Check the preview of 2nd version of this platform being developed by the open MLCommons taskforce on automation and reproducibility as a free, open-source and technology-agnostic on-prem platform.
package:lib-tensorflow-2.0.1-src-cuda (v3.0.0)
Copyright: See copyright in the source repository
License: See license in the source repository
Creation date: 2020-04-02
Source: GitHub
cID: 1dc07ee0f4742028:156ad44f65b95c83

Don't hesitate to get in touch if you encounter any issues or would like to discuss this community project!
Please report if this CK component works: 1  or fails: 0 
Sign up to be notified when artifacts are shared or updated!

Description  

This meta package is our attempt to provide a unified Python API, CLI and JSON meta description for different package managers and building tools to automatically download and install different components (models, data sets, libraries, frameworks, tools) necessary to run portable program pipelines across evolving platforms. Our on-going project is to make the onboarding process as simple as possible via this platform. Please check this CK white paper and don't hesitate to contact us if you have suggestions or feedback!

Dependencies    

ReadMe  

Install dependencies on Ubuntu

https://www.tensorflow.org/install/install_sources

Install Python 3 (recommended) or Python 2

$ sudo apt-get install libcupti-dev
$ sudo apt-get install python-dev python-pip python-wheel

Install Python packages (in user-space)

python -m pip install --user absl-py==0.8.0              # >=0.7.0
python -m pip install --user astor==0.7.1                # >=0.6.0
python -m pip install --user cachetools==4.0.0           # <5.0,>=2.0.0
python -m pip install --user gast==0.2.2                 # exact
python -m pip install --user google-auth==1.13.1         # <2,>=1.6.3
python -m pip install --user google-auth-oauthlib==0.4.1 # <0.5,>=0.4.1
python -m pip install --user google-pasta==0.2.0         # >=0.1.6
python -m pip install --user grpcio==1.27.2              # >=1.8.6
python -m pip install --user h5py==2.10.0                # exact?
python -m pip install --user keras-applications==1.0.8   # >=1.0.8
python -m pip install --user keras-preprocessing==1.1.0  # >=1.0.5
python -m pip install --user markdown==3.2.1             # >=2.6.8
python -m pip install --user numpy==1.17.2               # <2.0,>=1.16.0
python -m pip install --user oauthlib==3.1.0             # ?
python -m pip install --user opt-einsum==3.2.0           # >=2.3.2
python -m pip install --user protobuf==3.6.1             # >=3.6.1
python -m pip install --user pyasn1==0.4.8               # <0.5.0,>=0.4.6
python -m pip install --user pyasn1-modules==0.2.8       # >=0.2.1
python -m pip install --user requests==2.23.0            # <3,>=2.21.0
python -m pip install --user requests-oauthlib==1.3.0    # >=0.7.0
python -m pip install --user rsa==4.0                    # <4.1,>=3.1.4
python -m pip install --user setuptools==46.1.3          # >=41.0.0
python -m pip install --user tensorboard==2.0.2          # <2.1.0,>=2.0.0 (almost exact)
python -m pip install --user tensorflow-estimator==2.0.1 # <2.1.0,>=2.0.0 (almost exact)
python -m pip install --user termcolor==1.1.0            # >=1.1.0
python -m pip install --user wheel==0.30.0               # >=0.26
python -m pip install --user werkzeug==1.0.1             # >=0.11.15
python -m pip install --user wrapt==1.11.2               # >=1.11.1

Prevent running out of memory

To prevent running out of memory during a build, restrict the build to use e.g. 1 processor:

$ ck install package:lib-tensorflow-2.0.1-src-cuda --env.CK_HOST_CPU_NUMBER_OF_PROCESSORS=1

Known problems

CUDA 10.2 in not supported

Downgrade to 10.1.

Versions  

Files  

Comments  

Please log in to add your comments!
If you notice any inapropriate content that should not be here, please report us as soon as possible and we will try to remove it within 48 hours!