#include <filesystem>
namespace fs = std::filesystem;
#include "data.hpp"
#include "toolbox.hpp"
#include "config.hpp"
#include "gpu_unit_test.hpp"
#include "toolbox_unit_test.hpp"
#include "ViolaJones.hpp"
#include "ViolaJonesGPU.hpp"
#include "ViolaJonesCPU.hpp"

#if GPU_BOOSTED
#define LABEL "GPU"
#define apply_features apply_features_gpu
#define set_integral_image set_integral_image_gpu
#define argsort_2d argsort_2d_gpu
#else
#define LABEL "CPU"
#define apply_features apply_features_cpu
#define set_integral_image set_integral_image_cpu
#define argsort_2d argsort_2d_cpu
#endif

std::tuple<np::Array<int32_t>, np::Array<uint16_t>, np::Array<uint8_t>, np::Array<int32_t>, np::Array<uint8_t>> preprocessing() {
	// Creating state saver folders if they don't exist already
	if (SAVE_STATE)
		for (const char* const folder_name : { "models", "out" })
			fs::create_directory(folder_name);

	printf("| %-49s | %-18s | %-29s |\n", "Preprocessing", "Time spent (ns)", "Formatted time spent");
	printf("|%s|%s|%s|\n", S(51), S(20), S(31));

	const auto [ X_train, y_train, X_test, y_test ] = state_saver<uint8_t, 4>("Loading sets", {"X_train", "y_train", "X_test", "y_test"},
			FORCE_REDO, SAVE_STATE, OUT_DIR, load_datasets);

#if __DEBUG
	print("X_train");
	print(X_train.shape);
	print(X_train, { IDX_INSPECT });
	print("X_test");
	print(X_test.shape);
	print(X_test, { IDX_INSPECT });
	print("y_train");
	print(y_train.shape);
	print(y_train, { IDX_INSPECT, IDX_INSPECT + IDX_INSPECT_OFFSET });
	print("y_test");
	print(y_test.shape);
	print(y_test, { IDX_INSPECT, IDX_INSPECT + IDX_INSPECT_OFFSET });
#endif

	const np::Array<uint8_t> feats = state_saver<uint8_t>("Building features", "feats",
			FORCE_REDO, SAVE_STATE, OUT_DIR, build_features, X_train.shape[1], X_train.shape[2]);

#if __DEBUG
	print("feats");
	print(feats.shape);
	print_feat(feats, { IDX_INSPECT });
#endif

	const np::Array<uint32_t> X_train_ii = state_saver<uint32_t>("Converting training set to integral images (" LABEL ")", "X_train_ii_" LABEL,
			FORCE_REDO, SAVE_STATE, OUT_DIR, set_integral_image, X_train);
	const np::Array<uint32_t> X_test_ii = state_saver<uint32_t>("Converting testing set to integral images (" LABEL ")", "X_test_ii_" LABEL,
			FORCE_REDO, SAVE_STATE, OUT_DIR, set_integral_image, X_test);

#if __DEBUG
	print("X_train_ii");
	print(X_train_ii.shape);
	print(X_train_ii, { IDX_INSPECT });
	print("X_test_ii");
	print(X_test_ii.shape);
	print(X_test_ii, { IDX_INSPECT });
#endif

	const np::Array<int32_t> X_train_feat = state_saver<int32_t>("Applying features to training set (" LABEL ")", "X_train_feat_" LABEL,
			FORCE_REDO, SAVE_STATE, OUT_DIR, apply_features, feats, X_train_ii);
	const np::Array<int32_t> X_test_feat = state_saver<int32_t>("Applying features to testing set (" LABEL ")", "X_test_feat_" LABEL,
			FORCE_REDO, SAVE_STATE, OUT_DIR, apply_features, feats, X_test_ii);

#if __DEBUG
	print("X_train_feat");
	print(X_train_feat.shape);
	print(X_train_feat, { IDX_INSPECT, IDX_INSPECT + IDX_INSPECT_OFFSET });
	print("X_test_feat");
	print(X_test_feat.shape);
	print(X_test_feat, { IDX_INSPECT, IDX_INSPECT + IDX_INSPECT_OFFSET });
#endif

	// const Array<int> indices = measure_time_save<Array<int>>("Selecting best features", "indices", select_percentile, X_train_feat, d.y_train);
	// const Array<int> indices = measure_time<Array<int>>("Selecting best features", select_percentile, X_train_feat, d.y_train);

#if __DEBUG
	// print_feature(indices);
#endif

	const np::Array<uint16_t> X_train_feat_argsort = state_saver<uint16_t>("Precalculating training set argsort (" LABEL ")", "X_train_feat_argsort_" LABEL,
			FORCE_REDO, SAVE_STATE, OUT_DIR, argsort_2d, X_train_feat);

#if __DEBUG
	print("X_train_feat_argsort");
	print(X_train_feat_argsort.shape);
	print(X_train_feat_argsort, { IDX_INSPECT, IDX_INSPECT + IDX_INSPECT_OFFSET });
#endif

	const np::Array<uint16_t> X_test_feat_argsort = state_saver<uint16_t>("Precalculating testing set argsort (" LABEL ")", "X_test_feat_argsort_" LABEL,
			FORCE_REDO, SAVE_STATE, OUT_DIR, argsort_2d, X_test_feat);

#if __DEBUG
	print("X_test_feat_argsort");
	print(X_test_feat_argsort.shape);
	print(X_test_feat_argsort, { IDX_INSPECT, IDX_INSPECT + IDX_INSPECT_OFFSET });
#endif

	return { X_train_feat, X_train_feat_argsort, y_train, X_test_feat, y_test };
}

void train(const np::Array<int32_t>& X_train_feat, const np::Array<uint16_t>& X_train_feat_argsort, const np::Array<uint8_t>& y_train) {
	printf("\n| %-49s | %-18s | %-29s |\n", "Training", "Time spent (ns)", "Formatted time spent");
	printf("|%s|%s|%s|\n", S(51), S(20), S(31));

	for (const size_t T : TS) {
		char title[BUFFER_SIZE] = { 0 };
		char alphas_title[BUFFER_SIZE] = { 0 };
		char final_classifiers_title[BUFFER_SIZE] = { 0 };
		sprintf(title, "ViolaJones T = %-4lu (%s)", T, LABEL);
		sprintf(alphas_title, "alphas_%lu_%s", T, LABEL);
		sprintf(final_classifiers_title, "final_classifiers_%lu_%s", T, LABEL);

#if __DEBUG
		const auto [ alphas, final_classifiers ] = state_saver<float64_t, 2>(title, { alphas_title, final_classifiers_title },
#else
		state_saver<float64_t, 2>(title, { alphas_title, final_classifiers_title },
#endif
				FORCE_REDO, SAVE_STATE, MODEL_DIR, train_viola_jones, T, X_train_feat, X_train_feat_argsort, y_train);
#if __DEBUG
		print("alphas");
		print(alphas);
		print("final_classifiers");
		print(final_classifiers);
#endif
	}
}

void testing_and_evaluating(const np::Array<int32_t>& X_train_feat, const np::Array<uint8_t>& y_train, const np::Array<int32_t>& X_test_feat, const np::Array<uint8_t>& y_test) {
	printf("\n| %-26s | Time spent (ns) (E) | %-29s | Time spent (ns) (T) | %-29s |\n", "Testing", "Formatted time spent (E)", "Formatted time spent (T)");
	printf("|%s|%s|%s|%s|%s|\n", S(28), S(21), S(31), S(21), S(31));

	constexpr const size_t NT = sizeof(TS) / sizeof(size_t);
	std::array<std::array<float64_t, 8>, NT> results;

	size_t i = 0;
	for (const size_t T : TS) {
		char title[BUFFER_SIZE] = { 0 };
		char alphas_title[BUFFER_SIZE] = { 0 };
		char final_classifiers_title[BUFFER_SIZE] = { 0 };
		sprintf(title, "ViolaJones T = %-4lu (%s)", T, LABEL);
		sprintf(alphas_title, MODEL_DIR "/alphas_%lu_%s.bin", T, LABEL);
		sprintf(final_classifiers_title, MODEL_DIR "/final_classifiers_%lu_%s.bin", T, LABEL);

		const np::Array<float64_t> alphas = load<float64_t>(alphas_title);
		const np::Array<float64_t> final_classifiers = load<float64_t>(final_classifiers_title);

		std::chrono::system_clock::time_point start = perf_counter_ns();
		const np::Array<uint8_t> y_pred_train = classify_viola_jones(alphas, final_classifiers, X_train_feat);
		const long long t_pred_train = duration_ns(perf_counter_ns() - start);
		const float64_t e_acc = accuracy_score(y_train, y_pred_train);
		const float64_t e_f1 = f1_score(y_train, y_pred_train);
		float64_t e_FN, e_FP;
		std::tie(std::ignore, e_FN, e_FP, std::ignore) = confusion_matrix(y_train, y_pred_train);

		start = perf_counter_ns();
		const np::Array<uint8_t> y_pred_test = classify_viola_jones(alphas, final_classifiers, X_test_feat);
		const long long t_pred_test = duration_ns(perf_counter_ns() - start);
		const float64_t t_acc = accuracy_score(y_test, y_pred_test);
		const float64_t t_f1 = f1_score(y_test, y_pred_test);
		float64_t t_FN, t_FP;
		std::tie(std::ignore, t_FN, t_FP, std::ignore) = confusion_matrix(y_test, y_pred_test);
		results[i++] = { e_acc, e_f1, e_FN, e_FP, t_acc, t_f1, t_FN, t_FP };

		printf("| %-26s | %'19lld | %-29s | %'19lld | %-29s |\n", title, t_pred_train, format_time_ns(t_pred_train).c_str(), t_pred_test, format_time_ns(t_pred_test).c_str());
	}

	printf("\n| %-19s | ACC (E) | F1 (E) | FN (E) | FP (E) | ACC (T) | F1 (T) | FN (T) | FP (T) |\n", "Evaluating");
	printf("|%s|%s|%s|%s|%s|%s|%s|%s|%s|\n", S(21), S(9), S(8), S(8), S(8), S(9), S(8), S(8), S(8));

	i = 0;
	for (const size_t T : TS) {
		char title[BUFFER_SIZE] = { 0 };
		sprintf(title, "ViolaJones T = %-4lu", T);
		const auto [e_acc, e_f1, e_FN, e_FP, t_acc, t_f1, t_FN, t_FP] = results[i++];
		printf("| %-19s | %'6.2f%% | %'6.2f | %'6.0f | %'6.0f | %6.2f%% | %'6.2f | %'6.0f | %'6.0f |\n", title, e_acc * 100, e_f1, e_FN, e_FP, t_acc * 100, t_f1, t_FN, t_FP);
	}
}

void unit_test(void) {
	printf("\n| %-37s | %-10s | %-18s | %-29s |\n", "Unit testing", "Test state", "Time spent (ns)", "Formatted time spent");
	printf("|%s|%s|%s|%s|\n", S(39), S(12), S(20), S(31));

	char title[BUFFER_SIZE] = { 0 };
	char tmp_title[BUFFER_SIZE / 2] = { 0 };
	char file_cpu[BUFFER_SIZE] = { 0 };
	char file_gpu[BUFFER_SIZE] = { 0 };
	const std::chrono::system_clock::time_point fnc_s = perf_counter_ns();
	uint64_t n_total = 0, n_success = 0;

	auto test_fnc = [&n_total, &n_success](const char* title, const auto& fnc) {
		++n_total;
		const std::chrono::system_clock::time_point start = perf_counter_ns();
		const bool state = fnc();
		const long long time_spent = duration_ns(perf_counter_ns() - start);
		if(state){
			printf("| %-37s | %10s | %18s | %-29s |\n", title, "Passed", thousand_sep(time_spent).c_str(), format_time_ns(time_spent).c_str());
			++n_success;
		} else
			printf("| %-37s | %10s | %18s | %-29s |\n", title, "Failed", thousand_sep(time_spent).c_str(), format_time_ns(time_spent).c_str());
	};

	for (const char* label : { "train", "test" }) {
		sprintf(file_cpu, OUT_DIR "/X_%s_ii_CPU.bin", label);
		sprintf(file_gpu, OUT_DIR "/X_%s_ii_GPU.bin", label);
		if (fs::exists(file_cpu) && fs::exists(file_gpu)) {
			const np::Array<uint32_t> X_train_ii_cpu = load<uint32_t>(file_cpu);
			const np::Array<uint32_t> X_train_ii_gpu = load<uint32_t>(file_gpu);
			sprintf(tmp_title, "X_%s_ii", label);
			sprintf(title, "%-22s - CPU  vs GPU", tmp_title);
			test_fnc(title, [&X_train_ii_cpu, &X_train_ii_gpu]{ return unit_test_cpu_vs_gpu<uint32_t>(X_train_ii_cpu, X_train_ii_gpu); });
		}
		char file_feat[BUFFER_SIZE] = { 0 };
		sprintf(file_feat, OUT_DIR "/X_%s_feat_CPU.bin", label);
		if (fs::exists(file_feat)) {
			const np::Array<int32_t> X_feat = load<int32_t>(file_feat);
			sprintf(file_gpu, OUT_DIR "/X_%s_feat_GPU.bin", label);
			if (fs::exists(file_gpu)) {
				const np::Array<int32_t> X_feat_gpu = load<int32_t>(file_gpu);
				sprintf(tmp_title, "X_%s_feat", label);
				sprintf(title, "%-22s - CPU  vs GPU", tmp_title);
				test_fnc(title, [&X_feat, &X_feat_gpu]{ return unit_test_cpu_vs_gpu<int32_t>(X_feat, X_feat_gpu); });
			}
			sprintf(file_cpu, OUT_DIR "/X_%s_feat_argsort_CPU.bin", label);
			np::Array<uint16_t> X_feat_argsort_cpu;
			uint8_t loaded = 0;
			if (fs::exists(file_cpu)) {
				X_feat_argsort_cpu = std::move(load<uint16_t>(file_cpu));
				++loaded;
				sprintf(tmp_title, "X_%s_feat_argsort", label);
				sprintf(title, "%-22s - CPU  argsort", tmp_title);
				test_fnc(title, [&X_feat, &X_feat_argsort_cpu]{ return unit_test_argsort_2d<int32_t>(X_feat, X_feat_argsort_cpu); });
			}
			sprintf(file_gpu, OUT_DIR "/X_%s_feat_argsort_GPU.bin", label);
			np::Array<uint16_t> X_feat_argsort_gpu;
			if (fs::exists(file_gpu)) {
				X_feat_argsort_gpu = std::move(load<uint16_t>(file_gpu));
				++loaded;
				sprintf(tmp_title, "X_%s_feat_argsort", label);
				sprintf(title, "%-22s - GPU  argsort", tmp_title);
				test_fnc(title, [&X_feat, &X_feat_argsort_gpu]{ return unit_test_argsort_2d<int32_t>(X_feat, X_feat_argsort_gpu); });
			}
			if (loaded == 2){
				sprintf(tmp_title, "X_%s_feat_argsort", label);
				sprintf(title, "%-22s - CPU  vs GPU", tmp_title);
				test_fnc(title, [&X_feat_argsort_cpu, &X_feat_argsort_gpu]{ return unit_test_cpu_vs_gpu<uint16_t>(X_feat_argsort_cpu, X_feat_argsort_gpu); });
			}
		}
	}

	for (const size_t T : TS)
		for (const char* label : { "alphas", "final_classifiers" }) {
			sprintf(file_cpu, MODEL_DIR "/%s_%lu_CPU.bin", label, T);
			sprintf(file_gpu, MODEL_DIR "/%s_%lu_GPU.bin", label, T);
			if (fs::exists(file_cpu) && fs::exists(file_gpu)){
				const np::Array<float64_t> cpu = load<float64_t>(file_cpu);
				const np::Array<float64_t> gpu = load<float64_t>(file_gpu);
				sprintf(tmp_title, "%s_%ld", label, T);
				sprintf(title, "%-22s - CPU  vs GPU", tmp_title);
				test_fnc(title, [&cpu, &gpu]{ return unit_test_cpu_vs_gpu<float64_t>(cpu, gpu); });
			}
		}

	const long long time_spent = duration_ns(perf_counter_ns() - fnc_s);
	sprintf(title, "%ld/%ld", n_success, n_total);

	printf("|%s|%s|%s|%s|\n", S(39), S(12), S(20), S(31));
	printf("| %-37s | %10s | %18s | %-29s |\n", "Unit testing summary", title, thousand_sep(time_spent).c_str(), format_time_ns(time_spent).c_str());
}

int main(){
	setlocale(LC_NUMERIC, ""); // Allow proper number display

	printf("| %-49s | %-18s | %-29s |\n", "Unit testing", "Time spent (ns)", "Formatted time spent");
	printf("|%s|%s|%s|\n", S(51), S(20), S(31));
#if GPU_BOOSTED
	benchmark_function_void("Testing GPU capabilities 1D", test_working, 3 + (1<<29));
	benchmark_function_void("Testing GPU capabilities 2D", test_working_2d, 3 + (1<<15), 2 + (1<<14));
	benchmark_function_void("Testing GPU capabilities 3D", test_working_3d, 9 + (1<<10), 5 + (1<<10), 7 + (1<<9));
#endif
	benchmark_function_void("Testing format_time", format_time_test);
	benchmark_function_void("Testing format_time_ns", format_time_ns_test);
	benchmark_function_void("Testing format_byte_size", format_byte_size_test);
	benchmark_function_void("Testing thousand_sep", thousand_sep_test);
	printf("\n");

	const auto [ X_train_feat, X_train_feat_argsort, y_train, X_test_feat, y_test ] = preprocessing();
	train(X_train_feat, X_train_feat_argsort, y_train);
	testing_and_evaluating(X_train_feat, y_train, X_test_feat, y_test);
	unit_test();
	return EXIT_SUCCESS;
}