search

BlinkDL / RWKV-LM

RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference, saves VRAM, fast training, "infinite" ctx_len, and free sentence embedding.
Apache License 2.0
12.32k stars 838 forks source link

Ninja compile error how can i fix it? #97

Closed karingijs2727 closed 1 year ago

karingijs2727 commented 1 year ago

Hello i have this error below and i don't know why and how to fix when i try to run train file to train the model this error comes im using VSC and local env

Using C:\Users\hayda\AppData\Local\torch_extensions\torch_extensions\Cache\py311_cu117 as PyTorch extensions root... d:\VSC_PROJECTS\RWKV-3.venv\Lib\site-packages\torch\utils\cpp_extension.py:359: UserWarning: Error checking compiler version for cl: [WinError 2] Het systeem kan het opgegeven bestand niet vinden warnings.warn(f'Error checking compiler version for {compiler}: {error}') Detected CUDA files, patching ldflags Emitting ninja build file C:\Users\hayda\AppData\Local\torch_extensions\torch_extensions\Cache\py311_cu117\wkv\build.ninja... INFO: Could not find files for the given pattern(s). Traceback (most recent call last): File "d:\VSC_PROJECTS\RWKV-3.venv\train.py", line 194, in
from src.trainer import Trainer, TrainerConfig File "d:\VSC_PROJECTS\RWKV-3.venv\src\trainer.py", line 51, in from src.model import GPT, GPTConfig File "d:\VSC_PROJECTS\RWKV-3.venv\src\model.py", line 44, in wkv_cuda = load(name="wkv", sources=["cuda/wkv_op.cpp", "cuda/wkv_cuda.cu"], ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "d:\VSC_PROJECTS\RWKV-3.venv\Lib\site-packages\torch\utils\cpp_extension.py", line 1284, in load return _jit_compile( ^^^^^^^^^^^^^ File "d:\VSC_PROJECTS\RWKV-3.venv\Lib\site-packages\torch\utils\cpp_extension.py", line 1509, in _jit_compile _write_ninja_file_and_build_library( File "d:\VSC_PROJECTS\RWKV-3.venv\Lib\site-packages\torch\utils\cpp_extension.py", line 1611, in _write_ninja_file_and_build_library _write_ninja_file_to_build_library( File "d:\VSC_PROJECTS\RWKV-3.venv\Lib\site-packages\torch\utils\cpp_extension.py", line 2048, in _write_ninja_file_to_build_library _write_ninja_file( File "d:\VSC_PROJECTS\RWKV-3.venv\Lib\site-packages\torch\utils\cpp_extension.py", line 2188, in _write_ninja_file cl_paths = subprocess.check_output(['where', ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\Python.3.11.3\Lib\subprocess.py", line 466, in check_output return run(*popenargs, stdout=PIPE, timeout=timeout, check=True, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\Python.3.11.3\Lib\subprocess.py", line 571, in run raise CalledProcessError(retcode, process.args, subprocess.CalledProcessError: Command '['where', 'cl']' returned non-zero exit status 1.

BlinkDL commented 1 year ago

How to build in Linux: export PATH=/usr/local/cuda/bin:$PATH export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH Run train.py

How to build in win: Install VS2022 build tools (https://aka.ms/vs/17/release/vs_BuildTools.exe select Desktop C++). Reinstall CUDA 11.7 (install VC++ extensions). Run train.py in "x64 native tools command prompt" (in your start menu)