getnamo
commented
6 years ago Just tested a download of https://github.com/getnamo/tensorflow-ue4-examples/releases/tag/0.2.3 and https://github.com/getnamo/tensorflow-ue4/releases/tag/0.6.0 (nb: you need the files inside the .7z file not source).
I extracted the 0.2.3 project, dragged in plugins folder from the 0.6.0 archive and then opened the project and waited ~ 4min until the output log showed:
Collecting tensorflow==1.5.0
Using cached tensorflow-1.5.0-cp36-cp36m-win_amd64.whl
Collecting tensorflow-tensorboard<1.6.0,>=1.5.0 (from tensorflow==1.5.0)
Downloading tensorflow_tensorboard-1.5.1-py3-none-any.whl (3.0MB)
Collecting absl-py>=0.1.6 (from tensorflow==1.5.0)
Downloading absl-py-0.1.10.tar.gz (79kB)
Collecting numpy>=1.12.1 (from tensorflow==1.5.0)
Downloading numpy-1.14.1-cp36-none-win_amd64.whl (13.4MB)
Collecting six>=1.10.0 (from tensorflow==1.5.0)
Using cached six-1.11.0-py2.py3-none-any.whl
Requirement already satisfied: wheel>=0.26 in c:\users\admin\desktop\tensorflow-ue4-examples\plugins\unrealenginepython\binaries\win64\lib\site-packages (from tensorflow==1.5.0)
Collecting protobuf>=3.4.0 (from tensorflow==1.5.0)
Using cached protobuf-3.5.1-py2.py3-none-any.whl
Collecting bleach==1.5.0 (from tensorflow-tensorboard<1.6.0,>=1.5.0->tensorflow==1.5.0)
Using cached bleach-1.5.0-py2.py3-none-any.whl
Collecting html5lib==0.9999999 (from tensorflow-tensorboard<1.6.0,>=1.5.0->tensorflow==1.5.0)
Collecting markdown>=2.6.8 (from tensorflow-tensorboard<1.6.0,>=1.5.0->tensorflow==1.5.0)
Using cached Markdown-2.6.11-py2.py3-none-any.whl
Collecting werkzeug>=0.11.10 (from tensorflow-tensorboard<1.6.0,>=1.5.0->tensorflow==1.5.0)
Using cached Werkzeug-0.14.1-py2.py3-none-any.whl
Requirement already satisfied: setuptools in c:\users\admin\desktop\tensorflow-ue4-examples\plugins\unrealenginepython\binaries\win64\lib\site-packages (from protobuf>=3.4.0->tensorflow==1.5.0)
Building wheels for collected packages: absl-py
Running setup.py bdist_wheel for absl-py: started
Running setup.py bdist_wheel for absl-py: finished with status 'done'
Stored in directory: C:\Users\Admin\AppData\Local\pip\Cache\wheels\45\07\0e\6880381ca521796cf6cc18ba4ab502c2232e5777099b4df4ae
Successfully built absl-py
Installing collected packages: six, html5lib, bleach, numpy, markdown, werkzeug, protobuf, tensorflow-tensorboard, absl-py, tensorflow
Successfully installed absl-py-0.1.10 bleach-1.5.0 html5lib-0.9999999 markdown-2.6.11 numpy-1.14.1 protobuf-3.5.1 six-1.11.0 tensorflow-1.5.0 tensorflow-tensorboard-1.5.1 werkzeug-0.14.1then closed the project and re-opened it as per https://github.com/getnamo/tensorflow-ue4#no-module-named-tensorflow problem.
After the project loaded the second time, hitting play started training kerasCNN. Having used the cpu version of tensorflow it started training right away, albeit slowly. Stopped play and changed the Python TFModule to mnistSpawnSamples on the ConnectedTFMnistActor and hit play again, this time the much simpler regression model was trained in <3 seconds while streaming images. Classification tests also worked as expected.
From a basic test, it appears to be working. Where in this process are you hitting a snag? Kindly provide more information/step by step instructions so the issue can be reproduced as you experience it.
nonlin
I've downloaded the example project and copied into it the latest release of tensflow for UE4. I've made sure to enable all three plugins and I have no errors at all.
However no training is being done. None of the example projects work.