In some cases, a minor-version mismatch will not cause later errors. Pytorch binaries were compiled with Cuda 11.1. RuntimeError: Cuda extensions are being compiled with a version of Cuda that does not match the version used to compile Pytorch binaries. Raise RuntimeError("Cuda extensions are being compiled with a version of Cuda that does " +
#My nvidia has no clone option install#
Running command /scratch3/venv/proxy/bin/python3.8 -u -c 'import io, os, sys, setuptools, tokenize sys.argv = '"'"'/scratch/tmp/pip-req-build-fg_khhkt/setup.py'"'"' _file_='"'"'/scratch/tmp/pip-req-build-fg_khhkt/setup.py'"'"' f = getattr(tokenize, '"'"'open'"'"', open)(_file_) if os.path.exists(_file_) else io.StringIO('"'"'from setuptools import setup setup()'"'"') code = f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"') f.close() exec(compile(code, _file_, '"'"'exec'"'"'))' -cpp_ext -cuda_ext install -record /scratch/tmp/pip-record-u812zb2v/install-record.txt -single-version-externally-managed -compile -install-headers /scratch3/venv/proxy/include/site/python3.8/apexĬopyright (c) 2005-2018 NVIDIA CorporationĬuda compilation tools, release 10.0, V10.0.130įile "/scratch/tmp/pip-req-build-fg_khhkt/setup.py", line 159, in Ĭheck_cuda_torch_binary_vs_bare_metal(CUDA_HOME)įile "/scratch/tmp/pip-req-build-fg_khhkt/setup.py", line 99, in check_cuda_torch_binary_vs_bare_metal Skipping wheel build for apex, due to binaries being disabled for it. Warnings.warn("Option -pyprof not specified. scratch/tmp/pip-req-build-fg_khhkt/setup.py:67: UserWarning: Option -pyprof not specified. Reading manifest file '/scratch/tmp/pip-pip-egg-info-yc32vm37/apex.egg-info/SOURCES.txt' Writing manifest file '/scratch/tmp/pip-pip-egg-info-yc32vm37/apex.egg-info/SOURCES.txt' Writing top-level names to /scratch/tmp/pip-pip-egg-info-yc32vm37/apex.egg-info/top_level.txt Writing dependency_links to /scratch/tmp/pip-pip-egg-info-yc32vm37/apex.egg-info/dependency_links.txt Writing /scratch/tmp/pip-pip-egg-info-yc32vm37/apex.egg-info/PKG-INFO Ĭreating /scratch/tmp/pip-pip-egg-info-yc32vm37/apex.egg-info You can find discussion regarding this at. Pip 21.3 will remove support for this functionality. We recommend you use -use-feature=in-tree-build to test your packages with this new behavior before it becomes the default. Processing /scratch3/research/code/fashion/proxynca_pp/apexĭEPRECATION: A future pip version will change local packages to be built in-place without first copying to a temporary directory. Using pip 21.2.4 from /scratch3/venv/proxy/lib/python3.8/site-packages/pip (python 3.8) scratch3/venv/proxy/lib/python3.8/site-packages/pip/_internal/commands/install.py:229: UserWarning: Disabling all use of wheels due to the use of -build-option / -global-option / -install-option.Ĭmdoptions.check_install_build_global(options)
(proxy) apex]$ pip install -v -disable-pip-version-check -no-cache-dir \ Successfully uninstalled apex-0.9.10.dev0 scratch3/venv/proxy/lib/python3.8/site-packages/apex/* ImportError: cannot import name 'UnencryptedCookieSessionFactor圜onfig' from 'ssion' (unknown location)Īfter I got the above error, I tried this answer: (proxy) proxynca_pp]$ pip uninstall apexįound existing installation: apex 0.9.10.dev0 e('agg', warn=False, force=True)įile "/scratch3/venv/proxy/lib/python3.8/site-packages/apex/_init_.py", line 13, in įrom ssion import UnencryptedCookieSessionFactor圜onfig If any parameter follows 'warn', they should be pass as keyword, not positionally. Train.py:12: MatplotlibDeprecationWarning: The 'warn' parameter of use() is deprecated since Matplotlib 3.1 and will be removed in 3.3.
#My nvidia has no clone option how to#
I am unsure how to fix the final error: (proxy) proxynca_pp]$ CUDA_VISIBLE_DEVICES=0,1 python train.py -dataset cub -config config/cub.json -mode train -apex -seed 0 I am running a code that apparently requires NVIDIA apex (I initially didn't know and installed the wrong apex).
0 kommentar(er)
