Skip to content

Add more tokenizer tests #3742

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 7 commits into from
Oct 24, 2023
Merged

Add more tokenizer tests #3742

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
6919b8c
Add more tokenizer tests
Galunid Oct 23, 2023
769135b
Add starcoder
Galunid Oct 23, 2023
226f0a8
Update test vocab files
Galunid Oct 23, 2023
a06e82b
Restrict bpe tokenizer tests to unicode planes
Galunid Oct 23, 2023
1244b00
Update comment
Galunid Oct 23, 2023
38de5d0
Comment cosmetics
Galunid Oct 23, 2023
a01ddfb
Remove bloom vocab/test
Galunid Oct 23, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Binary file added models/ggml-vocab-baichuan.gguf
View file
Open in desktop
Binary file not shown.
Binary file added models/ggml-vocab-gpt-neox.gguf
View file
Open in desktop
Binary file not shown.
Binary file added models/ggml-vocab-refact.gguf
View file
Open in desktop
Binary file not shown.
Binary file added models/ggml-vocab-starcoder.gguf
View file
Open in desktop
Binary file not shown.
4 changes: 4 additions & 0 deletions tests/CMakeLists.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,10 +28,14 @@ llama_build_executable(test-tokenizer-0-falcon.cpp)
llama_test_executable (test-tokenizer-0-falcon test-tokenizer-0-falcon.cpp ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-falcon.gguf)
llama_build_executable(test-tokenizer-1-llama.cpp)
llama_test_executable (test-tokenizer-1-llama test-tokenizer-1-llama.cpp ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-llama.gguf)
llama_test_executable(test-tokenizer-1-baichuan test-tokenizer-1-llama.cpp ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-baichuan.gguf)
llama_build_executable(test-tokenizer-1-bpe.cpp)
llama_test_executable (test-tokenizer-1-falcon test-tokenizer-1-bpe.cpp ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-falcon.gguf)
llama_test_executable(test-tokenizer-1-aquila test-tokenizer-1-bpe.cpp ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-aquila.gguf)
llama_test_executable(test-tokenizer-1-mpt test-tokenizer-1-bpe.cpp ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-mpt.gguf)
llama_test_executable(test-tokenizer-1-gpt-neox test-tokenizer-1-bpe.cpp ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-gpt-neox.gguf)
llama_test_executable(test-tokenizer-1-refact test-tokenizer-1-bpe.cpp ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-refact.gguf)
llama_test_executable(test-tokenizer-1-starcoder test-tokenizer-1-bpe.cpp ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab-starcoder.gguf)
llama_build_and_test_executable(test-grammar-parser.cpp)
llama_build_and_test_executable(test-llama-grammar.cpp)
llama_build_and_test_executable(test-grad0.cpp) # SLOW
Expand Down
15 changes: 12 additions & 3 deletions tests/test-tokenizer-1-bpe.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -91,9 +91,19 @@ int main(int argc, char **argv) {
}
}
}
// TODO: why doesn't this work for the full range of Unicodes?
// Restrict to assigned unicode planes
// for (uint32_t cp = 0x10000; cp < 0x0010ffff; ++cp) {
for (uint32_t cp = 0x10000; cp < 0x00080000; ++cp) {
for (uint32_t cp = 0x10000; cp < 0x00040000; ++cp) {
std::string str = codepoint_to_utf8(cp);
std::vector<llama_token> tokens = llama_tokenize(ctx, str, false);
std::string check = llama_detokenize_bpe(ctx, tokens);
if (str != check) {
fprintf(stderr, "%s : error: codepoint %x detokenizes to '%s'(%zu) instead of '%s'(%zu)\n",
__func__, cp, check.c_str(), check.length(), str.c_str(), str.length());
return 4;
}
}
for (uint32_t cp = 0x000e0000; 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_bpe(ctx, tokens);
Expand All @@ -103,7 +113,6 @@ int main(int argc, char **argv) {
return 4;
}
}

llama_free_model(model);
llama_free(ctx);

Expand Down