Function main

subprojects/llama.cpp/tests/test-tokenizer-1-bpe.cpp:17–155

#include <atomic>

int main(int argc, char **argv) {
    if (argc < 2 || argc > 3) {
        fprintf(stderr, "Usage: %s <vocab-file> [--ignore-merges]\n", argv[0]);
        return 1;
    }

    const std::string fname = argv[1];
    bool ignore_merges = false;
    if (argc == 3) {
        if (std::strcmp(argv[2], "--ignore-merges") != 0) {
            fprintf(stderr, "Usage: %s <vocab-file> [--ignore-merges]\n", argv[0]);
            return 1;
        }
        ignore_merges = true;
    }

    fprintf(stderr, "%s : reading vocab from: '%s'\n", __func__, fname.c_str());

    if (ignore_merges) {
        fprintf(stderr, "%s : ignoring merges for tokens inside vocab\n", __func__);
    }

    llama_model * model;
    llama_context * ctx;

    llama_backend_init();

    // load the vocab
    {
        auto mparams = llama_model_default_params();

        mparams.vocab_only = true;

        model = llama_model_load_from_file(fname.c_str(), mparams);

        if (model == NULL) {
            fprintf(stderr, "%s: error: failed to load vocab '%s'\n", __func__, fname.c_str());
            return 1;
        }

        auto cparams = llama_context_default_params();

        ctx = llama_init_from_model(model, cparams);

        if (ctx == NULL) {
            fprintf(stderr, "%s: error: failed to load vocab '%s'\n", __func__, fname.c_str());
            llama_model_free(model);
            return 1;
        }
    }

    const llama_vocab * vocab = llama_model_get_vocab(model);

    //GGML_ASSERT(llama_vocab_type(vocab) == LLAMA_VOCAB_TYPE_BPE);
    if (llama_vocab_type(vocab) != LLAMA_VOCAB_TYPE_BPE) {
        return 99;
    }