portlint: no error; poudriere testport pass: 11.0R-amd64-py27, 11.0R-amd64-py3, 11.0R-i386-py27, 10.3R-amd64, 12current-amd64-py27
Note : this port depend on PR https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=219840 for build,and PR https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=220206 for runtime depend
There are tensorflow poudriere testport logs(11amd64,11i386,11amd64-py3,10.3amd64,12current):
[[ https://github.com/amutu/data/blob/master/poudriere_log/103amd64-default.log | 103amd64-default.log ]]
[[ https://github.com/amutu/data/blob/master/poudriere_log/11amd64-default-py3.log | 11amd64-default-py3.log ]]
[[ https://github.com/amutu/data/blob/master/poudriere_log/11amd64-default.log | 11amd64-default.log ]]
[[ https://github.com/amutu/data/blob/master/poudriere_log/11i386-default.log | 11i386-default.log ]]
[[ https://github.com/amutu/data/blob/master/poudriere_log/12current-20170619-default.log | 12current-20170619-default.log ]]
tensorflow functional test inputs steps：
from https://github.com/amutu/tensorflow-port/tree/master/test，just download the *.py files, after install tensorflow,you can test it like this:
python2.7 ./mnist_deep.py,output will like this:
some warning...
can't determine number of CPU cores: assuming 4
can't determine number of CPU cores: assuming 4
step 0, training accuracy 0.1
step 0, training accuracy 0.1
step 100, training accuracy 0.82
step 200, training accuracy 0.94
...
python2.7 ./mnist_softmax.py,output will like this:
some warning...
can't determine number of CPU cores: assuming 4
can't determine number of CPU cores: assuming 4
0.9188
python2.7 ./linear_regression.py,output will like this:
some warning...
train loss: {'loss': 3.355257e-08, 'global_step': 1000}
eval loss: {'loss': 0.0025368345, 'global_step': 1000}