merged the changes from deepseeker models to main branch
This commit is contained in:
Jaggzh
Georgi Gerganov
parent
83b72cb086
commit
6fbab2dbc8
15 changed files with 886 additions and 151 deletions
|
|
@ -41,21 +41,24 @@ llama_test(test-quantize-perf.cpp)
|
|||
llama_test(test-sampling.cpp)
|
||||
llama_test(test-chat-template.cpp)
|
||||
|
||||
llama_test(test-tokenizer-0-llama.cpp NAME test-tokenizer-0-llama ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-llama.gguf)
|
||||
llama_test(test-tokenizer-0-falcon.cpp NAME test-tokenizer-0-falcon ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-falcon.gguf)
|
||||
|
||||
llama_test(test-tokenizer-1-llama.cpp NAME test-tokenizer-1-llama ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-llama.gguf)
|
||||
llama_test(test-tokenizer-1-llama.cpp NAME test-tokenizer-1-baichuan ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-baichuan.gguf)
|
||||
llama_test(test-tokenizer-0-llama.cpp NAME test-tokenizer-0-llama ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-llama.gguf)
|
||||
llama_test(test-tokenizer-0-falcon.cpp NAME test-tokenizer-0-falcon ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-falcon.gguf)
|
||||
|
||||
llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-falcon ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-falcon.gguf)
|
||||
llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-aquila ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-aquila.gguf)
|
||||
llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-mpt ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-mpt.gguf)
|
||||
llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-stablelm-3b-4e1t ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-stablelm-3b-4e1t.gguf)
|
||||
llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-gpt-neox ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-gpt-neox.gguf)
|
||||
llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-refact ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-refact.gguf)
|
||||
llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-starcoder ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-starcoder.gguf)
|
||||
llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-gpt2 ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-gpt2.gguf)
|
||||
#llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-bloom ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-bloom.gguf) # BIG
|
||||
llama_test(test-tokenizer-0-deepseek-coder.cpp NAME test-tokenizer-0-deepseek-coder ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-deepseek-coder.gguf)
|
||||
|
||||
llama_test(test-tokenizer-1-llama.cpp NAME test-tokenizer-1-llama ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-llama.gguf)
|
||||
llama_test(test-tokenizer-1-llama.cpp NAME test-tokenizer-1-baichuan ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-baichuan.gguf)
|
||||
|
||||
llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-falcon ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-falcon.gguf)
|
||||
llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-aquila ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-aquila.gguf)
|
||||
llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-mpt ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-mpt.gguf)
|
||||
llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-stablelm-3b-4e1t ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-stablelm-3b-4e1t.gguf)
|
||||
llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-gpt-neox ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-gpt-neox.gguf)
|
||||
llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-refact ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-refact.gguf)
|
||||
llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-starcoder ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-starcoder.gguf)
|
||||
llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-gpt2 ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-gpt2.gguf)
|
||||
#llama_test(test-tokenizer-1-bpe.cpp NAME test-tokenizer-1-bloom ARGS ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-bloom.gguf) # BIG
|
||||
|
||||
llama_test(test-grammar-parser.cpp)
|
||||
llama_test(test-llama-grammar.cpp)
|
||||
|
|
|
|||
188
tests/test-tokenizer-0-deepseek-coder.cpp
Normal file
188
tests/test-tokenizer-0-deepseek-coder.cpp
Normal file
|
|
@ -0,0 +1,188 @@
|
|||
#include "llama.h"
|
||||
#include "common.h"
|
||||
#include "console.h"
|
||||
|
||||
#include <cstdio>
|
||||
#include <string>
|
||||
#include <map>
|
||||
#include <vector>
|
||||
#include <fstream>
|
||||
|
||||
// generate using test-tokenizer-0-falcon.py
|
||||
static const std::map<std::string, std::vector<llama_token>> & k_tests() {
|
||||
static std::map<std::string, std::vector<llama_token>> _k_tests = {
|
||||
{ "" , { }, },
|
||||
{ " " , { 207, }, },
|
||||
{ " " , { 243, }, },
|
||||
{ " " , { 315, }, },
|
||||
{ "\t" , { 184, }, },
|
||||
{ "\n" , { 185, }, },
|
||||
{ "\t\n" , { 184, 185, }, },
|
||||
{ "Hello world" , { 17535, 1835, }, },
|
||||
{ " Hello world" , { 414, 9489, 1835, }, },
|
||||
{ "Hello World" , { 17535, 5414, }, },
|
||||
{ " Hello World" , { 414, 9489, 5414, }, },
|
||||
{ " Hello World!" , { 414, 9489, 5414, 0, }, },
|
||||
{ "Hello, world!" , { 17535, 11, 1835, 0, }, },
|
||||
{ " Hello, world!" , { 414, 9489, 11, 1835, 0, }, },
|
||||
{ " this is 🦙.cpp" , { 437, 317, 12394, 99, 234, 13, 14789, }, },
|
||||
{ "w048 7tuijk dsdfhu" , { 86, 15, 19, 23, 207, 22, 83, 3963, 27659, 26078, 3934, 14072, }, },
|
||||
{ "нещо на Български" , { 1593, 6478, 616, 2251, 14994, }, },
|
||||
{ "កាន់តែពិសេសអាចខលចេញ" , { 155, 239, 209, 155, 239, 114, 155, 239, 228, 155, 240, 220, 155, 239, 224, 155, 240, 211, 155, 239, 231, 155, 239, 115, 155, 239, 240, 155, 240, 210, 155, 239, 240, 155, 239, 95, 155, 239, 114, 155, 239, 214, 155, 239, 210, 155, 239, 236, 155, 239, 214, 155, 240, 210, 155, 239, 218, }, },
|
||||
{ "🚀 (normal) 😶🌫️ (multiple emojis concatenated) ✅ (only emoji that has its own token)", { 10047, 235, 209, 334, 8760, 8, 12394, 233, 114, 350, 222, 10047, 221, 104, 169, 116, 224, 334, 4684, 3909, 992, 24330, 262, 29651, 612, 8, 207, 156, 237, 214, 334, 5950, 992, 78, 12896, 344, 638, 891, 1372, 10736, 8, }, },
|
||||
{ "Hello" , { 17535, }, },
|
||||
{ " Hello" , { 414, 9489, }, },
|
||||
{ " Hello" , { 207, 414, 9489, }, },
|
||||
{ " Hello" , { 243, 414, 9489, }, },
|
||||
{ " Hello" , { 315, 414, 9489, }, },
|
||||
{ " Hello\n Hello" , { 315, 414, 9489, 185, 315, 414, 9489, }, },
|
||||
{ "\n =" , { 185, 405, }, },
|
||||
{ "' era" , { 6, 2895, }, },
|
||||
{ "Hello, y'all! How are you 😁 ?我想在apple工作1314151天～", { 17535, 11, 320, 6, 435, 0, 1717, 417, 340, 12394, 233, 210, 3015, 19100, 608, 9413, 2668, 16, 18, 16, 19, 16, 20, 16, 1393, 169, 121, 239, }, },
|
||||
|
||||
};
|
||||
|
||||
return _k_tests;
|
||||
}
|
||||
|
||||
int main(int argc, char **argv) {
|
||||
if (argc < 2) {
|
||||
fprintf(stderr, "Usage: %s vocab-file [text-file]\n", argv[0]);
|
||||
return 1;
|
||||
}
|
||||
|
||||
const std::string fname = argv[1];
|
||||
|
||||
std::string fname_text;
|
||||
if (argc > 2) {
|
||||
fname_text = argv[2];
|
||||
}
|
||||
|
||||
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;
|
||||
}
|
||||
}
|
||||
|
||||
if (llama_vocab_type(model) != LLAMA_VOCAB_TYPE_DEEPSEEKCODER) {
|
||||
fprintf(stderr, "%s : error: vocab type is not DEEPSEEKCODER\n", __func__);
|
||||
llama_free_model(model);
|
||||
llama_free(ctx);
|
||||
return 2;
|
||||
}
|
||||
|
||||
#ifdef _WIN32
|
||||
// We need this for unicode console support
|
||||
console::init(false, false);
|
||||
atexit([]() { console::cleanup(); });
|
||||
#endif
|
||||
|
||||
bool success = true;
|
||||
|
||||
for (const auto & test_kv : k_tests()) {
|
||||
const std::vector<llama_token> res = llama_tokenize(ctx, test_kv.first, false);
|
||||
|
||||
printf("\n");
|
||||
printf("src: '%s'\n", test_kv.first.c_str());
|
||||
printf("res: '%s'\n", llama_detokenize_bpe(ctx, res).c_str());
|
||||
printf("tok: ");
|
||||
for (const auto & tok : res) {
|
||||
printf("%d ", tok);
|
||||
}
|
||||
printf("\n");
|
||||
|
||||
bool correct = res.size() == test_kv.second.size();
|
||||
for (int i = 0; i < (int) res.size() && correct; ++i) {
|
||||
if (test_kv.second[i] != res[i]) {
|
||||
correct = false;
|
||||
}
|
||||
}
|
||||
|
||||
if (!correct) {
|
||||
fprintf(stderr, "%s : failed test: '%s'\n", __func__, test_kv.first.c_str());
|
||||
fprintf(stderr, "%s : detokenized to: '%s' instead of '%s'\n", __func__,
|
||||
llama_detokenize_bpe(ctx, res).c_str(),
|
||||
llama_detokenize_bpe(ctx, test_kv.second).c_str());
|
||||
fprintf(stderr, "%s : expected tokens: ", __func__);
|
||||
for (const auto & t : test_kv.second) {
|
||||
fprintf(stderr, "%6d, ", t);
|
||||
}
|
||||
fprintf(stderr, "\n");
|
||||
fprintf(stderr, "%s : got tokens: ", __func__);
|
||||
for (const auto & t : res) {
|
||||
fprintf(stderr, "%6d, ", t);
|
||||
}
|
||||
fprintf(stderr, "\n");
|
||||
|
||||
success = false;
|
||||
}
|
||||
}
|
||||
|
||||
if (!fname_text.empty()) {
|
||||
fprintf(stderr, "%s : tokenizing: '%s'\n", __func__, fname_text.c_str());
|
||||
|
||||
std::string text;
|
||||
{
|
||||
std::ifstream ifs(fname_text);
|
||||
if (!ifs) {
|
||||
fprintf(stderr, "%s : error: could not open file '%s'\n", __func__, fname_text.c_str());
|
||||
return 1;
|
||||
}
|
||||
text = std::string(std::istreambuf_iterator<char>(ifs), std::istreambuf_iterator<char>());
|
||||
}
|
||||
|
||||
fprintf(stderr, "%s : text size: %zu\n", __func__, text.size());
|
||||
|
||||
const std::vector<llama_token> res = llama_tokenize(ctx, text, false);
|
||||
|
||||
fprintf(stderr, "%s : tokens: %zu\n", __func__, res.size());
|
||||
|
||||
{
|
||||
const std::string fname_out = fname_text + ".tokcpp";
|
||||
|
||||
std::ofstream ofs(fname_out);
|
||||
if (!ofs) {
|
||||
fprintf(stderr, "%s : error: could not open file '%s'\n", __func__, fname_out.c_str());
|
||||
return 1;
|
||||
}
|
||||
|
||||
for (const auto & tok : res) {
|
||||
ofs << tok << " '" << llama_detokenize_bpe(ctx, std::vector<int>{tok}) << "'" << std::endl;
|
||||
}
|
||||
}
|
||||
|
||||
fprintf(stderr, "%s : tokens written to '%s'\n", __func__, (fname_text + ".tokcpp").c_str());
|
||||
}
|
||||
|
||||
llama_free_model(model);
|
||||
llama_free(ctx);
|
||||
|
||||
llama_backend_free();
|
||||
|
||||
return success ? 0 : 3;
|
||||
}
|
||||
83
tests/test-tokenizer-0-deepseek-coder.py
Normal file
83
tests/test-tokenizer-0-deepseek-coder.py
Normal file
|
|
@ -0,0 +1,83 @@
|
|||
# tests with BPE tokenizer
|
||||
|
||||
import argparse
|
||||
|
||||
from transformers import AutoTokenizer
|
||||
|
||||
parser = argparse.ArgumentParser()
|
||||
parser.add_argument("dir_tokenizer", help="directory containing 'tokenizer.model' file")
|
||||
parser.add_argument("--fname-tok", help="path to a text file to tokenize")
|
||||
args = parser.parse_args()
|
||||
|
||||
dir_tokenizer = args.dir_tokenizer
|
||||
|
||||
tokenizer = AutoTokenizer.from_pretrained(dir_tokenizer)
|
||||
|
||||
tests = [
|
||||
"",
|
||||
" ",
|
||||
" ",
|
||||
" ",
|
||||
"\t",
|
||||
"\n",
|
||||
"\t\n",
|
||||
"Hello world",
|
||||
" Hello world",
|
||||
"Hello World",
|
||||
" Hello World",
|
||||
" Hello World!",
|
||||
"Hello, world!",
|
||||
" Hello, world!",
|
||||
" this is 🦙.cpp",
|
||||
"w048 7tuijk dsdfhu",
|
||||
"нещо на Български",
|
||||
"កាន់តែពិសេសអាចខលចេញ",
|
||||
"🚀 (normal) 😶🌫️ (multiple emojis concatenated) ✅ (only emoji that has its own token)",
|
||||
"Hello",
|
||||
" Hello",
|
||||
" Hello",
|
||||
" Hello",
|
||||
" Hello",
|
||||
" Hello\n Hello",
|
||||
"\n =",
|
||||
"' era",
|
||||
"Hello, y'all! How are you 😁 ?我想在apple工作1314151天～",
|
||||
]
|
||||
|
||||
for text in tests:
|
||||
print('text: ', text)
|
||||
print(tokenizer.encode(text))
|
||||
print(tokenizer.decode(tokenizer.encode(text)))
|
||||
|
||||
print("\n\ntests for C++:\n")
|
||||
for text in tests:
|
||||
res = tokenizer.encode(text)
|
||||
|
||||
k = text.replace('\n', '\\n')
|
||||
k = k.replace('\t', '\\t')
|
||||
k = '"' + k + '"'
|
||||
print("{ %-24s, { " % k, end='')
|
||||
for x in res:
|
||||
print("%7d," % x, end='')
|
||||
print(" }, },")
|
||||
|
||||
print(tokenizer.encode('hello'))
|
||||
print(tokenizer.encode('world'))
|
||||
print(tokenizer.encode(' world'))
|
||||
print(tokenizer.encode('hello world'))
|
||||
|
||||
fname_tok = args.fname_tok
|
||||
if fname_tok:
|
||||
print('tokenizing file: ', fname_tok)
|
||||
fname_out = fname_tok + '.tok'
|
||||
with open(fname_tok, 'r', encoding='utf-8') as f:
|
||||
lines = f.readlines()
|
||||
s = ''.join(lines)
|
||||
res = tokenizer.encode(s)
|
||||
# write to file
|
||||
with open(fname_out, 'w', encoding='utf-8') as f:
|
||||
for x in res:
|
||||
f.write(str(x) + ' \'' + tokenizer.decode(x) + '\'\n')
|
||||
print('len(res): ', len(res))
|
||||
print('len(lines): ', len(lines))
|
||||
print('results written to: ', fname_out)
|
||||
188
tests/test-tokenizer-0-deepseek-llm.cpp
Normal file
188
tests/test-tokenizer-0-deepseek-llm.cpp
Normal file
|
|
@ -0,0 +1,188 @@
|
|||
#include "llama.h"
|
||||
#include "common.h"
|
||||
#include "console.h"
|
||||
|
||||
#include <cstdio>
|
||||
#include <string>
|
||||
#include <map>
|
||||
#include <vector>
|
||||
#include <fstream>
|
||||
|
||||
// generate using test-tokenizer-0-falcon.py
|
||||
static const std::map<std::string, std::vector<llama_token>> & k_tests() {
|
||||
static std::map<std::string, std::vector<llama_token>> _k_tests = {
|
||||
{ "" , { }, },
|
||||
{ " " , { 207, }, },
|
||||
{ " " , { 243, }, },
|
||||
{ " " , { 300, }, },
|
||||
{ "\t" , { 184, }, },
|
||||
{ "\n" , { 185, }, },
|
||||
{ "\t\n" , { 184, 185, }, },
|
||||
{ "Hello world" , { 17464, 1843, }, },
|
||||
{ " Hello world" , { 37727, 1843, }, },
|
||||
{ "Hello World" , { 17464, 5427, }, },
|
||||
{ " Hello World" , { 37727, 5427, }, },
|
||||
{ " Hello World!" , { 37727, 5427, 0, }, },
|
||||
{ "Hello, world!" , { 17464, 11, 1843, 0, }, },
|
||||
{ " Hello, world!" , { 37727, 11, 1843, 0, }, },
|
||||
{ " this is 🦙.cpp" , { 437, 317, 12356, 99, 234, 13, 14743, }, },
|
||||
{ "w048 7tuijk dsdfhu" , { 86, 15, 19, 23, 207, 22, 83, 3970, 27519, 26016, 3944, 14025, }, },
|
||||
{ "нещо на Български" , { 1603, 6476, 620, 91754, }, },
|
||||
{ "កាន់តែពិសេសអាចខលចេញ" , { 71374, 209, 71374, 114, 71374, 228, 155, 240, 220, 71374, 224, 155, 240, 211, 71374, 231, 71374, 115, 71374, 240, 155, 240, 210, 71374, 240, 71374, 95, 71374, 114, 71374, 214, 71374, 210, 71374, 236, 71374, 214, 155, 240, 210, 71374, 218, }, },
|
||||
{ "🚀 (normal) 😶🌫️ (multiple emojis concatenated) ✅ (only emoji that has its own token)", { 10044, 95300, 334, 8754, 8, 33701, 114, 350, 222, 10044, 221, 104, 46713, 334, 34732, 996, 24250, 262, 80923, 8, 207, 37103, 214, 334, 5956, 89213, 344, 643, 895, 1377, 10728, 8, }, },
|
||||
{ "Hello" , { 17464, }, },
|
||||
{ " Hello" , { 37727, }, },
|
||||
{ " Hello" , { 207, 37727, }, },
|
||||
{ " Hello" , { 243, 37727, }, },
|
||||
{ " Hello" , { 300, 37727, }, },
|
||||
{ " Hello\n Hello" , { 300, 37727, 185, 300, 37727, }, },
|
||||
{ "\n =" , { 185, 403, }, },
|
||||
{ "' era" , { 6, 2906, }, },
|
||||
{ "Hello, y'all! How are you 😁 ?我想在apple工作1314151天～", { 17464, 11, 320, 6, 436, 0, 1724, 418, 340, 33701, 210, 3025, 19017, 612, 9407, 2681, 16, 18, 16, 19, 16, 20, 16, 1398, 68940, 239, }, },
|
||||
|
||||
};
|
||||
|
||||
return _k_tests;
|
||||
}
|
||||
|
||||
int main(int argc, char **argv) {
|
||||
if (argc < 2) {
|
||||
fprintf(stderr, "Usage: %s vocab-file [text-file]\n", argv[0]);
|
||||
return 1;
|
||||
}
|
||||
|
||||
const std::string fname = argv[1];
|
||||
|
||||
std::string fname_text;
|
||||
if (argc > 2) {
|
||||
fname_text = argv[2];
|
||||
}
|
||||
|
||||
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;
|
||||
}
|
||||
}
|
||||
|
||||
if (llama_vocab_type(model) != LLAMA_VOCAB_TYPE_DEEPSEEKLLM) {
|
||||
fprintf(stderr, "%s : error: vocab type is not DEEPSEEKLLM\n", __func__);
|
||||
llama_free_model(model);
|
||||
llama_free(ctx);
|
||||
return 2;
|
||||
}
|
||||
|
||||
#ifdef _WIN32
|
||||
// We need this for unicode console support
|
||||
console::init(false, false);
|
||||
atexit([]() { console::cleanup(); });
|
||||
#endif
|
||||
|
||||
bool success = true;
|
||||
|
||||
for (const auto & test_kv : k_tests()) {
|
||||
const std::vector<llama_token> res = llama_tokenize(ctx, test_kv.first, false);
|
||||
|
||||
printf("\n");
|
||||
printf("src: '%s'\n", test_kv.first.c_str());
|
||||
printf("res: '%s'\n", llama_detokenize_bpe(ctx, res).c_str());
|
||||
printf("tok: ");
|
||||
for (const auto & tok : res) {
|
||||
printf("%d ", tok);
|
||||
}
|
||||
printf("\n");
|
||||
|
||||
bool correct = res.size() == test_kv.second.size();
|
||||
for (int i = 0; i < (int) res.size() && correct; ++i) {
|
||||
if (test_kv.second[i] != res[i]) {
|
||||
correct = false;
|
||||
}
|
||||
}
|
||||
|
||||
if (!correct) {
|
||||
fprintf(stderr, "%s : failed test: '%s'\n", __func__, test_kv.first.c_str());
|
||||
fprintf(stderr, "%s : detokenized to: '%s' instead of '%s'\n", __func__,
|
||||
llama_detokenize_bpe(ctx, res).c_str(),
|
||||
llama_detokenize_bpe(ctx, test_kv.second).c_str());
|
||||
fprintf(stderr, "%s : expected tokens: ", __func__);
|
||||
for (const auto & t : test_kv.second) {
|
||||
fprintf(stderr, "%6d, ", t);
|
||||
}
|
||||
fprintf(stderr, "\n");
|
||||
fprintf(stderr, "%s : got tokens: ", __func__);
|
||||
for (const auto & t : res) {
|
||||
fprintf(stderr, "%6d, ", t);
|
||||
}
|
||||
fprintf(stderr, "\n");
|
||||
|
||||
success = false;
|
||||
}
|
||||
}
|
||||
|
||||
if (!fname_text.empty()) {
|
||||
fprintf(stderr, "%s : tokenizing: '%s'\n", __func__, fname_text.c_str());
|
||||
|
||||
std::string text;
|
||||
{
|
||||
std::ifstream ifs(fname_text);
|
||||
if (!ifs) {
|
||||
fprintf(stderr, "%s : error: could not open file '%s'\n", __func__, fname_text.c_str());
|
||||
return 1;
|
||||
}
|
||||
text = std::string(std::istreambuf_iterator<char>(ifs), std::istreambuf_iterator<char>());
|
||||
}
|
||||
|
||||
fprintf(stderr, "%s : text size: %zu\n", __func__, text.size());
|
||||
|
||||
const std::vector<llama_token> res = llama_tokenize(ctx, text, false);
|
||||
|
||||
fprintf(stderr, "%s : tokens: %zu\n", __func__, res.size());
|
||||
|
||||
{
|
||||
const std::string fname_out = fname_text + ".tokcpp";
|
||||
|
||||
std::ofstream ofs(fname_out);
|
||||
if (!ofs) {
|
||||
fprintf(stderr, "%s : error: could not open file '%s'\n", __func__, fname_out.c_str());
|
||||
return 1;
|
||||
}
|
||||
|
||||
for (const auto & tok : res) {
|
||||
ofs << tok << " '" << llama_detokenize_bpe(ctx, std::vector<int>{tok}) << "'" << std::endl;
|
||||
}
|
||||
}
|
||||
|
||||
fprintf(stderr, "%s : tokens written to '%s'\n", __func__, (fname_text + ".tokcpp").c_str());
|
||||
}
|
||||
|
||||
llama_free_model(model);
|
||||
llama_free(ctx);
|
||||
|
||||
llama_backend_free();
|
||||
|
||||
return success ? 0 : 3;
|
||||
}
|
||||
83
tests/test-tokenizer-0-deepseek-llm.py
Normal file
83
tests/test-tokenizer-0-deepseek-llm.py
Normal file
|
|
@ -0,0 +1,83 @@
|
|||
# tests with BPE tokenizer
|
||||
|
||||
import argparse
|
||||
|
||||
from transformers import AutoTokenizer
|
||||
|
||||
parser = argparse.ArgumentParser()
|
||||
parser.add_argument("dir_tokenizer", help="directory containing 'tokenizer.model' file")
|
||||
parser.add_argument("--fname-tok", help="path to a text file to tokenize")
|
||||
args = parser.parse_args()
|
||||
|
||||
dir_tokenizer = args.dir_tokenizer
|
||||
|
||||
tokenizer = AutoTokenizer.from_pretrained(dir_tokenizer)
|
||||
|
||||
tests = [
|
||||
"",
|
||||
" ",
|
||||
" ",
|
||||
" ",
|
||||
"\t",
|
||||
"\n",
|
||||
"\t\n",
|
||||
"Hello world",
|
||||
" Hello world",
|
||||
"Hello World",
|
||||
" Hello World",
|
||||
" Hello World!",
|
||||
"Hello, world!",
|
||||
" Hello, world!",
|
||||
" this is 🦙.cpp",
|
||||
"w048 7tuijk dsdfhu",
|
||||
"нещо на Български",
|
||||
"កាន់តែពិសេសអាចខលចេញ",
|
||||
"🚀 (normal) 😶🌫️ (multiple emojis concatenated) ✅ (only emoji that has its own token)",
|
||||
"Hello",
|
||||
" Hello",
|
||||
" Hello",
|
||||
" Hello",
|
||||
" Hello",
|
||||
" Hello\n Hello",
|
||||
"\n =",
|
||||
"' era",
|
||||
"Hello, y'all! How are you 😁 ?我想在apple工作1314151天～",
|
||||
]
|
||||
|
||||
for text in tests:
|
||||
print('text: ', text)
|
||||
print(tokenizer.encode(text))
|
||||
print(tokenizer.decode(tokenizer.encode(text)))
|
||||
|
||||
print("\n\ntests for C++:\n")
|
||||
for text in tests:
|
||||
res = tokenizer.encode(text)
|
||||
|
||||
k = text.replace('\n', '\\n')
|
||||
k = k.replace('\t', '\\t')
|
||||
k = '"' + k + '"'
|
||||
print("{ %-24s, { " % k, end='')
|
||||
for x in res:
|
||||
print("%7d," % x, end='')
|
||||
print(" }, },")
|
||||
|
||||
print(tokenizer.encode('hello'))
|
||||
print(tokenizer.encode('world'))
|
||||
print(tokenizer.encode(' world'))
|
||||
print(tokenizer.encode('hello world'))
|
||||
|
||||
fname_tok = args.fname_tok
|
||||
if fname_tok:
|
||||
print('tokenizing file: ', fname_tok)
|
||||
fname_out = fname_tok + '.tok'
|
||||
with open(fname_tok, 'r', encoding='utf-8') as f:
|
||||
lines = f.readlines()
|
||||
s = ''.join(lines)
|
||||
res = tokenizer.encode(s)
|
||||
# write to file
|
||||
with open(fname_out, 'w', encoding='utf-8') as f:
|
||||
for x in res:
|
||||
f.write(str(x) + ' \'' + tokenizer.decode(x) + '\'\n')
|
||||
print('len(res): ', len(res))
|
||||
print('len(lines): ', len(lines))
|
||||
print('results written to: ', fname_out)
|
||||
|
|
@ -38,6 +38,7 @@ static const std::map<std::string, std::vector<llama_token>> & k_tests() {
|
|||
{ " Hello\n Hello" , { 466, 23090, 742, 23090, }, },
|
||||
{ "\n =" , { 1212, 40, }, },
|
||||
{ "' era" , { 18, 4932, }, },
|
||||
{ "Hello, y'all! How are you 😁 ?我想在apple工作1314151天～", { 9856, 23, 291, 18, 436, 12, 1265, 362, 299, 8196, 207, 204, 42, 50087, 123, 2727, 20300, 32022, 133, 234, 17419, 30137, 28, 7858, 181, 133, 236, }, },
|
||||
};
|
||||
|
||||
return _k_tests;
|
||||
|
|
@ -115,7 +116,6 @@ int main(int argc, char **argv) {
|
|||
printf("\n");
|
||||
|
||||
bool correct = res.size() == test_kv.second.size();
|
||||
|
||||
for (int i = 0; i < (int) res.size() && correct; ++i) {
|
||||
if (test_kv.second[i] != res[i]) {
|
||||
correct = false;
|
||||
|
|
|
|||
|
|
@ -41,6 +41,7 @@ tests = [
|
|||
" Hello\n Hello",
|
||||
"\n =",
|
||||
"' era",
|
||||
"Hello, y'all! How are you 😁 ?我想在apple工作1314151天～",
|
||||
]
|
||||
|
||||
for text in tests:
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue