lora.hpp

#ifndef __LORA_HPP__
#define __LORA_HPP__

#include "ggml_extend.hpp"

#define LORA_GRAPH_SIZE 10240

struct LoraModel : public GGMLModule {
    float multiplier = 1.0f;
    std::map<std::string, struct ggml_tensor*> lora_tensors;
    std::string file_path;
    ModelLoader model_loader;
    bool load_failed = false;
    bool applied     = false;

    LoraModel(ggml_backend_t backend,
              ggml_type wtype,
              const std::string& file_path = "",
              const std::string& prefix    = "")
        : file_path(file_path), GGMLModule(backend, wtype) {
        if (!model_loader.init_from_file(file_path, prefix)) {
            load_failed = true;
        }
    }

    std::string get_desc() {
        return "lora";
    }

    bool load_from_file(bool filter_tensor = false) {
        LOG_INFO("loading LoRA from '%s'", file_path.c_str());

        if (load_failed) {
            LOG_ERROR("init lora model loader from file failed: '%s'", file_path.c_str());
            return false;
        }

        bool dry_run          = true;
        auto on_new_tensor_cb = [&](const TensorStorage& tensor_storage, ggml_tensor** dst_tensor) -> bool {
            const std::string& name = tensor_storage.name;

            if (filter_tensor && !contains(name, "lora")) {
                // LOG_INFO("skipping LoRA tesnor '%s'", name.c_str());
                return true;
            }

            if (dry_run) {
                struct ggml_tensor* real = ggml_new_tensor(params_ctx,
                                                           tensor_storage.type,
                                                           tensor_storage.n_dims,
                                                           tensor_storage.ne);
                lora_tensors[name]       = real;
            } else {
                auto real   = lora_tensors[name];
                *dst_tensor = real;
            }

            return true;
        };

        model_loader.load_tensors(on_new_tensor_cb, backend);
        alloc_params_buffer();

        dry_run = false;
        model_loader.load_tensors(on_new_tensor_cb, backend);

        LOG_DEBUG("finished loaded lora");
        return true;
    }

    struct ggml_cgraph* build_lora_graph(std::map<std::string, struct ggml_tensor*> model_tensors) {
        struct ggml_cgraph* gf = ggml_new_graph_custom(compute_ctx, LORA_GRAPH_SIZE, false);

        std::set<std::string> applied_lora_tensors;
        for (auto it : model_tensors) {
            std::string k_tensor       = it.first;
            struct ggml_tensor* weight = model_tensors[it.first];

            size_t k_pos = k_tensor.find(".weight");
            if (k_pos == std::string::npos) {
                continue;
            }
            k_tensor = k_tensor.substr(0, k_pos);
            replace_all_chars(k_tensor, '.', '_');
            // LOG_DEBUG("k_tensor %s", k_tensor.c_str());
            std::string lora_up_name = "lora." + k_tensor + ".lora_up.weight";
            if (lora_tensors.find(lora_up_name) == lora_tensors.end()) {
                if (k_tensor == "model_diffusion_model_output_blocks_2_2_conv") {
                    // fix for some sdxl lora, like lcm-lora-xl
                    k_tensor     = "model_diffusion_model_output_blocks_2_1_conv";
                    lora_up_name = "lora." + k_tensor + ".lora_up.weight";
                }
            }

            std::string lora_down_name = "lora." + k_tensor + ".lora_down.weight";
            std::string alpha_name     = "lora." + k_tensor + ".alpha";
            std::string scale_name     = "lora." + k_tensor + ".scale";

            ggml_tensor* lora_up   = NULL;
            ggml_tensor* lora_down = NULL;

            if (lora_tensors.find(lora_up_name) != lora_tensors.end()) {
                lora_up = lora_tensors[lora_up_name];
            }

            if (lora_tensors.find(lora_down_name) != lora_tensors.end()) {
                lora_down = lora_tensors[lora_down_name];
            }

            if (lora_up == NULL || lora_down == NULL) {
                continue;
            }

            applied_lora_tensors.insert(lora_up_name);
            applied_lora_tensors.insert(lora_down_name);
            applied_lora_tensors.insert(alpha_name);
            applied_lora_tensors.insert(scale_name);

            // calc_cale
            int64_t dim       = lora_down->ne[ggml_n_dims(lora_down) - 1];
            float scale_value = 1.0f;
            if (lora_tensors.find(scale_name) != lora_tensors.end()) {
                scale_value = ggml_backend_tensor_get_f32(lora_tensors[scale_name]);
            } else if (lora_tensors.find(alpha_name) != lora_tensors.end()) {
                float alpha = ggml_backend_tensor_get_f32(lora_tensors[alpha_name]);
                scale_value = alpha / dim;
            }
            scale_value *= multiplier;

            // flat lora tensors to multiply it
            int64_t lora_up_rows   = lora_up->ne[ggml_n_dims(lora_up) - 1];
            lora_up                = ggml_reshape_2d(compute_ctx, lora_up, ggml_nelements(lora_up) / lora_up_rows, lora_up_rows);
            int64_t lora_down_rows = lora_down->ne[ggml_n_dims(lora_down) - 1];
            lora_down              = ggml_reshape_2d(compute_ctx, lora_down, ggml_nelements(lora_down) / lora_down_rows, lora_down_rows);

            // ggml_mul_mat requires tensor b transposed
            lora_down                  = ggml_cont(compute_ctx, ggml_transpose(compute_ctx, lora_down));
            struct ggml_tensor* updown = ggml_mul_mat(compute_ctx, lora_up, lora_down);
            updown                     = ggml_cont(compute_ctx, ggml_transpose(compute_ctx, updown));
            updown                     = ggml_reshape(compute_ctx, updown, weight);
            GGML_ASSERT(ggml_nelements(updown) == ggml_nelements(weight));
            updown = ggml_scale_inplace(compute_ctx, updown, scale_value);
            ggml_tensor* final_weight;
            // if (weight->type != GGML_TYPE_F32 && weight->type != GGML_TYPE_F16) {
            //     final_weight = ggml_new_tensor(compute_ctx, GGML_TYPE_F32, weight->n_dims, weight->ne);
            //     final_weight = ggml_cpy_inplace(compute_ctx, weight, final_weight);
            //     final_weight = ggml_add_inplace(compute_ctx, final_weight, updown);
            //     final_weight = ggml_cpy_inplace(compute_ctx, final_weight, weight);
            // } else {
            //     final_weight = ggml_add_inplace(compute_ctx, weight, updown);
            // }
            final_weight = ggml_add_inplace(compute_ctx, weight, updown);  // apply directly
            ggml_build_forward_expand(gf, final_weight);
        }

        size_t total_lora_tensors_count   = 0;
        size_t applied_lora_tensors_count = 0;

        for (auto& kv : lora_tensors) {
            total_lora_tensors_count++;
            if (applied_lora_tensors.find(kv.first) == applied_lora_tensors.end()) {
                LOG_WARN("unused lora tensor %s", kv.first.c_str());
            } else {
                applied_lora_tensors_count++;
            }
        }
        /* Don't worry if this message shows up twice in the logs per LoRA,
         * this function is called once to calculate the required buffer size
         * and then again to actually generate a graph to be used */
        if (applied_lora_tensors_count != total_lora_tensors_count) {
            LOG_WARN("Only (%lu / %lu) LoRA tensors have been applied",
                     applied_lora_tensors_count, total_lora_tensors_count);
        } else {
            LOG_DEBUG("(%lu / %lu) LoRA tensors applied successfully",
                      applied_lora_tensors_count, total_lora_tensors_count);
        }

        return gf;
    }

    void apply(std::map<std::string, struct ggml_tensor*> model_tensors, int n_threads) {
        auto get_graph = [&]() -> struct ggml_cgraph* {
            return build_lora_graph(model_tensors);
        };
        GGMLModule::compute(get_graph, n_threads, true);
    }
};

#endif  // __LORA_HPP__