basic.cpp

#include "ggml.h"
#include "ggml-easy.h"
#include <iostream>

/**
 * This example demonstrates how to perform matrix multiplication using ggml-easy.h
 * 
 * Given 2 matrices A and B, the result matrix C is calculated as follows:
 *   C = (A x B) * 2
 *
 * We will use utils.debug_print() to debug the intermediate result of (A x B)
 * Then, we will use utils.mark_output() to get the final result of C
 *
 * The final result can be printed using ggml_easy::debug::print_tensor_data()
 * Or, can be used to perform further computations
 */

int main() {
    ggml_easy::ctx_params params;
    ggml_easy::ctx ctx(params);

    // initialize data of matrices to perform matrix multiplication
    const int rows_A = 4, cols_A = 2;
    float matrix_A[rows_A * cols_A] = {
        2, 8,
        5, 1,
        4, 2,
        8, 6
    };
    const int rows_B = 3, cols_B = 2;
    float matrix_B[rows_B * cols_B] = {
        10, 5,
        9, 9,
        5, 4
    };

    // create cgraph
    ctx.build_graph([&](ggml_context * ctx_gf, ggml_cgraph * gf, auto & utils) {
        ggml_tensor * a = utils.new_input("a", GGML_TYPE_F32, cols_A, rows_A);
        ggml_tensor * b = utils.new_input("b", GGML_TYPE_F32, cols_B, rows_B);
        ggml_tensor * a_mul_b = ggml_mul_mat(ctx_gf, a, b);
        utils.debug_print(a_mul_b, "a_mul_b");
        ggml_tensor * result = ggml_scale(ctx_gf, a_mul_b, 2);
        utils.mark_output(result, "result");
    });

    // set data
    ctx.set_tensor_data("a", matrix_A);
    ctx.set_tensor_data("b", matrix_B);

    // optional: print backend buffer info
    ggml_easy::debug::print_backend_buffer_info(ctx);

    // compute
    ggml_status status = ctx.compute();
    if (status != GGML_STATUS_SUCCESS) {
        std::cerr << "error: ggml compute return status: " << status << std::endl;
        return 1;
    }

    // get result
    auto result = ctx.get_tensor_data("result");
    ggml_tensor * result_tensor        = result.first;
    std::vector<uint8_t> & result_data = result.second;

    // print result
    ggml_easy::debug::print_tensor_data(result_tensor, result_data.data());

    return 0;
}