info, warn, error, critical, or none");

    po::variables_map vm;
    po::store(po::parse_command_line(argc, argv, desc), vm);
    po::notify(vm);

    if (vm.count("help"))
    {
        std::cout << desc << std::endl;
        return std::nullopt;
    }

    ModelDescription model_desc;
    auto common_path = std::filesystem::weakly_canonical(std::filesystem::path(argv[0]).parent_path());

    if (vm.count("model"))
    {
        std::string model_type = vm["model"].as<std::string>();
        if (model_type == "blifat")
        {
            model_desc.type_ = SupportedModelType::BLIFAT;
        }
        else if (model_type == "altai")
        {
            model_desc.type_ = SupportedModelType::AltAI;
        }
        else
        {
            std::cout << "Not supported model type." << std::endl;
            std::cout << desc << std::endl;
            return std::nullopt;
        }
    }
    else
    {
        std::cout << "Model type not specified." << std::endl;
        std::cout << desc << std::endl;
        return std::nullopt;
    }

    if (vm.count("train_iters"))
    {
        model_desc.train_images_amount_ = vm["train_iters"].as<size_t>();
    }
    else
    {
        std::cout << "Train iterations not specified." << std::endl;
        std::cout << desc << std::endl;
        return std::nullopt;
    }

    if (vm.count("inference_iters"))
    {
        model_desc.inference_images_amount_ = vm["inference_iters"].as<size_t>();
    }
    else
    {
        std::cout << "Inference iterations not specified." << std::endl;
        std::cout << desc << std::endl;
        return std::nullopt;
    }

    if (vm.count("images"))
    {
        model_desc.images_file_path_ = vm["images"].as<std::string>();
    }
    else
    {
        std::cout << "Images path not specified." << std::endl;
        std::cout << desc << std::endl;
        return std::nullopt;
    }

    if (vm.count("labels"))
    {
        model_desc.labels_file_path_ = vm["labels"].as<std::string>();
    }
    else
    {
        std::cout << "Labels path not specified." << std::endl;
        std::cout << desc << std::endl;
        return std::nullopt;
    }

    if (vm.count("training_backend"))
    {
        model_desc.training_backend_path_ = common_path / vm["training_backend"].as<std::string>();
    }
    else
    {
        std::cout << "Training backend path not specified." << std::endl;
        std::cout << desc << std::endl;
        return std::nullopt;
    }

    if (!vm.count("inference_backend") || vm["inference_backend"].as<std::string>().empty())
    {
        model_desc.inference_backend_path_ = model_desc.training_backend_path_;
    }
    else
    {
        model_desc.inference_backend_path_ = common_path / vm["inference_backend"].as<std::string>();
    }

    if (vm.count("extensive_logs_path"))
    {
        model_desc.log_path_ = vm["extensive_logs_path"].as<std::string>();
    }
    else
    {
        model_desc.log_path_ = "";
    }

    if (vm.count("model_path"))
    {
        model_desc.model_saving_path_ = vm["model_path"].as<std::string>();
    }
    else
    {
        model_desc.model_saving_path_ = "";
    }

    model_desc.inference_only_ = vm.count("inference_only") > 0;

    if (vm.count("logging_level"))
    {
        knp::framework::logging::Level logging_level =
            knp::framework::logging::str_to_level(vm["logging_level"].as<std::string>());
        knp::framework::logging::set_level(logging_level);
        std::cout << "Set logging level to \"" << knp::framework::logging::level_to_str(logging_level) << "\""
                  << std::endl;
    }
    else
    {
        std::cout << "Logging level not specified." << std::endl;
        std::cout << desc << std::endl;
        return std::nullopt;
    }

    return model_desc;
}