fix inference.py for TF 1.0 && reformat using pep8 code style

[corenel]

• fix inference.py for TF 1.0
• reformat using pep8 code style

[corenel]

Need more fix.

[corenel]

That's strange that I test the script in .travis.yml on local computer successfully, but Travis.org returns a failure.

UPDATE: It seems that some commands run out of memory on travis-ci server. Don't know how to fix it.

[DrSleep]

Thank you, will look at it

[zhixy]

Hi, I suffered a Segmentation fault (core dumped) error, even though I finally got the segmentation result.
python inference.py ../elephant.jpg ./deeplab_resnet.ckpt W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE3 instructions, but these are available on your machine and could speed up CPU computations. W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.1 instructions, but these are available on your machine and could speed up CPU computations. W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use SSE4.2 instructions, but these are available on your machine and could speed up CPU computations. W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX instructions, but these are available on your machine and could speed up CPU computations. W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use AVX2 instructions, but these are available on your machine and could speed up CPU computations. W tensorflow/core/platform/cpu_feature_guard.cc:45] The TensorFlow library wasn't compiled to use FMA instructions, but these are available on your machine and could speed up CPU computations. Restored model parameters from ./deeplab_resnet.ckpt The output file has been saved to ./output/mask.png Segmentation fault (core dumped)

[bhack]

/cc @speedinghzl