ff5a3f0c09
* Work on the BPE tokenizer Tokenizer tests work for Falcon-7B * Try to fix build problem * Fix debug assertion failure * Fix MSVC Unicode BOM problem * Cleanup and an improvement * Fix compiler warning * Cleanup * Test doesn't work over the full range of Unicodes * Update .gitignore and Makefile * Another Makefile rule * Testing Aquila * Moving byte decoding back to `token_to_piece` ... ... because everyone is using it. * Guarding some unusable code pathes * Streamlining code and adding some more assertions Important change: I'm classifying added tokens as control tokens now for BPE. * Adding a comment * Adding another assertion * Fixed vocabulary guarding assertions * Fix PR for recent change * Fix PR for recent change * Fix for compiler warning * Fix PR for recent change * Fix PR for recent change * Fix PR for recent change * Fix for compiler warning * Fixes for more compiler warnings * Remove unused code * Fix initialization of static maps * Add scores and token types back, adapt gptneox * Update llama.cpp Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> * Update unicode.h Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> * Update unicode.h Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> * Ported Starcoder and added some assertions * Fix coding style * Apply @jploski 's fix for missing tokens --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
105 lines
3.2 KiB
C++
105 lines
3.2 KiB
C++
#include "llama.h"
|
|
#include "common.h"
|
|
#include "unicode.h"
|
|
#include "console.h"
|
|
|
|
#include <cassert>
|
|
#include <cstdio>
|
|
#include <cstring>
|
|
#include <string>
|
|
#include <codecvt>
|
|
#include <map>
|
|
#include <vector>
|
|
#include <locale>
|
|
|
|
int main(int argc, char **argv) {
|
|
if (argc < 2) {
|
|
fprintf(stderr, "Usage: %s <vocab-file>\n", argv[0]);
|
|
return 1;
|
|
}
|
|
|
|
const std::string fname = argv[1];
|
|
|
|
fprintf(stderr, "%s : reading vocab from: '%s'\n", __func__, fname.c_str());
|
|
|
|
llama_model * model;
|
|
llama_context * ctx;
|
|
|
|
llama_backend_init(false);
|
|
|
|
// load the vocab
|
|
{
|
|
auto mparams = llama_model_default_params();
|
|
|
|
mparams.vocab_only = true;
|
|
|
|
model = llama_load_model_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_new_context_with_model(model, cparams);
|
|
|
|
if (ctx == NULL) {
|
|
fprintf(stderr, "%s: error: failed to load vocab '%s'\n", __func__, fname.c_str());
|
|
llama_free_model(model);
|
|
return 1;
|
|
}
|
|
}
|
|
|
|
GGML_ASSERT(llama_vocab_type(model) == LLAMA_VOCAB_TYPE_SPM);
|
|
|
|
#ifdef _WIN32
|
|
// We need this for unicode console support
|
|
console::init(false, false);
|
|
atexit([]() { console::cleanup(); });
|
|
#endif
|
|
|
|
const int n_vocab = llama_n_vocab(model);
|
|
|
|
for (int i = 0; i < n_vocab; ++i) {
|
|
std::string str = llama_detokenize_spm(ctx, std::vector<int>(1, i));
|
|
std::vector<llama_token> tokens = llama_tokenize(ctx, str, false);
|
|
std::string check = llama_detokenize_spm(ctx, tokens);
|
|
if (check != str) {
|
|
fprintf(stderr, "%s : error: token %d detokenizes to '%s'(%zu) but tokenization of this detokenizes to '%s'(%zu)\n",
|
|
__func__, i, str.c_str(), str.length(), check.c_str(), check.length());
|
|
return 2;
|
|
}
|
|
}
|
|
|
|
for (uint32_t cp = 0x0000; cp < 0xffff; ++cp) {
|
|
if (cp < 0xd800 || cp > 0xdfff) {
|
|
std::string str = codepoint_to_utf8(cp);
|
|
std::vector<llama_token> tokens = llama_tokenize(ctx, str, false);
|
|
std::string check = llama_detokenize_spm(ctx, tokens);
|
|
if (cp != 9601 && str != check) {
|
|
fprintf(stderr, "%s : error: codepoint %d detokenizes to '%s'(%zu) instead of '%s'(%zu)\n",
|
|
__func__, cp, check.c_str(), check.length(), str.c_str(), str.length());
|
|
return 3;
|
|
}
|
|
}
|
|
}
|
|
for (uint32_t cp = 0x10000; cp < 0x0010ffff; ++cp) {
|
|
std::string str = codepoint_to_utf8(cp);
|
|
std::vector<llama_token> tokens = llama_tokenize(ctx, str, false);
|
|
std::string check = llama_detokenize_spm(ctx, tokens);
|
|
if (str != check) {
|
|
fprintf(stderr, "%s : error: codepoint %d detokenizes to '%s'(%zu) instead of '%s'(%zu)\n",
|
|
__func__, cp, check.c_str(), check.length(), str.c_str(), str.length());
|
|
return 4;
|
|
}
|
|
}
|
|
|
|
llama_free_model(model);
|
|
llama_free(ctx);
|
|
|
|
llama_backend_free();
|
|
|
|
return 0;
|
|
}
|