-
-
Save kayush2O6/c9b81a25062ddb1616e6f6d8ff767093 to your computer and use it in GitHub Desktop.
BEFORE INSTALLING THE CUML, PLEASE MAKE SURE YOU HAVE FOLLOWED THE ABOVE STEPS FOR CUDF. CUDF SHOULD BE WORKING... | |
Step 1: Install the cuml and its depandencies. | |
!apt install libopenblas-base libomp-dev | |
!pip install cuml-cuda100 | |
# import cuml at this point, will give libcuml.so not found error. # | |
NOTE: Step2 is optional and is just for information, you can fast forward to Step3 directly to work quickely. | |
Step 2: find the location of libcuml.so | |
!find / -name libcuml.so* | |
# You should get something like this: /usr/local/lib/python3.6/dist-packages/libcuml.so # | |
**IMPORTANT**: At this point, if you are using terminal, then you update the $LD_LIBRARY_PATH environment variable | |
with the location of libcuml.so, and it will work. | |
checkout out my git comment regarding this: https://github.com/rapidsai/cuml/issues/404#issuecomment-487054041 | |
But, jupyter notebook load this $LDLIBRARY_PATH at the start time only, so even if you update the LD_LIBRARY_PATH in notebook | |
cell, it will not be loaded by notebook. | |
Step 3: In order to solve this issue in colab, run the below command. | |
!cp /usr/local/lib/python3.6/dist-packages/libcuml.so /usr/lib64-nvidia/ | |
Now, you can use import cuml and run you code. |
on running " !pip install cuml-cuda100" command in colab gives below error:
Collecting cuml-cuda100 Using cached https://files.pythonhosted.org/packages/87/9e/4329e8a39672ed011b372981ba337ca0a0d923c7cbd9e96725f757094167/cuml-cuda100-0.6.1.post1.tar.gz Building wheels for collected packages: cuml-cuda100 Building wheel for cuml-cuda100 (setup.py) ... error ERROR: Failed building wheel for cuml-cuda100 Running setup.py clean for cuml-cuda100 Failed to build cuml-cuda100 Installing collected packages: cuml-cuda100 Running setup.py install for cuml-cuda100 ... error ERROR: Command errored out with exit status 1: /usr/bin/python3 -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-2eewpayy/cuml-cuda100/setup.py'"'"'; __file__='"'"'/tmp/pip-install-2eewpayy/cuml-cuda100/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' install --record /tmp/pip-record-d50wb7pa/install-record.txt --single-version-externally-managed --compile Check the logs for full command output.
Not sure what am i missing here.
I got the same error
Hi @tonewhoo @TaherLkdw, I am afraid these instructions are obsolete now.
Please refer these new instructions mentioned in official page of rapids.ai colab
on running " !pip install cuml-cuda100" command in colab gives below error:
Collecting cuml-cuda100 Using cached https://files.pythonhosted.org/packages/87/9e/4329e8a39672ed011b372981ba337ca0a0d923c7cbd9e96725f757094167/cuml-cuda100-0.6.1.post1.tar.gz Building wheels for collected packages: cuml-cuda100 Building wheel for cuml-cuda100 (setup.py) ... error ERROR: Failed building wheel for cuml-cuda100 Running setup.py clean for cuml-cuda100 Failed to build cuml-cuda100 Installing collected packages: cuml-cuda100 Running setup.py install for cuml-cuda100 ... error ERROR: Command errored out with exit status 1: /usr/bin/python3 -u -c 'import sys, setuptools, tokenize; sys.argv[0] = '"'"'/tmp/pip-install-2eewpayy/cuml-cuda100/setup.py'"'"'; __file__='"'"'/tmp/pip-install-2eewpayy/cuml-cuda100/setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(__file__);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, __file__, '"'"'exec'"'"'))' install --record /tmp/pip-record-d50wb7pa/install-record.txt --single-version-externally-managed --compile Check the logs for full command output.
Not sure what am i missing here.